From a2de06f0d6c54de8d7255063fe69876512021c86 Mon Sep 17 00:00:00 2001 From: Nabarun Dasgupta Date: Wed, 22 Apr 2020 21:22:08 -0400 Subject: [PATCH] Upload HTML version --- .DS_Store | Bin 10244 -> 10244 bytes ...locationcovid_exploratory-checkpoint.ipynb | 3076 +++++++++----- dl_x_valid.dta | Bin 8729380 -> 8729380 bytes docs/.DS_Store | Bin 6148 -> 6148 bytes ...id.html => locationcovid_exploratory.html} | 3733 +++++++++++++++-- docs/output_13_1.pdf | Bin 1525 -> 0 bytes docs/output_14_1.pdf | Bin 1420 -> 1335 bytes docs/output_14_1.svg | 26 +- docs/output_17_1.pdf | Bin 0 -> 1528 bytes docs/{output_13_1.svg => output_17_1.svg} | 56 +- docs/output_18_1.pdf | Bin 1417 -> 1317 bytes docs/output_18_1.svg | 46 +- docs/{output_16_1.pdf => output_22_1.pdf} | Bin 1423 -> 1335 bytes docs/{output_16_1.svg => output_22_1.svg} | 28 +- docs/{output_8_1.pdf => output_26_1.pdf} | Bin 1428 -> 1334 bytes docs/output_26_1.svg | 34 + docs/{output_10_1.pdf => output_29_1.pdf} | Bin 1431 -> 1318 bytes docs/output_29_1.svg | 32 + docs/{output_5_1.pdf => output_32_1.pdf} | Bin 1451 -> 1333 bytes docs/output_32_1.svg | 34 + docs/output_36_1.pdf | Bin 0 -> 1339 bytes docs/output_36_1.svg | 36 + docs/output_39_1.pdf | Bin 0 -> 1325 bytes docs/output_39_1.svg | 34 + docs/output_42_1.pdf | Bin 0 -> 1340 bytes docs/output_42_1.svg | 36 + docs/output_46_1.pdf | Bin 0 -> 1324 bytes docs/output_46_1.svg | 32 + docs/output_50_1.pdf | Bin 0 -> 1418 bytes docs/{output_10_1.svg => output_50_1.svg} | 34 +- docs/output_52_1.pdf | Bin 0 -> 1374 bytes docs/output_52_1.svg | 34 + docs/output_53_1.pdf | Bin 0 -> 1425 bytes docs/{output_5_1.svg => output_53_1.svg} | 32 +- docs/output_8_1.svg | 36 - docs/output_9_1.pdf | Bin 0 -> 1350 bytes docs/output_9_1.svg | 36 + google_x_valid.dta | Bin 5022962 -> 5022962 bytes locationcovid_exploratory.ipynb | 619 +-- 39 files changed, 6205 insertions(+), 1789 deletions(-) rename docs/{locationcovid.html => locationcovid_exploratory.html} (61%) delete mode 100644 docs/output_13_1.pdf create mode 100644 docs/output_17_1.pdf rename docs/{output_13_1.svg => output_17_1.svg} (67%) rename docs/{output_16_1.pdf => output_22_1.pdf} (53%) rename docs/{output_16_1.svg => output_22_1.svg} (58%) rename docs/{output_8_1.pdf => output_26_1.pdf} (52%) create mode 100644 docs/output_26_1.svg rename docs/{output_10_1.pdf => output_29_1.pdf} (52%) create mode 100644 docs/output_29_1.svg rename docs/{output_5_1.pdf => output_32_1.pdf} (52%) create mode 100644 docs/output_32_1.svg create mode 100644 docs/output_36_1.pdf create mode 100644 docs/output_36_1.svg create mode 100644 docs/output_39_1.pdf create mode 100644 docs/output_39_1.svg create mode 100644 docs/output_42_1.pdf create mode 100644 docs/output_42_1.svg create mode 100644 docs/output_46_1.pdf create mode 100644 docs/output_46_1.svg create mode 100644 docs/output_50_1.pdf rename docs/{output_10_1.svg => output_50_1.svg} (60%) create mode 100644 docs/output_52_1.pdf create mode 100644 docs/output_52_1.svg create mode 100644 docs/output_53_1.pdf rename docs/{output_5_1.svg => output_53_1.svg} (62%) delete mode 100644 docs/output_8_1.svg create mode 100644 docs/output_9_1.pdf create mode 100644 docs/output_9_1.svg diff --git a/.DS_Store b/.DS_Store index 527455ec931bae85cd3eea09e5017eaf00c8e6c0..6d9a4e3db6feff78eb8aba0caa91834e1b3fb25e 100644 GIT binary patch delta 1220 zcmeH{T}YEr7{{N#TkCym=GoRZKbkEpS!U{5>Qt0jmYR~5Ql>=eoVs*Qm>=kx%d#lw zqC84iL`D(xMUo0#l!8$~g+vHZ23>j6MO|bG71irzK^H}L-E{Yy^W*S5&+iYXr?{th zXf>;ps~jJQNKOe^_ne~)Oh2DWV0{c=uNXG z#T!$SQ++;B(62FP=2yBpI=u&cU1xZrdL_>CB6WmJHTj7*s4YvCg=J2=yQ#$4>EJ$X zt!5u%tscitM`Klsv(1rH)#C8*LA6$lvDpG+5Kq(CLNLI+npwZdKmr*kmaJ4rrBo|T z*i9$sJdMy@8l!P~OjGolX6PNw(HHs#0Ib6n-Xk5iLjNl&b;{hJxDPCe4uP}qRc#m1G(H4c} z<>zfv#{Jcf(LZ+-#TQ{rrPeIh>h$5OV%H^ZOv~ELbvo{k6WOT1FclFQwZdRBBnScv z2^^GWgGFRj0+R(YHWDu}S@*5?&T`PG_MRvez0Sd7lyCt{fsK8#-q7F{U z@4VeMw4(!ua2Q8$RC0Vua@>c0oW>w7;36(bu5V%pw=m2wiZP61BH(-y&-s`>T~Xo5 p>hCr2ulmRzWpsE{h-PRwBmBwYVTJLtT=hSke-ut=Fr0-i={rphC+`3N delta 1174 zcmeH{T}YE*6vxkhoAZ6?Egv`Mbh?_AVy5C;LmP=RmZm~37hx2*Pix`aXiL+%Oz@&8 zi+Z6Xi|7NRQV2}@z&?n&2(udv>c%d-sEaPTNGPf3WwW3g>888x&Ut=3JkR<4;q=-2 z?6=o4Ck^d7M??xLroH5+NFZplxm$y7>5My5G&j|4cv|%*_b$mp`-~1xFc6aN5Z$tq zIKlG8VYXbJRNW%gBU#EgQIMB{#}CF5oC VXtH5Jq52=gKaW93#E{rBKLPXdB{u*7 diff --git a/.ipynb_checkpoints/locationcovid_exploratory-checkpoint.ipynb b/.ipynb_checkpoints/locationcovid_exploratory-checkpoint.ipynb index 9687bbb..d37908d 100644 --- a/.ipynb_checkpoints/locationcovid_exploratory-checkpoint.ipynb +++ b/.ipynb_checkpoints/locationcovid_exploratory-checkpoint.ipynb @@ -47,7 +47,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "Notebook generated on 22 Apr 2020 at 09:49:01 ET\n" + "Notebook generated on 22 Apr 2020 at 21:10:47 ET\n" ] } ], @@ -181,17 +181,17 @@ "\n", " rucc | Freq. Percent Cum.\n", "------------+-----------------------------------\n", - " 1 | 344 16.23 16.23\n", - " 2 | 300 14.15 30.38\n", - " 3 | 273 12.88 43.25\n", - " 4 | 167 7.88 51.13\n", - " 5 | 76 3.58 54.72\n", - " 6 | 447 21.08 75.80\n", - " 7 | 318 15.00 90.80\n", - " 8 | 101 4.76 95.57\n", - " 9 | 94 4.43 100.00\n", + " 1 | 426 16.18 16.18\n", + " 2 | 368 13.98 30.16\n", + " 3 | 336 12.76 42.92\n", + " 4 | 214 8.13 51.04\n", + " 5 | 92 3.49 54.54\n", + " 6 | 572 21.72 76.26\n", + " 7 | 387 14.70 90.96\n", + " 8 | 124 4.71 95.67\n", + " 9 | 114 4.33 100.00\n", "------------+-----------------------------------\n", - " Total | 2,120 100.00\n", + " Total | 2,633 100.00\n", "\n", "\n", "Distancing: |\n", @@ -199,89 +199,136 @@ " to Highest |\n", " (5) | Freq. Percent Cum.\n", "------------+-----------------------------------\n", - " 1 | 415 19.58 19.58\n", - " 2 | 418 19.72 39.29\n", - " 3 | 437 20.61 59.91\n", - " 4 | 416 19.62 79.53\n", - " 5 | 434 20.47 100.00\n", + " 1 | 519 19.71 19.71\n", + " 2 | 526 19.98 39.69\n", + " 3 | 513 19.48 59.17\n", + " 4 | 540 20.51 79.68\n", + " 5 | 535 20.32 100.00\n", "------------+-----------------------------------\n", - " Total | 2,120 100.00\n", + " Total | 2,633 100.00\n", "\n", "\n", - "-------------------------\n", + " Variable | Obs Mean Std. Dev. Min Max\n", + "-------------+---------------------------------------------------------\n", + "last3_sample | 2,633 24893.38 41699.36 100 285845\n", + "\n", + "Quintile boundaries:\n", + "\n", + "\n", + "\n", + "------------------------------------\n", "Distancin |\n", "g: Lowest |\n", "(1) to |\n", "Highest |\n", - "(5) | sum(last3_~e)\n", - "----------+--------------\n", - " 1 | 1006757\n", - " 2 | 1554145\n", - " 3 | 1989986\n", - " 4 | 3043564\n", - " 5 | 5696915\n", - "-------------------------\n", + "(5) | min(last3i) max(last3i)\n", + "----------+-------------------------\n", + " 1 | -45.83333 193\n", + " 2 | -55.4 -46\n", + " 3 | -62.33333 -55.40625\n", + " 4 | -74.83334 -62.4058\n", + " 5 | -100 -75\n", + "------------------------------------\n", "\n", + "Places with positive movement:\n", "\n", - " Variable | Obs Mean Std. Dev. Min Max\n", - "-------------+---------------------------------------------------------\n", - "last3_sample | 2,120 6269.513 14490.88 101 285845\n", "\n", - "Total 3-day mobile traces: 13291367\n", + " state | Freq. Percent Cum.\n", + "---------------------+-----------------------------------\n", + " Alaska | 1 14.29 14.29\n", + " Arkansas | 1 14.29 28.57\n", + " Iowa | 1 14.29 42.86\n", + " Mississippi | 1 14.29 57.14\n", + " New Mexico | 1 14.29 71.43\n", + " Texas | 2 28.57 100.00\n", + "---------------------+-----------------------------------\n", + " Total | 7 100.00\n", + "\n", + "\n", + "--------------------------------------\n", + " state | med(last3i) mean(rucc)\n", + "------------+-------------------------\n", + " Alaska | 26.33334 7\n", + " Arkansas | 6.666664 6\n", + " Iowa | 119.3333 9\n", + "Mississippi | 3.666664 5\n", + " New Mexico | 63.66667 7\n", + " Texas | 105.5 8\n", + "--------------------------------------\n", "\n", "US average change in mobility:\n", "\n", "\n", "\n", "scalars:\n", - " r(N) = 2120\n", - " r(sum_w) = 2120\n", - " r(mean) = 42.12680818487732\n", - " r(Var) = 558.800535528917\n", - " r(sd) = 23.63896223460152\n", + " r(N) = 2633\n", + " r(sum_w) = 2633\n", + " r(mean) = 40.27232371120683\n", + " r(Var) = 378.1592527463074\n", + " r(sd) = 19.44631720265581\n", " r(min) = 0\n", - " r(max) = 378.3333435058594\n", - " r(sum) = 89308.83335193992\n", + " r(max) = 293\n", + " r(sum) = 106037.0283316076\n", "\n", "US median distance traveled km):\n", "\n", "\n", - "3.9808334\n", + "4.3673334\n", "\n", - "US mean change in mobility\n", + "US mean change in mobility:\n", "\n", "\n", - "-57.873192\n", + "-59.727676\n", "\n", "Mean mobility change by quintile:\n", "\n", "\n", - "--------------------------------------------------------------------------------\n", + "-------------------------------------------------------\n", + "Distancin |\n", + "g: Lowest |\n", + "(1) to |\n", + "Highest |\n", + "(5) | N(last3_~x) mean(last3i) med(last3~50)\n", + "----------+--------------------------------------------\n", + " 1 | 519 -33.2338 6.991\n", + " 2 | 526 -51.11703 5.701638\n", + " 3 | 513 -59.35949 4.606933\n", + " 4 | 540 -67.38671 3.749153\n", + " 5 | 535 -86.51745 1.14\n", + "-------------------------------------------------------\n", + "\n", + "\n", + "------------------------\n", "Distancin |\n", "g: Lowest |\n", "(1) to |\n", "Highest |\n", - "(5) | N(last3_i~x) mean(last~x) med(last3~x) sum(samples) med(last~50)\n", - "----------+---------------------------------------------------------------------\n", - " 1 | 415 73.21445 68.66666 1.76e+07 7.707334\n", - " 2 | 418 54.89673 55 2.80e+07 6.091667\n", - " 3 | 437 42.8032 43 3.37e+07 4.656667\n", - " 4 | 416 29.80529 29.66667 5.27e+07 2.757667\n", - " 5 | 434 11.23041 11.33333 9.62e+07 .8596666\n", - "--------------------------------------------------------------------------------\n", + "(5) | sum(last3~e)\n", + "----------+-------------\n", + " 1 | 2,289,735\n", + " 2 | 12,686,755\n", + " 3 | 21,464,240\n", + " 4 | 17,229,129\n", + " 5 | 11,874,409\n", + "------------------------\n", + "\n", + "\n", + "Total traces in 3 days: \n", + "\n", + " 65,544,268\n", "\n", "Mean mobility change by status of homeorder:\n", "\n", "\n", - "----------------------------------------------------------\n", + "----------------------------------------------------\n", "Stay at |\n", "home |\n", "order for |\n", - "COVID-19 | N(rucc) mean(last3_~x) sem(last3_~x)\n", - "----------+-----------------------------------------------\n", - " 0 | 268 52.17537 1.469372\n", - " 1 | 1,852 40.6727 .539732\n", - "----------------------------------------------------------\n", + "COVID-19 | N(rucc) mean(last3i) sem(last3i)\n", + "----------+-----------------------------------------\n", + " 0 | 337 -52.30764 1.047917\n", + " 1 | 2,296 -60.81676 .4015633\n", + "----------------------------------------------------\n", "\n", "Rurality/urbanicity by quintile:\n", "\n", @@ -294,8 +341,8 @@ "(5) | med(rucc)\n", "----------+-----------\n", " 1 | 6\n", - " 2 | 6\n", - " 3 | 4\n", + " 2 | 5\n", + " 3 | 5\n", " 4 | 3\n", " 5 | 3\n", "----------------------\n", @@ -305,72 +352,76 @@ "\n", " state | Freq. Percent Cum.\n", "---------------------+-----------------------------------\n", - " California | 3 2.19 2.19\n", - " Colorado | 3 2.19 4.38\n", - " Georgia | 1 0.73 5.11\n", - " Idaho | 2 1.46 6.57\n", - " Illinois | 3 2.19 8.76\n", - " Indiana | 6 4.38 13.14\n", - " Iowa | 4 2.92 16.06\n", - " Kansas | 1 0.73 16.79\n", - " Kentucky | 2 1.46 18.25\n", - " Maine | 2 1.46 19.71\n", - " Massachusetts | 2 1.46 21.17\n", - " Michigan | 30 21.90 43.07\n", - " Minnesota | 13 9.49 52.55\n", - " Mississippi | 1 0.73 53.28\n", - " Missouri | 1 0.73 54.01\n", - " Montana | 2 1.46 55.47\n", - " Nebraska | 1 0.73 56.20\n", - " New Hampshire | 1 0.73 56.93\n", - " New Mexico | 2 1.46 58.39\n", - " New York | 6 4.38 62.77\n", - " North Dakota | 3 2.19 64.96\n", - " Oklahoma | 3 2.19 67.15\n", - " Oregon | 1 0.73 67.88\n", - " Pennsylvania | 7 5.11 72.99\n", - " Texas | 12 8.76 81.75\n", - " Utah | 1 0.73 82.48\n", - " Vermont | 7 5.11 87.59\n", - " Virginia | 5 3.65 91.24\n", - " Washington | 2 1.46 92.70\n", - " West Virginia | 1 0.73 93.43\n", - " Wisconsin | 9 6.57 100.00\n", + " California | 4 2.29 2.29\n", + " Colorado | 3 1.71 4.00\n", + " Idaho | 3 1.71 5.71\n", + " Illinois | 2 1.14 6.86\n", + " Indiana | 10 5.71 12.57\n", + " Iowa | 5 2.86 15.43\n", + " Kansas | 5 2.86 18.29\n", + " Kentucky | 2 1.14 19.43\n", + " Maine | 2 1.14 20.57\n", + " Maryland | 1 0.57 21.14\n", + " Massachusetts | 2 1.14 22.29\n", + " Michigan | 31 17.71 40.00\n", + " Minnesota | 14 8.00 48.00\n", + " Mississippi | 1 0.57 48.57\n", + " Missouri | 2 1.14 49.71\n", + " Montana | 2 1.14 50.86\n", + " Nebraska | 3 1.71 52.57\n", + " New Mexico | 3 1.71 54.29\n", + " New York | 7 4.00 58.29\n", + " North Carolina | 2 1.14 59.43\n", + " North Dakota | 4 2.29 61.71\n", + " Ohio | 2 1.14 62.86\n", + " Oklahoma | 5 2.86 65.71\n", + " Oregon | 1 0.57 66.29\n", + " Pennsylvania | 9 5.14 71.43\n", + " South Dakota | 1 0.57 72.00\n", + " Tennessee | 1 0.57 72.57\n", + " Texas | 12 6.86 79.43\n", + " Utah | 2 1.14 80.57\n", + " Vermont | 8 4.57 85.14\n", + " Virginia | 9 5.14 90.29\n", + " Washington | 3 1.71 92.00\n", + " West Virginia | 2 1.14 93.14\n", + " Wisconsin | 10 5.71 98.86\n", + " Wyoming | 2 1.14 100.00\n", "---------------------+-----------------------------------\n", - " Total | 137 100.00\n", + " Total | 175 100.00\n", "\n", "Large municipalities in lowest tier:\n", "\n", "\n", " state | Freq. Percent Cum.\n", "---------------------+-----------------------------------\n", - " Alabama | 14 13.08 13.08\n", - " Arizona | 1 0.93 14.02\n", - " Arkansas | 6 5.61 19.63\n", - " Colorado | 1 0.93 20.56\n", - " Florida | 3 2.80 23.36\n", - " Georgia | 17 15.89 39.25\n", - " Illinois | 1 0.93 40.19\n", - " Iowa | 1 0.93 41.12\n", - " Kansas | 1 0.93 42.06\n", - " Kentucky | 3 2.80 44.86\n", - " Louisiana | 7 6.54 51.40\n", - " Minnesota | 1 0.93 52.34\n", - " Mississippi | 4 3.74 56.07\n", - " Missouri | 3 2.80 58.88\n", - " North Carolina | 5 4.67 63.55\n", - " North Dakota | 2 1.87 65.42\n", - " Ohio | 1 0.93 66.36\n", - " Oklahoma | 1 0.93 67.29\n", - " Pennsylvania | 1 0.93 68.22\n", - " South Carolina | 8 7.48 75.70\n", - " South Dakota | 1 0.93 76.64\n", - " Tennessee | 8 7.48 84.11\n", - " Texas | 12 11.21 95.33\n", - " Utah | 1 0.93 96.26\n", - " Virginia | 4 3.74 100.00\n", + " Alabama | 15 10.00 10.00\n", + " Arizona | 1 0.67 10.67\n", + " Arkansas | 9 6.00 16.67\n", + " California | 1 0.67 17.33\n", + " Florida | 1 0.67 18.00\n", + " Georgia | 22 14.67 32.67\n", + " Idaho | 2 1.33 34.00\n", + " Indiana | 1 0.67 34.67\n", + " Iowa | 2 1.33 36.00\n", + " Kansas | 1 0.67 36.67\n", + " Kentucky | 2 1.33 38.00\n", + " Louisiana | 17 11.33 49.33\n", + " Mississippi | 6 4.00 53.33\n", + " Missouri | 8 5.33 58.67\n", + " North Carolina | 11 7.33 66.00\n", + " North Dakota | 2 1.33 67.33\n", + " Oklahoma | 2 1.33 68.67\n", + " Oregon | 1 0.67 69.33\n", + " Pennsylvania | 1 0.67 70.00\n", + " South Carolina | 8 5.33 75.33\n", + " South Dakota | 1 0.67 76.00\n", + " Tennessee | 8 5.33 81.33\n", + " Texas | 19 12.67 94.00\n", + " Utah | 1 0.67 94.67\n", + " Virginia | 8 5.33 100.00\n", "---------------------+-----------------------------------\n", - " Total | 107 100.00\n" + " Total | 150 100.00\n" ] } ], @@ -380,9 +431,14 @@ "\n", "tab rucc, m\n", "tab iso5, m\n", - "table iso5, c(sum last3_sample)\n", "su last3_sample\n", - "di \"Total 3-day mobile traces: \" r(sum)\n", + " \n", + "di \"Quintile boundaries:\"\n", + " gen last3i=last3_index-100\n", + " table iso5, c(min last3i max last3i)\n", + " di \"Places with positive movement:\"\n", + " tab state if last3i>=0\n", + " table state if last3i>=0, c(median last3i mean rucc)\n", "\n", "di \"US average change in mobility:\"\n", " qui: su last3_index\n", @@ -390,15 +446,19 @@ "di \"US median distance traveled km):\"\n", " qui: su last3_m50, d\n", " di r(p50)\n", - "di \"US mean change in mobility\"\n", - " qui: su last3_index\n", - " di r(mean)-100\n", + "di \"US mean change in mobility:\"\n", + " qui: su last3i\n", + " di r(mean)\n", " \n", "di \"Mean mobility change by quintile:\"\n", - " table iso5, c(count last3_index mean last3_index median last3_index sum samples median last3_m50) \n", + " table iso5, c(count last3_index mean last3i median last3_m50) \n", + " table iso5, c(sum last3_sample) format(%12.0fc)\n", + " qui: su last3_sample\n", + " di \"Total traces in 3 days: \"\n", + " di %12.0fc r(sum)\n", "\n", "di \"Mean mobility change by status of homeorder:\"\n", - " table homeorder, c(count rucc mean last3_index sem last3_index)\n", + " table homeorder, c(count rucc mean last3i sem last3i)\n", "\n", "di \"Rurality/urbanicity by quintile:\"\n", " table iso5, c(median rucc)\n", @@ -438,49 +498,49 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -258204.94 \n", - "Iteration 1: log pseudolikelihood = -39468.839 \n", - "Iteration 2: log pseudolikelihood = -24923.731 \n", - "Iteration 3: log pseudolikelihood = -21328.635 \n", - "Iteration 4: log pseudolikelihood = -21327.584 \n", - "Iteration 5: log pseudolikelihood = -21327.584 \n", + "Iteration 0: log pseudolikelihood = -320092.99 \n", + "Iteration 1: log pseudolikelihood = -48882.026 \n", + "Iteration 2: log pseudolikelihood = -30820.243 \n", + "Iteration 3: log pseudolikelihood = -26330.47 \n", + "Iteration 4: log pseudolikelihood = -26328.915 \n", + "Iteration 5: log pseudolikelihood = -26328.915 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -10380.593 \n", - "Iteration 1: log pseudolikelihood = -9921.5674 \n", - "Iteration 2: log pseudolikelihood = -9713.3941 \n", - "Iteration 3: log pseudolikelihood = -9712.983 \n", - "Iteration 4: log pseudolikelihood = -9712.9829 \n", + "Iteration 0: log pseudolikelihood = -12906.655 \n", + "Iteration 1: log pseudolikelihood = -12333.437 \n", + "Iteration 2: log pseudolikelihood = -12073.753 \n", + "Iteration 3: log pseudolikelihood = -12073.291 \n", + "Iteration 4: log pseudolikelihood = -12073.291 \n", "\n", - "Negative binomial regression Number of obs = 2,074\n", - "Dispersion = mean Wald chi2(14) = 124005.68\n", - "Log pseudolikelihood = -9712.9829 Prob > chi2 = 0.0000\n", + "Negative binomial regression Number of obs = 2,579\n", + "Dispersion = mean Wald chi2(14) = 158357.87\n", + "Log pseudolikelihood = -12073.291 Prob > chi2 = 0.0000\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " pcp_rate | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 48.03302 2.404615 77.34 0.000 43.5439 52.98495\n", - " levels2 | 50.28804 2.545734 77.39 0.000 45.53803 55.53351\n", - " levels3 | 56.80765 2.835276 80.94 0.000 51.51376 62.64558\n", - " levels4 | 61.25371 3.008065 83.79 0.000 55.63286 67.44247\n", - " levels5 | 71.49253 3.509299 86.98 0.000 64.93493 78.71236\n", - " homeorder | .9970538 .0326169 -0.09 0.928 .9351321 1.063076\n", + " levels1 | 49.92198 2.325957 83.93 0.000 45.56514 54.69541\n", + " levels2 | 54.89074 2.441119 90.06 0.000 50.30882 59.88996\n", + " levels3 | 57.89873 2.709556 86.73 0.000 52.82437 63.46053\n", + " levels4 | 61.49219 2.753973 91.97 0.000 56.32462 67.13387\n", + " levels5 | 73.62823 3.207945 98.67 0.000 67.60175 80.19195\n", + " homeorder | .9716535 .0285067 -0.98 0.327 .9173575 1.029163\n", " |\n", " rucc |\n", - " 2 | 1.033211 .0435763 0.77 0.439 .9512374 1.122248\n", - " 3 | 1.129862 .0582734 2.37 0.018 1.021232 1.250048\n", - " 4 | .9547024 .0432737 -1.02 0.306 .8735459 1.043399\n", - " 5 | 1.302038 .0704226 4.88 0.000 1.171077 1.447646\n", - " 6 | .8129643 .0335999 -5.01 0.000 .7497065 .8815597\n", - " 7 | 1.058695 .0465956 1.30 0.195 .9711979 1.154076\n", - " 8 | .5952729 .0474725 -6.50 0.000 .5091357 .6959831\n", - " 9 | .6727804 .0487545 -5.47 0.000 .5836992 .7754567\n", + " 2 | 1.031687 .039499 0.81 0.415 .9571034 1.112082\n", + " 3 | 1.120497 .0519795 2.45 0.014 1.023113 1.22715\n", + " 4 | .9427865 .0368889 -1.51 0.132 .8731884 1.017932\n", + " 5 | 1.279302 .0611998 5.15 0.000 1.164805 1.405055\n", + " 6 | .8080024 .0295491 -5.83 0.000 .7521142 .8680436\n", + " 7 | 1.027799 .0403114 0.70 0.484 .9517511 1.109924\n", + " 8 | .5639108 .0416387 -7.76 0.000 .487931 .651722\n", + " 9 | .6694049 .0419761 -6.40 0.000 .591988 .756946\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -1.350341 .0387661 -1.426321 -1.274361\n", + " /lnalpha | -1.354027 .0344678 -1.421582 -1.286471\n", "-------------+----------------------------------------------------------------\n", - " alpha | .2591519 .0100463 .240191 .2796097\n", + " alpha | .2581985 .0088995 .2413319 .2762439\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Compare to tabular data:\n", @@ -492,11 +552,11 @@ "Highest |\n", "(5) | N(pcp_rate) mean(pcp_rate) sem(pcp_rate)\n", "----------+-----------------------------------------------\n", - " 1 | 396 45.918473 1.253123\n", - " 2 | 410 48.266257 1.37516\n", - " 3 | 428 54.377764 1.485897\n", - " 4 | 410 59.639797 1.572321\n", - " 5 | 430 69.71894 2.039235\n", + " 1 | 499 46.57405 1.122683\n", + " 2 | 520 50.705057 1.227417\n", + " 3 | 504 53.536733 1.405582\n", + " 4 | 526 57.650394 1.358795\n", + " 5 | 530 69.016454 1.764024\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: you are responsible for interpretation of non-count dep. variable\n", @@ -504,56 +564,56 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -21327.614 \n", - "Iteration 1: log pseudolikelihood = -21327.584 \n", - "Iteration 2: log pseudolikelihood = -21327.584 \n", + "Iteration 0: log pseudolikelihood = -26328.97 \n", + "Iteration 1: log pseudolikelihood = -26328.915 \n", + "Iteration 2: log pseudolikelihood = -26328.915 \n", "\n", "Fitting constant-only model:\n", "\n", - "Iteration 0: log pseudolikelihood = -10432.724 \n", - "Iteration 1: log pseudolikelihood = -10021.901 \n", - "Iteration 2: log pseudolikelihood = -9889.366 \n", - "Iteration 3: log pseudolikelihood = -9889.3209 \n", - "Iteration 4: log pseudolikelihood = -9889.3209 \n", + "Iteration 0: log pseudolikelihood = -12966.731 \n", + "Iteration 1: log pseudolikelihood = -12449.434 \n", + "Iteration 2: log pseudolikelihood = -12278.236 \n", + "Iteration 3: log pseudolikelihood = -12278.172 \n", + "Iteration 4: log pseudolikelihood = -12278.172 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -9728.7005 \n", - "Iteration 1: log pseudolikelihood = -9713.1574 \n", - "Iteration 2: log pseudolikelihood = -9712.9829 \n", - "Iteration 3: log pseudolikelihood = -9712.9829 \n", + "Iteration 0: log pseudolikelihood = -12090.402 \n", + "Iteration 1: log pseudolikelihood = -12073.452 \n", + "Iteration 2: log pseudolikelihood = -12073.291 \n", + "Iteration 3: log pseudolikelihood = -12073.291 \n", "\n", - "Negative binomial regression Number of obs = 2,074\n", - " Wald chi2(13) = 366.31\n", + "Negative binomial regression Number of obs = 2,579\n", + " Wald chi2(13) = 423.89\n", "Dispersion = mean Prob > chi2 = 0.0000\n", - "Log pseudolikelihood = -9712.9829 Pseudo R2 = 0.0178\n", + "Log pseudolikelihood = -12073.291 Pseudo R2 = 0.0167\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " pcp_rate | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | .6718608 .027037 -9.88 0.000 .6209052 .7269981\n", - " levels2 | .7034027 .0285788 -8.66 0.000 .6495615 .7617068\n", - " levels3 | .7945957 .0310854 -5.88 0.000 .7359466 .8579186\n", - " levels4 | .8567848 .0330574 -4.01 0.000 .7943828 .9240888\n", + " levels1 | .6780277 .0239497 -11.00 0.000 .6326751 .7266314\n", + " levels2 | .7455121 .0261976 -8.36 0.000 .6958942 .7986679\n", + " levels3 | .7863659 .0279794 -6.75 0.000 .7333958 .8431619\n", + " levels4 | .8351714 .0287434 -5.23 0.000 .7806934 .893451\n", " levels5 | 1 (omitted)\n", - " homeorder | .9970538 .0326169 -0.09 0.928 .9351321 1.063076\n", + " homeorder | .9716535 .0285067 -0.98 0.327 .9173575 1.029163\n", " |\n", " rucc |\n", - " 2 | 1.033211 .0435763 0.77 0.439 .9512373 1.122248\n", - " 3 | 1.129862 .0582734 2.37 0.018 1.021232 1.250048\n", - " 4 | .9547023 .0432736 -1.02 0.306 .8735458 1.043399\n", - " 5 | 1.302038 .0704226 4.88 0.000 1.171076 1.447645\n", - " 6 | .8129642 .0335999 -5.01 0.000 .7497064 .8815596\n", - " 7 | 1.058695 .0465956 1.30 0.195 .9711978 1.154076\n", - " 8 | .5952729 .0474725 -6.50 0.000 .5091357 .695983\n", - " 9 | .6727803 .0487545 -5.47 0.000 .5836992 .7754566\n", + " 2 | 1.031687 .0394989 0.81 0.415 .9571034 1.112082\n", + " 3 | 1.120497 .0519795 2.45 0.014 1.023113 1.22715\n", + " 4 | .9427864 .0368889 -1.51 0.132 .8731884 1.017932\n", + " 5 | 1.279302 .0611998 5.15 0.000 1.164805 1.405055\n", + " 6 | .8080024 .0295491 -5.83 0.000 .7521141 .8680435\n", + " 7 | 1.027799 .0403114 0.70 0.484 .951751 1.109924\n", + " 8 | .5639107 .0416387 -7.76 0.000 .487931 .651722\n", + " 9 | .6694049 .0419761 -6.40 0.000 .591988 .756946\n", " |\n", - " _cons | 71.49253 3.509299 86.98 0.000 64.93493 78.71237\n", + " _cons | 73.62823 3.207945 98.67 0.000 67.60175 80.19196\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -1.350341 .0387661 -1.426321 -1.274361\n", + " /lnalpha | -1.354026 .0344678 -1.421582 -1.286471\n", "-------------+----------------------------------------------------------------\n", - " alpha | .259152 .0100463 .240191 .2796097\n", + " alpha | .2581985 .0088995 .2413319 .276244\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Note: _cons estimates baseline incidence rate.\n" @@ -561,7 +621,7 @@ }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -571,17 +631,19 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t50\n", - "\t55\n", - "\t60\n", - "\t65\n", - "\t70\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t50\n", + "\t55\n", + "\t60\n", + "\t65\n", + "\t70\n", + "\t75\n", "\tavg\n", "\t\n", "\t\n", @@ -607,17 +669,19 @@ "\t<rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/>\n", "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", - "\t<line x1="55.44" y1="1273.49" x2="1342.53" y2="1273.49" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1030.74" x2="1342.53" y2="1030.74" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="787.99" x2="1342.53" y2="787.99" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="545.29" x2="1342.53" y2="545.29" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="302.54" x2="1342.53" y2="302.54" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 1370.61 L381.95 1258.93 L698.96 943.37 L1015.97 724.93 L1333.03 229.73" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1364.51" y="1294.67" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">50</text>\n", - "\t<text x="1364.51" y="1051.92" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">55</text>\n", - "\t<text x="1364.51" y="809.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">60</text>\n", - "\t<text x="1364.51" y="566.48" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">65</text>\n", - "\t<text x="1364.51" y="323.73" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">70</text>\n", + "\t<line x1="55.44" y1="1366.00" x2="1342.53" y2="1366.00" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1138.75" x2="1342.53" y2="1138.75" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="911.49" x2="1342.53" y2="911.49" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="684.24" x2="1342.53" y2="684.24" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="456.98" x2="1342.53" y2="456.98" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="229.73" x2="1342.53" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<path d=" M64.90 1370.61 L381.95 1143.35 L698.96 1006.98 L1015.97 843.33 L1333.03 293.39" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1364.51" y="1387.19" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">50</text>\n", + "\t<text x="1364.51" y="1159.93" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">55</text>\n", + "\t<text x="1364.51" y="932.68" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">60</text>\n", + "\t<text x="1364.51" y="705.42" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">65</text>\n", + "\t<text x="1364.51" y="478.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">70</text>\n", + "\t<text x="1364.51" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">75</text>\n", "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1342.53" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -665,15 +729,15 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 48.03302 2.404615 77.34 0.000 43.5439 52.98495\n", - " levels2 | 50.28804 2.545734 77.39 0.000 45.53803 55.53351\n", - " levels3 | 56.80765 2.835276 80.94 0.000 51.51376 62.64558\n", - " levels4 | 61.25371 3.008065 83.79 0.000 55.63286 67.44247\n", - " levels5 | 71.49253 3.509299 86.98 0.000 64.93493 78.71236\n", + " levels1 | 49.92198 2.325957 83.93 0.000 45.56514 54.69541\n", + " levels2 | 54.89074 2.441119 90.06 0.000 50.30882 59.88996\n", + " levels3 | 57.89873 2.709556 86.73 0.000 52.82437 63.46053\n", + " levels4 | 61.49219 2.753973 91.97 0.000 56.32462 67.13387\n", + " levels5 | 73.62823 3.207945 98.67 0.000 67.60175 80.19195\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | .6718608 .027037 -9.88 0.000 .6209052 .7269981`" + "` levels1 | .6780277 .0239497 -11.00 0.000 .6326751 .7266314`" ] }, { @@ -685,14 +749,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "48.8\n", - "LL: 61.1\n", + "47.5\n", + "LL: 58.1\n", "UL: 37.6\n" ] } ], "source": [ - "invert .6718608 .6209052 .7269981" + "invert .6780277 .6326751 .7266314" ] }, { @@ -700,7 +764,7 @@ "metadata": {}, "source": [ "\n", - "The counties showing the smallest declines in mobility had 48 primary care providers per 100,000, whereas the most social distancing counties had 71 per 100,000 after adjusting for rurality and stay-at-home orders, a 49% (95% CI: 38%, 61%) difference. \n", + "The counties showing the smallest declines in mobility had 50 primary care providers per 100,000, whereas the most social distancing counties had 74 per 100,000 after adjusting for rurality and stay-at-home orders, a 47% (95% CI: 38%, 58%) difference. \n", "\n", "---" ] @@ -722,48 +786,55 @@ "name": "stdout", "output_type": "stream", "text": [ - "----- RURALITY-ADJUSTED POISSON MODEL -----\n", - "note: uninsured_p has noninteger values\n", + "----- RURALITY-ADJUSTED NEGBIN MODEL -----\n", + "note: you are responsible for interpretation of non-count dep. variable\n", "\n", - "Iteration 0: log likelihood = -6278.5262 \n", - "Iteration 1: log likelihood = -6271.627 \n", - "Iteration 2: log likelihood = -6271.6259 \n", - "Iteration 3: log likelihood = -6271.6259 \n", + "Fitting Poisson model:\n", "\n", - "Generalized linear models Number of obs = 2,120\n", - "Optimization : ML Residual df = 2,106\n", - " Scale parameter = 1\n", - "Deviance = 3729.423569 (1/df) Deviance = 1.770856\n", - "Pearson = 3922.25363 (1/df) Pearson = 1.862419\n", + "Iteration 0: log pseudolikelihood = -25368.195 \n", + "Iteration 1: log pseudolikelihood = -7847.5733 \n", + "Iteration 2: log pseudolikelihood = -7718.5574 \n", + "Iteration 3: log pseudolikelihood = -7718.2864 \n", + "Iteration 4: log pseudolikelihood = -7718.2864 \n", "\n", - "Variance function: V(u) = u [Poisson]\n", - "Link function : g(u) = ln(u) [Log]\n", + "Fitting full model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -8982.7909 \n", + "Iteration 1: log pseudolikelihood = -7464.8863 \n", + "Iteration 2: log pseudolikelihood = -7462.9819 \n", + "Iteration 3: log pseudolikelihood = -7462.981 \n", + "Iteration 4: log pseudolikelihood = -7462.981 \n", "\n", - " AIC = 5.929836\n", - "Log likelihood = -6271.62591 BIC = -12400.79\n", + "Negative binomial regression Number of obs = 2,633\n", + "Dispersion = mean Wald chi2(14) = 94084.62\n", + "Log pseudolikelihood = -7462.981 Prob > chi2 = 0.0000\n", "\n", "------------------------------------------------------------------------------\n", - " | OIM\n", + " | Robust\n", " uninsured_p | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 11.0404 .4479299 59.19 0.000 10.19647 11.95417\n", - " levels2 | 10.5449 .4268925 58.19 0.000 9.74054 11.41568\n", - " levels3 | 9.052178 .3649673 54.64 0.000 8.364388 9.796524\n", - " levels4 | 8.318344 .3368231 52.32 0.000 7.683699 9.005408\n", - " levels5 | 7.144656 .291952 48.12 0.000 6.594755 7.74041\n", - " homeorder | 1.135978 .0317841 4.56 0.000 1.07536 1.200013\n", + " levels1 | 10.74092 .3917303 65.09 0.000 9.999938 11.5368\n", + " levels2 | 9.464582 .3546025 59.99 0.000 8.794479 10.18574\n", + " levels3 | 8.480079 .3192442 56.78 0.000 7.876899 9.129449\n", + " levels4 | 8.000203 .3006194 55.34 0.000 7.432174 8.611645\n", + " levels5 | 7.044127 .2647529 51.94 0.000 6.543873 7.582624\n", + " homeorder | 1.156171 .0313348 5.35 0.000 1.096358 1.219246\n", " |\n", " rucc |\n", - " 2 | 1.021169 .0353039 0.61 0.545 .9542663 1.092761\n", - " 3 | 1.003037 .0356308 0.09 0.932 .9355773 1.07536\n", - " 4 | 1.047498 .0423684 1.15 0.251 .9676641 1.133919\n", - " 5 | 1.047333 .0556754 0.87 0.384 .9437034 1.162342\n", - " 6 | 1.14665 .0356718 4.40 0.000 1.078823 1.218741\n", - " 7 | 1.114966 .0379895 3.19 0.001 1.042939 1.191966\n", - " 8 | 1.156837 .0533969 3.16 0.002 1.056776 1.266373\n", - " 9 | 1.132531 .0543145 2.60 0.009 1.030927 1.24415\n", + " 2 | 1.036269 .0311154 1.19 0.235 .9770437 1.099084\n", + " 3 | 1.02913 .0307182 0.96 0.336 .9706509 1.091133\n", + " 4 | 1.050335 .0360446 1.43 0.152 .9820124 1.123411\n", + " 5 | 1.05925 .0478817 1.27 0.203 .9694404 1.157379\n", + " 6 | 1.170248 .0327939 5.61 0.000 1.107706 1.236321\n", + " 7 | 1.123108 .0352113 3.70 0.000 1.056173 1.194286\n", + " 8 | 1.237528 .0477472 5.52 0.000 1.147396 1.33474\n", + " 9 | 1.1618 .0488127 3.57 0.000 1.069962 1.261521\n", + "-------------+----------------------------------------------------------------\n", + " /lnalpha | -2.682096 .0589782 -2.797691 -2.566501\n", + "-------------+----------------------------------------------------------------\n", + " alpha | .0684196 .0040353 .0609506 .0768038\n", "------------------------------------------------------------------------------\n", - "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", + "Note: Estimates are transformed only in the first equation.\n", "Compare to tabular data:\n", "\n", "----------------------------------------------------------\n", @@ -773,63 +844,75 @@ "Highest |\n", "(5) | N(uninsu~p) mean(uninsu~p) sem(uninsu~p)\n", "----------+-----------------------------------------------\n", - " 1 | 415 13.348753 .2311543\n", - " 2 | 418 12.678412 .2322624\n", - " 3 | 437 10.838269 .2318229\n", - " 4 | 416 9.8819901 .2159591\n", - " 5 | 434 8.4965582 .1963992\n", + " 1 | 519 13.379365 .2038213\n", + " 2 | 526 11.729307 .2069097\n", + " 3 | 513 10.520381 .2010726\n", + " 4 | 540 9.8374245 .1876119\n", + " 5 | 535 8.6297361 .1775762\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", + "note: you are responsible for interpretation of non-count dep. variable\n", "note: levels5 omitted because of collinearity\n", - "note: uninsured_p has noninteger values\n", "\n", - "Iteration 0: log likelihood = -6278.5262 \n", - "Iteration 1: log likelihood = -6271.627 \n", - "Iteration 2: log likelihood = -6271.6259 \n", - "Iteration 3: log likelihood = -6271.6259 \n", + "Fitting Poisson model:\n", "\n", - "Generalized linear models Number of obs = 2,120\n", - "Optimization : ML Residual df = 2,106\n", - " Scale parameter = 1\n", - "Deviance = 3729.423569 (1/df) Deviance = 1.770856\n", - "Pearson = 3922.25363 (1/df) Pearson = 1.862419\n", + "Iteration 0: log pseudolikelihood = -7718.2865 \n", + "Iteration 1: log pseudolikelihood = -7718.2864 \n", "\n", - "Variance function: V(u) = u [Poisson]\n", - "Link function : g(u) = ln(u) [Log]\n", + "Fitting constant-only model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -9016.9274 \n", + "Iteration 1: log pseudolikelihood = -7680.503 \n", + "Iteration 2: log pseudolikelihood = -7675.9149 \n", + "Iteration 3: log pseudolikelihood = -7675.8619 \n", + "Iteration 4: log pseudolikelihood = -7675.8619 \n", + "\n", + "Fitting full model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -7480.4687 \n", + "Iteration 1: log pseudolikelihood = -7463.1407 \n", + "Iteration 2: log pseudolikelihood = -7462.981 \n", + "Iteration 3: log pseudolikelihood = -7462.981 \n", "\n", - " AIC = 5.929836\n", - "Log likelihood = -6271.62591 BIC = -12400.79\n", + "Negative binomial regression Number of obs = 2,633\n", + " Wald chi2(13) = 482.26\n", + "Dispersion = mean Prob > chi2 = 0.0000\n", + "Log pseudolikelihood = -7462.981 Pseudo R2 = 0.0277\n", "\n", "------------------------------------------------------------------------------\n", - " | OIM\n", + " | Robust\n", " uninsured_p | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 1.545267 .0467263 14.39 0.000 1.456346 1.639617\n", - " levels2 | 1.475914 .0443857 12.94 0.000 1.391434 1.565524\n", - " levels3 | 1.266986 .0384861 7.79 0.000 1.193756 1.344708\n", - " levels4 | 1.164275 .0362991 4.88 0.000 1.09526 1.237639\n", + " levels1 | 1.524804 .0395603 16.26 0.000 1.449206 1.604347\n", + " levels2 | 1.343613 .0368115 10.78 0.000 1.273367 1.417735\n", + " levels3 | 1.203851 .0339756 6.57 0.000 1.139068 1.272318\n", + " levels4 | 1.135727 .0316513 4.57 0.000 1.075355 1.199488\n", " levels5 | 1 (omitted)\n", - " homeorder | 1.135978 .0317841 4.56 0.000 1.07536 1.200013\n", + " homeorder | 1.156171 .0313348 5.35 0.000 1.096358 1.219246\n", " |\n", " rucc |\n", - " 2 | 1.021169 .0353039 0.61 0.545 .9542663 1.092761\n", - " 3 | 1.003037 .0356308 0.09 0.932 .9355773 1.07536\n", - " 4 | 1.047498 .0423684 1.15 0.251 .9676641 1.133919\n", - " 5 | 1.047333 .0556754 0.87 0.384 .9437034 1.162342\n", - " 6 | 1.14665 .0356718 4.40 0.000 1.078823 1.218741\n", - " 7 | 1.114966 .0379895 3.19 0.001 1.042939 1.191966\n", - " 8 | 1.156837 .0533969 3.16 0.002 1.056776 1.266373\n", - " 9 | 1.132531 .0543145 2.60 0.009 1.030927 1.24415\n", + " 2 | 1.036269 .0311154 1.19 0.235 .9770437 1.099084\n", + " 3 | 1.02913 .0307182 0.96 0.336 .9706509 1.091133\n", + " 4 | 1.050335 .0360446 1.43 0.152 .9820124 1.123411\n", + " 5 | 1.05925 .0478817 1.27 0.203 .9694404 1.157379\n", + " 6 | 1.170248 .0327939 5.61 0.000 1.107706 1.236321\n", + " 7 | 1.123108 .0352113 3.70 0.000 1.056173 1.194286\n", + " 8 | 1.237528 .0477472 5.52 0.000 1.147396 1.33474\n", + " 9 | 1.1618 .0488127 3.57 0.000 1.069962 1.261521\n", " |\n", - " _cons | 7.144656 .291952 48.12 0.000 6.594755 7.74041\n", + " _cons | 7.044127 .2647529 51.94 0.000 6.543873 7.582624\n", + "-------------+----------------------------------------------------------------\n", + " /lnalpha | -2.682096 .0589782 -2.797691 -2.566501\n", + "-------------+----------------------------------------------------------------\n", + " alpha | .0684196 .0040353 .0609506 .0768038\n", "------------------------------------------------------------------------------\n", - "Note: _cons estimates baseline incidence rate.\n", - "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" + "Note: Estimates are transformed only in the first equation.\n", + "Note: _cons estimates baseline incidence rate.\n" ] }, { "data": { - "application/pdf": "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", + "application/pdf": "JVBERi0xLjMKJbe+raoKMSAwIG9iago8PAovVHlwZSAvQ2F0YWxvZwovUGFnZXMgMiAwIFIKPj4KZW5kb2JqCjIgMCBvYmoKPDwKL1R5cGUgL1BhZ2VzCi9LaWRzIFsgNCAwIFIgXQovQ291bnQgMQo+PgplbmRvYmoKMyAwIG9iago8PAovUHJvZHVjZXIgKEhhcnUgRnJlZSBQREYgTGlicmFyeSAyLjQuMGRldikKPj4KZW5kb2JqCjQgMCBvYmoKPDwKL1R5cGUgL1BhZ2UKL01lZGlhQm94IFsgMCAwIDE1OC4zOTk5OSAxNTguMzk5OTkgXQovQ29udGVudHMgNSAwIFIKL1Jlc291cmNlcyA8PAovUHJvY1NldCBbIC9QREYgL1RleHQgL0ltYWdlQiAvSW1hZ2VDIC9JbWFnZUkgXQovRm9udCA8PAovRjEgNyAwIFIKPj4KPj4KL1BhcmVudCAyIDAgUgo+PgplbmRvYmoKNSAwIG9iago8PAovTGVuZ3RoIDYgMCBSCi9GaWx0ZXIgWyAvRmxhdGVEZWNvZGUgXQo+PgpzdHJlYW0NCnicpVRNk9MwDL37V/gIB4IkW7bMcYevAxfYHJlhOqVbstMPaLvs30e2m7S7k0Jbkhlbcvzek2Qr0MQYfLTQCEXPedYJ7WZu4MSnLx8MWLDI0rikz5G1mZkbxQH4pPvze2RPlya4JiWxjye5/y3rMAvVUeXujMOmruyn5FhFUUWPbd16axLrnqjyaPOrjIgE1jskuzQUkbA6C908xHoChQDRD7DqnYVjcXzAFe8cHCHrao+r3lm46DgdcMV7gnOilghggd7rAZAWwWvhXSIMOvsEVPgcKgMHB8oXo35DyePCoC+pUeJcBAImX0rishcSpVxXLLqgEoP2L3PTmtfvNV4isu1dOWLI1wCLkQPmoEcpwrZdmhfxpW3vzbvWfD4bm/Jtyli5HIvsQQo4XQ4mxAgFjHAFOiRHFY3H6GPo39gy06tM6YCSOl5YuzGIC4V08nt+YK2goR/yMPRDduq59Xtyh3gO+fJXu99yfnB9qilSQ+I1Ru9qriNBlZu2F6z29YLRxZAVUQ6aNFYIX5LfZ+mfVOJyVXRR8LmsG5HNncO97N65XpYAhVVW79Gg6sdUa4fuVavzH6qBJT1T5Wtu8EDoQc8sBRGrDZUK4e162k0W9m233U1W0241f2M/rR9n2539CgyoA9rd2n7s5j/6Rc6LY32YNNXRUxP9i5GIwyL5sOpW24fN7Pu3nweWP/bbd5oKZW5kc3RyZWFtCmVuZG9iago2IDAgb2JqCjUyOQplbmRvYmoKNyAwIG9iago8PAovVHlwZSAvRm9udAovQmFzZUZvbnQgL0hlbHZldGljYQovU3VidHlwZSAvVHlwZTEKL0VuY29kaW5nIC9XaW5BbnNpRW5jb2RpbmcKPj4KZW5kb2JqCnhyZWYKMCA4CjAwMDAwMDAwMDAgNjU1MzUgZg0KMDAwMDAwMDAxNSAwMDAwMCBuDQowMDAwMDAwMDY0IDAwMDAwIG4NCjAwMDAwMDAxMjMgMDAwMDAgbg0KMDAwMDAwMDE4NyAwMDAwMCBuDQowMDAwMDAwMzc1IDAwMDAwIG4NCjAwMDAwMDA5ODMgMDAwMDAgbg0KMDAwMDAwMTAwMiAwMDAwMCBuDQp0cmFpbGVyCjw8Ci9Sb290IDEgMCBSCi9JbmZvIDMgMCBSCi9TaXplIDgKPj4Kc3RhcnR4cmVmCjEwOTkKJSVFT0YK", "image/svg+xml": [ "\n", "\n", @@ -844,7 +927,7 @@ "\t\n", "\t\n", "\t\n", - "\t\n", + "\t\n", "\t7\n", "\t8\n", "\t9\n", @@ -880,7 +963,7 @@ "\t<line x1="55.44" y1="800.17" x2="1342.53" y2="800.17" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="514.95" x2="1342.53" y2="514.95" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="229.73" x2="1342.53" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 229.73 L381.95 372.34 L698.96 771.66 L1015.97 999.80 L1333.03 1342.09" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<path d=" M64.90 315.32 L381.95 657.56 L698.96 942.78 L1015.97 1085.39 L1333.03 1370.61" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">7</text>\n", "\t<text x="1364.51" y="1106.52" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">8</text>\n", "\t<text x="1364.51" y="821.35" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">9</text>\n", @@ -922,7 +1005,7 @@ ], "source": [ "// Comparing percent uninsured to social distancing\n", - "modelpoisson uninsured_p " + "modelrun uninsured_p " ] }, { @@ -934,17 +1017,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 11.0404 .4479299 59.19 0.000 10.19647 11.95417\n", - " levels2 | 10.5449 .4268925 58.19 0.000 9.74054 11.41568\n", - " levels3 | 9.052178 .3649673 54.64 0.000 8.364388 9.796524\n", - " levels4 | 8.318344 .3368231 52.32 0.000 7.683699 9.005408\n", - " levels5 | 7.144656 .291952 48.12 0.000 6.594755 7.74041\n", + " levels1 | 10.74092 .3917303 65.09 0.000 9.999938 11.5368\n", + " levels2 | 9.464582 .3546025 59.99 0.000 8.794479 10.18574\n", + " levels3 | 8.480079 .3192442 56.78 0.000 7.876899 9.129449\n", + " levels4 | 8.000203 .3006194 55.34 0.000 7.432174 8.611645\n", + " levels5 | 7.044127 .2647529 51.94 0.000 6.543873 7.582624\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | 1.545267 .0467263 14.39 0.000 1.456346 1.639617`\n", + "` levels1 | 1.524804 .0395603 16.26 0.000 1.449206 1.604347`\n", "\n", - "Counties with lower social distancing also had a higher proportion of people without health insurance. The lowest social distancing counties had 11.0% uninsured adults, whereas the most social distancing counties had only 7.1% uninsured after adjusting for rurality and social distancing orders, a 54% (95% CI: 46%, 64%) difference. \n", + "Counties with lower social distancing also had a higher proportion of people without health insurance. The lowest social distancing counties had 10.7% uninsured adults, whereas the most social distancing counties had only 7.0% uninsured after adjusting for rurality and social distancing orders, a 52% (95% CI: 45%, 60%) difference. \n", "\n", "---" ] @@ -973,24 +1056,24 @@ " % Medicare Beneficiaries Getting Flu Vaccine\n", "-------------------------------------------------------------\n", " Percentiles Smallest\n", - " 1% 18 9\n", - " 5% 26 9\n", - "10% 30 12 Obs 2,117\n", - "25% 37 12 Sum of Wgt. 2,117\n", + " 1% 19 9\n", + " 5% 26 12\n", + "10% 30 12 Obs 2,630\n", + "25% 37 13 Sum of Wgt. 2,630\n", "\n", - "50% 43 Mean 41.87907\n", - " Largest Std. Dev. 8.698995\n", + "50% 43 Mean 42.15703\n", + " Largest Std. Dev. 8.51741\n", "75% 48 62\n", - "90% 52 62 Variance 75.67252\n", - "95% 54 63 Skewness -.6357286\n", - "99% 57 65 Kurtosis 3.27119\n", + "90% 52 62 Variance 72.54627\n", + "95% 54 63 Skewness -.6164957\n", + "99% 57 65 Kurtosis 3.235287\n", "\n", "(bin=10, start=9, width=5.6)\n" ] }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -1001,37 +1084,37 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", "\t0\n", - "\t.01\n", - "\t.02\n", - "\t.03\n", - "\t.04\n", - "\t.05\n", + "\t.01\n", + "\t.02\n", + "\t.03\n", + "\t.04\n", + "\t.05\n", "\tDensity\n", "\t\n", "\t\n", @@ -1059,37 +1142,37 @@ "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", "\t<line x1="55.44" y1="1380.06" x2="1324.12" y2="1380.06" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1124.89" x2="1324.12" y2="1124.89" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="869.76" x2="1324.12" y2="869.76" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="614.59" x2="1324.12" y2="614.59" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="359.42" x2="1324.12" y2="359.42" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="104.25" x2="1324.12" y2="104.25" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<rect x="124.45" y="1358.53" width="119.00" height="21.53" style="fill:#CAC27E"/>\n", - "\t<rect x="126.82" y="1360.90" width="114.25" height="16.78" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="243.45" y="1326.25" width="119.05" height="53.81" style="fill:#CAC27E"/>\n", - "\t<rect x="245.82" y="1328.63" width="114.30" height="49.05" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="362.50" y="1231.56" width="119.00" height="148.50" style="fill:#CAC27E"/>\n", - "\t<rect x="364.88" y="1233.94" width="114.25" height="143.75" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="481.50" y="1022.77" width="119.00" height="357.29" style="fill:#CAC27E"/>\n", - "\t<rect x="483.88" y="1025.14" width="114.25" height="352.54" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="600.50" y="839.82" width="119.05" height="540.24" style="fill:#CAC27E"/>\n", - "\t<rect x="602.88" y="842.19" width="114.30" height="535.49" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="719.55" y="334.03" width="119.00" height="1046.03" style="fill:#CAC27E"/>\n", - "\t<rect x="721.93" y="336.40" width="114.25" height="1041.28" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1123.90" x2="1324.12" y2="1123.90" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="867.69" x2="1324.12" y2="867.69" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="611.47" x2="1324.12" y2="611.47" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="355.26" x2="1324.12" y2="355.26" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="99.10" x2="1324.12" y2="99.10" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<rect x="124.45" y="1366.15" width="119.00" height="13.91" style="fill:#CAC27E"/>\n", + "\t<rect x="126.82" y="1368.53" width="114.25" height="9.16" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="243.45" y="1331.35" width="119.05" height="48.71" style="fill:#CAC27E"/>\n", + "\t<rect x="245.82" y="1333.73" width="114.30" height="43.96" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="362.50" y="1240.92" width="119.00" height="139.14" style="fill:#CAC27E"/>\n", + "\t<rect x="364.88" y="1243.29" width="114.25" height="134.39" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="481.50" y="1049.55" width="119.00" height="330.51" style="fill:#CAC27E"/>\n", + "\t<rect x="483.88" y="1051.92" width="114.25" height="325.76" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="600.50" y="844.32" width="119.05" height="535.74" style="fill:#CAC27E"/>\n", + "\t<rect x="602.88" y="846.70" width="114.30" height="530.99" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="719.55" y="310.27" width="119.00" height="1069.79" style="fill:#CAC27E"/>\n", + "\t<rect x="721.93" y="312.64" width="114.25" height="1065.04" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", "\t<rect x="838.56" y="64.99" width="119.00" height="1315.07" style="fill:#CAC27E"/>\n", "\t<rect x="840.93" y="67.37" width="114.25" height="1310.31" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="957.56" y="564.35" width="119.00" height="815.71" style="fill:#CAC27E"/>\n", - "\t<rect x="959.93" y="566.73" width="114.25" height="810.96" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="1076.56" y="1138.99" width="119.05" height="241.07" style="fill:#CAC27E"/>\n", - "\t<rect x="1078.94" y="1141.37" width="114.30" height="236.31" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", - "\t<rect x="1195.61" y="1362.83" width="119.00" height="17.23" style="fill:#CAC27E"/>\n", - "\t<rect x="1197.99" y="1365.21" width="114.25" height="12.47" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="957.56" y="529.45" width="119.00" height="850.61" style="fill:#CAC27E"/>\n", + "\t<rect x="959.93" y="531.83" width="114.25" height="845.86" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="1076.56" y="1126.08" width="119.05" height="253.98" style="fill:#CAC27E"/>\n", + "\t<rect x="1078.94" y="1128.45" width="114.30" height="249.23" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", + "\t<rect x="1195.61" y="1362.69" width="119.00" height="17.37" style="fill:#CAC27E"/>\n", + "\t<rect x="1197.99" y="1365.06" width="114.25" height="12.62" style="fill:none;stroke:#D7D29E;stroke-width:4.75"/>\n", "\t<text x="1346.10" y="1401.24" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">0</text>\n", - "\t<text x="1346.10" y="1146.07" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.01</text>\n", - "\t<text x="1346.10" y="890.95" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.02</text>\n", - "\t<text x="1346.10" y="635.78" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.03</text>\n", - "\t<text x="1346.10" y="380.60" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.04</text>\n", - "\t<text x="1346.10" y="125.43" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.05</text>\n", + "\t<text x="1346.10" y="1145.08" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.01</text>\n", + "\t<text x="1346.10" y="888.87" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.02</text>\n", + "\t<text x="1346.10" y="632.66" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.03</text>\n", + "\t<text x="1346.10" y="376.45" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.04</text>\n", + "\t<text x="1346.10" y="120.28" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">.05</text>\n", "\t<text x="1499.40" y="717.80" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,717.80)" text-anchor="middle">Density</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1324.12" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="145.68" y1="1358.08" x2="145.68" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -1150,42 +1233,42 @@ "text": [ "----- RURALITY-ADJUSTED POISSON MODEL -----\n", "\n", - "Iteration 0: log likelihood = -7578.5889 \n", - "Iteration 1: log likelihood = -7576.9644 \n", - "Iteration 2: log likelihood = -7576.9644 \n", + "Iteration 0: log likelihood = -9326.8988 \n", + "Iteration 1: log likelihood = -9325.1183 \n", + "Iteration 2: log likelihood = -9325.1183 \n", "\n", - "Generalized linear models Number of obs = 2,117\n", - "Optimization : ML Residual df = 2,103\n", + "Generalized linear models Number of obs = 2,630\n", + "Optimization : ML Residual df = 2,616\n", " Scale parameter = 1\n", - "Deviance = 3402.605707 (1/df) Deviance = 1.617977\n", - "Pearson = 3223.276093 (1/df) Pearson = 1.532704\n", + "Deviance = 4029.442144 (1/df) Deviance = 1.540307\n", + "Pearson = 3820.490076 (1/df) Pearson = 1.460432\n", "\n", "Variance function: V(u) = u [Poisson]\n", "Link function : g(u) = ln(u) [Log]\n", "\n", - " AIC = 7.171435\n", - "Log likelihood = -7576.964367 BIC = -12701.65\n", + " AIC = 7.101991\n", + "Log likelihood = -9325.118318 BIC = -16570.88\n", "\n", "------------------------------------------------------------------------------\n", " | OIM\n", " fluvaccine | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 48.23875 .8912549 209.80 0.000 46.52317 50.01759\n", - " levels2 | 47.49881 .8681976 211.22 0.000 45.82729 49.23129\n", - " levels3 | 48.55726 .8674856 217.34 0.000 46.88644 50.28762\n", - " levels4 | 49.43355 .8763789 220.02 0.000 47.74538 51.18141\n", - " levels5 | 50.63066 .8872871 223.94 0.000 48.92114 52.39993\n", - " homeorder | .9191356 .011747 -6.60 0.000 .8963979 .94245\n", + " levels1 | 47.99562 .7799153 238.23 0.000 46.4911 49.54882\n", + " levels2 | 48.2899 .7663704 244.31 0.000 46.81096 49.81556\n", + " levels3 | 49.35107 .7754712 248.13 0.000 47.85434 50.89462\n", + " levels4 | 49.66889 .7746225 250.41 0.000 48.17363 51.21056\n", + " levels5 | 50.96012 .7778188 257.55 0.000 49.4582 52.50765\n", + " homeorder | .9198825 .0101901 -7.54 0.000 .9001255 .9400732\n", " |\n", " rucc |\n", - " 2 | .9875665 .0145302 -0.85 0.395 .9594946 1.01646\n", - " 3 | .9974339 .0151559 -0.17 0.866 .968167 1.027586\n", - " 4 | .9772802 .0171871 -1.31 0.191 .944168 1.011554\n", - " 5 | .9281343 .0227396 -3.04 0.002 .8846187 .9737905\n", - " 6 | .8742476 .0122593 -9.58 0.000 .850547 .8986085\n", - " 7 | .8190064 .0128867 -12.69 0.000 .7941344 .8446573\n", - " 8 | .8347185 .0188863 -7.98 0.000 .7985108 .872568\n", - " 9 | .7724262 .0187278 -10.65 0.000 .736579 .8100181\n", + " 2 | .9804055 .012666 -1.53 0.126 .9558922 1.005547\n", + " 3 | .9893659 .0131858 -0.80 0.422 .9638569 1.01555\n", + " 4 | .9732812 .0148226 -1.78 0.075 .9446587 1.002771\n", + " 5 | .9212544 .0199685 -3.78 0.000 .8829365 .9612352\n", + " 6 | .8735529 .0106074 -11.13 0.000 .8530082 .8945924\n", + " 7 | .8206082 .0113219 -14.33 0.000 .7987151 .8431015\n", + " 8 | .8344541 .0164987 -9.15 0.000 .8027359 .8674256\n", + " 9 | .7862014 .0165937 -11.40 0.000 .7543418 .8194065\n", "------------------------------------------------------------------------------\n", "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", "Compare to tabular data:\n", @@ -1197,53 +1280,53 @@ "Highest |\n", "(5) | N(fluvac~e) mean(fluvac~e) sem(fluvac~e)\n", "----------+-----------------------------------------------\n", - " 1 | 413 40.246975 .4141614\n", - " 2 | 418 40.442585 .4208429\n", - " 3 | 437 41.636154 .4359385\n", - " 4 | 415 42.906025 .4171079\n", - " 5 | 434 44.078342 .3957995\n", + " 1 | 517 40.218567 .3511693\n", + " 2 | 526 41.165398 .3810684\n", + " 3 | 513 42.111111 .3742107\n", + " 4 | 539 42.922077 .3666867\n", + " 5 | 535 44.278503 .3587537\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: levels5 omitted because of collinearity\n", "\n", - "Iteration 0: log likelihood = -7578.5889 \n", - "Iteration 1: log likelihood = -7576.9644 \n", - "Iteration 2: log likelihood = -7576.9644 \n", + "Iteration 0: log likelihood = -9326.8988 \n", + "Iteration 1: log likelihood = -9325.1183 \n", + "Iteration 2: log likelihood = -9325.1183 \n", "\n", - "Generalized linear models Number of obs = 2,117\n", - "Optimization : ML Residual df = 2,103\n", + "Generalized linear models Number of obs = 2,630\n", + "Optimization : ML Residual df = 2,616\n", " Scale parameter = 1\n", - "Deviance = 3402.605707 (1/df) Deviance = 1.617977\n", - "Pearson = 3223.276093 (1/df) Pearson = 1.532704\n", + "Deviance = 4029.442144 (1/df) Deviance = 1.540307\n", + "Pearson = 3820.490076 (1/df) Pearson = 1.460432\n", "\n", "Variance function: V(u) = u [Poisson]\n", "Link function : g(u) = ln(u) [Log]\n", "\n", - " AIC = 7.171435\n", - "Log likelihood = -7576.964367 BIC = -12701.65\n", + " AIC = 7.101991\n", + "Log likelihood = -9325.118318 BIC = -16570.88\n", "\n", "------------------------------------------------------------------------------\n", " | OIM\n", " fluvaccine | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | .9527577 .013099 -3.52 0.000 .927427 .9787802\n", - " levels2 | .9381431 .0126406 -4.74 0.000 .9136922 .9632483\n", - " levels3 | .9590484 .0124945 -3.21 0.001 .9348697 .9838525\n", - " levels4 | .976356 .0127096 -1.84 0.066 .9517607 1.001587\n", + " levels1 | .9418269 .0112491 -5.02 0.000 .9200352 .9641349\n", + " levels2 | .9476017 .0110092 -4.63 0.000 .926268 .9694268\n", + " levels3 | .9684253 .0112147 -2.77 0.006 .9466925 .990657\n", + " levels4 | .9746619 .0110035 -2.27 0.023 .9533323 .9964687\n", " levels5 | 1 (omitted)\n", - " homeorder | .9191356 .011747 -6.60 0.000 .8963979 .94245\n", + " homeorder | .9198825 .0101901 -7.54 0.000 .9001255 .9400732\n", " |\n", " rucc |\n", - " 2 | .9875665 .0145302 -0.85 0.395 .9594946 1.01646\n", - " 3 | .9974339 .0151559 -0.17 0.866 .968167 1.027586\n", - " 4 | .9772802 .0171871 -1.31 0.191 .944168 1.011554\n", - " 5 | .9281343 .0227396 -3.04 0.002 .8846187 .9737905\n", - " 6 | .8742476 .0122593 -9.58 0.000 .850547 .8986085\n", - " 7 | .8190064 .0128867 -12.69 0.000 .7941344 .8446573\n", - " 8 | .8347185 .0188863 -7.98 0.000 .7985108 .872568\n", - " 9 | .7724262 .0187278 -10.65 0.000 .736579 .8100181\n", + " 2 | .9804055 .012666 -1.53 0.126 .9558922 1.005547\n", + " 3 | .9893659 .0131858 -0.80 0.422 .9638569 1.01555\n", + " 4 | .9732812 .0148226 -1.78 0.075 .9446587 1.002771\n", + " 5 | .9212544 .0199685 -3.78 0.000 .8829365 .9612352\n", + " 6 | .8735529 .0106074 -11.13 0.000 .8530082 .8945924\n", + " 7 | .8206082 .0113219 -14.33 0.000 .7987151 .8431015\n", + " 8 | .8344541 .0164987 -9.15 0.000 .8027359 .8674256\n", + " 9 | .7862014 .0165937 -11.40 0.000 .7543418 .8194065\n", " |\n", - " _cons | 50.63066 .8872871 223.94 0.000 48.92114 52.39993\n", + " _cons | 50.96012 .7778188 257.55 0.000 49.4582 52.50765\n", "------------------------------------------------------------------------------\n", "Note: _cons estimates baseline incidence rate.\n", "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" @@ -1251,7 +1334,7 @@ }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -1262,15 +1345,13 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", + "\t\n", + "\t\n", "\t\n", - "\t\n", - "\t47\n", - "\t48\n", - "\t49\n", - "\t50\n", + "\t\n", + "\t48\n", + "\t49\n", + "\t50\n", "\t51\n", "\tavg\n", "\t\n", @@ -1298,15 +1379,13 @@ "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", "\t<line x1="55.44" y1="1370.56" x2="1342.53" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1085.34" x2="1342.53" y2="1085.34" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="800.17" x2="1342.53" y2="800.17" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="514.95" x2="1342.53" y2="514.95" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="990.30" x2="1342.53" y2="990.30" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="609.99" x2="1342.53" y2="609.99" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="229.73" x2="1342.53" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 1028.31 L381.95 1228.00 L698.96 914.26 L1015.97 686.07 L1333.03 343.83" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">47</text>\n", - "\t<text x="1364.51" y="1106.52" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">48</text>\n", - "\t<text x="1364.51" y="821.35" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">49</text>\n", - "\t<text x="1364.51" y="536.13" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">50</text>\n", + "\t<path d=" M64.90 1370.61 L381.95 1256.51 L698.96 838.18 L1015.97 724.14 L1333.03 229.73" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">48</text>\n", + "\t<text x="1364.51" y="1011.48" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">49</text>\n", + "\t<text x="1364.51" y="631.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">50</text>\n", "\t<text x="1364.51" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">51</text>\n", "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1342.53" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -1356,17 +1435,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 48.23875 .8912549 209.80 0.000 46.52317 50.01759\n", - " levels2 | 47.49881 .8681976 211.22 0.000 45.82729 49.23129\n", - " levels3 | 48.55726 .8674856 217.34 0.000 46.88644 50.28762\n", - " levels4 | 49.43355 .8763789 220.02 0.000 47.74538 51.18141\n", - " levels5 | 50.63066 .8872871 223.94 0.000 48.92114 52.39993\n", + " levels1 | 47.99562 .7799153 238.23 0.000 46.4911 49.54882\n", + " levels2 | 48.2899 .7663704 244.31 0.000 46.81096 49.81556\n", + " levels3 | 49.35107 .7754712 248.13 0.000 47.85434 50.89462\n", + " levels4 | 49.66889 .7746225 250.41 0.000 48.17363 51.21056\n", + " levels5 | 50.96012 .7778188 257.55 0.000 49.4582 52.50765\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | .9527577 .013099 -3.52 0.000 .927427 .9787802`\n", + "` levels1 | .9418269 .0112491 -5.02 0.000 .9200352 .9641349`\n", "\n", - "The lowest social distancing counties had 48.2% flu vaccine coverage among Medicare beneficiaries, whereas the most social distancing counties had 50.6% after adjusting for rurality and social distancing orders, a 5.0% (95% CI: 2.2%, 7.8%) difference. " + "The lowest social distancing counties had 48.0% flu vaccine coverage among Medicare beneficiaries, whereas the most social distancing counties had 51.0% after adjusting for rurality and social distancing orders, a 6.2% (95% CI: 3.7%, 8.7%) difference. " ] }, { @@ -1378,14 +1457,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "5\n", - "LL: 7.8\n", - "UL: 2.2\n" + "6.2\n", + "LL: 8.7\n", + "UL: 3.7\n" ] } ], "source": [ - "invert .9527577 .927427 .9787802" + "invert .9418269 .9200352 .9641349" ] }, { @@ -1413,51 +1492,51 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -1.937e+09 \n", - "Iteration 1: log pseudolikelihood = -2.197e+08 (backed up)\n", - "Iteration 2: log pseudolikelihood = -1.647e+08 \n", - "Iteration 3: log pseudolikelihood = -6735833.8 \n", - "Iteration 4: log pseudolikelihood = -3153019.9 \n", - "Iteration 5: log pseudolikelihood = -3096238.4 \n", - "Iteration 6: log pseudolikelihood = -3096228 \n", - "Iteration 7: log pseudolikelihood = -3096228 \n", + "Iteration 0: log pseudolikelihood = -2.396e+09 \n", + "Iteration 1: log pseudolikelihood = -2.765e+08 (backed up)\n", + "Iteration 2: log pseudolikelihood = -2.287e+08 \n", + "Iteration 3: log pseudolikelihood = -7797007.7 \n", + "Iteration 4: log pseudolikelihood = -3712358 \n", + "Iteration 5: log pseudolikelihood = -3655049.8 \n", + "Iteration 6: log pseudolikelihood = -3655041 \n", + "Iteration 7: log pseudolikelihood = -3655041 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -26426.306 \n", - "Iteration 1: log pseudolikelihood = -23517.281 \n", - "Iteration 2: log pseudolikelihood = -23506.346 \n", - "Iteration 3: log pseudolikelihood = -23506.226 \n", - "Iteration 4: log pseudolikelihood = -23506.226 \n", + "Iteration 0: log pseudolikelihood = -32813.382 \n", + "Iteration 1: log pseudolikelihood = -29141.104 \n", + "Iteration 2: log pseudolikelihood = -29125.526 \n", + "Iteration 3: log pseudolikelihood = -29125.351 \n", + "Iteration 4: log pseudolikelihood = -29125.351 \n", "\n", - "Negative binomial regression Number of obs = 2,120\n", - "Dispersion = mean Wald chi2(14) = 1.08e+07\n", - "Log pseudolikelihood = -23506.226 Prob > chi2 = 0.0000\n", + "Negative binomial regression Number of obs = 2,633\n", + "Dispersion = mean Wald chi2(14) = 1.42e+07\n", + "Log pseudolikelihood = -29125.351 Prob > chi2 = 0.0000\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " income80 | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 120492.2 2165.852 650.87 0.000 116321.1 124812.9\n", - " levels2 | 121736 2143.795 664.93 0.000 117605.9 126011.1\n", - " levels3 | 126707.4 2289.893 650.15 0.000 122297.8 131275.9\n", - " levels4 | 132681.5 2399.382 652.28 0.000 128061.2 137468.6\n", - " levels5 | 140242.2 2690.91 617.65 0.000 135066.1 145616.7\n", - " homeorder | .9299983 .010703 -6.31 0.000 .9092555 .9512142\n", + " levels1 | 118675.2 1873.364 740.18 0.000 115059.7 122404.3\n", + " levels2 | 122077.6 1869.469 764.83 0.000 118468 125797.3\n", + " levels3 | 125227.2 2022.182 726.89 0.000 121325.9 129254\n", + " levels4 | 129432 1971.797 772.66 0.000 125624.5 133354.9\n", + " levels5 | 139390 2261.967 729.93 0.000 135026.4 143894.6\n", + " homeorder | .9368546 .008621 -7.09 0.000 .9201093 .9539046\n", " |\n", " rucc |\n", - " 2 | .8401934 .01312 -11.15 0.000 .8148683 .8663057\n", - " 3 | .8106106 .0125033 -13.61 0.000 .7864712 .8354908\n", - " 4 | .7547631 .0122427 -17.35 0.000 .7311453 .7791438\n", - " 5 | .7986015 .0175574 -10.23 0.000 .7649204 .8337657\n", - " 6 | .7158771 .0104136 -22.98 0.000 .695755 .7365811\n", - " 7 | .7316058 .0121274 -18.85 0.000 .7082186 .7557654\n", - " 8 | .6945982 .0131431 -19.26 0.000 .6693099 .7208419\n", - " 9 | .6897351 .0162861 -15.73 0.000 .6585423 .7224053\n", + " 2 | .8432181 .0117573 -12.23 0.000 .8204863 .8665797\n", + " 3 | .8122305 .0110667 -15.26 0.000 .7908273 .8342129\n", + " 4 | .7601974 .0110199 -18.91 0.000 .7389027 .7821058\n", + " 5 | .793445 .0154531 -11.88 0.000 .7637282 .8243182\n", + " 6 | .7209769 .009222 -25.58 0.000 .7031268 .7392802\n", + " 7 | .72983 .0106562 -21.57 0.000 .7092403 .7510174\n", + " 8 | .6908203 .0120309 -21.24 0.000 .6676381 .7148074\n", + " 9 | .6910585 .0140465 -18.18 0.000 .6640691 .7191448\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -3.574182 .0386558 -3.649946 -3.498418\n", + " /lnalpha | -3.62168 .0344743 -3.689249 -3.554112\n", "-------------+----------------------------------------------------------------\n", - " alpha | .0280384 .0010838 .0259925 .0302452\n", + " alpha | .0267377 .0009218 .0249908 .0286068\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Compare to tabular data:\n", @@ -1469,67 +1548,67 @@ "Highest |\n", "(5) | N(income80) mean(income80) sem(income80)\n", "----------+-----------------------------------------------\n", - " 1 | 415 86193.88 718.9255\n", - " 2 | 418 89069.581 763.5601\n", - " 3 | 437 94655.206 933.8522\n", - " 4 | 416 101217.26 1099.554\n", - " 5 | 434 110185.07 1470.521\n", + " 1 | 519 85968.287 590.9798\n", + " 2 | 526 90843.228 733.2388\n", + " 3 | 513 93619.183 872.0948\n", + " 4 | 540 98925.328 890.1288\n", + " 5 | 535 110406.09 1298.318\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: levels5 omitted because of collinearity\n", "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -3096229.9 \n", - "Iteration 1: log pseudolikelihood = -3096228 \n", - "Iteration 2: log pseudolikelihood = -3096228 \n", + "Iteration 0: log pseudolikelihood = -3655043.1 \n", + "Iteration 1: log pseudolikelihood = -3655041 \n", + "Iteration 2: log pseudolikelihood = -3655041 \n", "\n", "Fitting constant-only model:\n", "\n", - "Iteration 0: log pseudolikelihood = -26448.903 \n", - "Iteration 1: log pseudolikelihood = -24107.639 \n", - "Iteration 2: log pseudolikelihood = -24107.251 \n", - "Iteration 3: log pseudolikelihood = -24107.251 \n", + "Iteration 0: log pseudolikelihood = -32840.589 \n", + "Iteration 1: log pseudolikelihood = -29881.7 \n", + "Iteration 2: log pseudolikelihood = -29881.116 \n", + "Iteration 3: log pseudolikelihood = -29881.116 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -23649.497 \n", - "Iteration 1: log pseudolikelihood = -23530.823 \n", - "Iteration 2: log pseudolikelihood = -23506.244 \n", - "Iteration 3: log pseudolikelihood = -23506.226 \n", - "Iteration 4: log pseudolikelihood = -23506.226 \n", + "Iteration 0: log pseudolikelihood = -29307.258 \n", + "Iteration 1: log pseudolikelihood = -29157.452 \n", + "Iteration 2: log pseudolikelihood = -29125.374 \n", + "Iteration 3: log pseudolikelihood = -29125.351 \n", + "Iteration 4: log pseudolikelihood = -29125.351 \n", "\n", - "Negative binomial regression Number of obs = 2,120\n", - " Wald chi2(13) = 1115.68\n", + "Negative binomial regression Number of obs = 2,633\n", + " Wald chi2(13) = 1399.51\n", "Dispersion = mean Prob > chi2 = 0.0000\n", - "Log pseudolikelihood = -23506.226 Pseudo R2 = 0.0249\n", + "Log pseudolikelihood = -29125.351 Pseudo R2 = 0.0253\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " income80 | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | .8591722 .011219 -11.62 0.000 .8374625 .8814448\n", - " levels2 | .8680411 .0108797 -11.29 0.000 .846977 .889629\n", - " levels3 | .9034895 .0116578 -7.87 0.000 .8809272 .9266297\n", - " levels4 | .9460884 .0124211 -4.22 0.000 .922054 .9707493\n", + " levels1 | .8513898 .0096039 -14.26 0.000 .832773 .8704228\n", + " levels2 | .8757993 .0097536 -11.91 0.000 .8568898 .8951261\n", + " levels3 | .8983949 .010434 -9.23 0.000 .8781757 .9190797\n", + " levels4 | .9285604 .0105342 -6.53 0.000 .9081416 .9494384\n", " levels5 | 1 (omitted)\n", - " homeorder | .9299983 .010703 -6.31 0.000 .9092555 .9512142\n", + " homeorder | .9368545 .008621 -7.09 0.000 .9201092 .9539046\n", " |\n", " rucc |\n", - " 2 | .8401934 .01312 -11.15 0.000 .8148683 .8663057\n", - " 3 | .8106106 .0125033 -13.61 0.000 .7864712 .8354908\n", - " 4 | .7547631 .0122427 -17.35 0.000 .7311453 .7791438\n", - " 5 | .7986015 .0175574 -10.23 0.000 .7649204 .8337657\n", - " 6 | .7158771 .0104136 -22.98 0.000 .695755 .7365811\n", - " 7 | .7316058 .0121274 -18.85 0.000 .7082186 .7557654\n", - " 8 | .6945982 .0131431 -19.26 0.000 .6693099 .7208419\n", - " 9 | .6897351 .0162861 -15.73 0.000 .6585423 .7224053\n", + " 2 | .8432181 .0117573 -12.23 0.000 .8204863 .8665797\n", + " 3 | .8122305 .0110667 -15.26 0.000 .7908273 .834213\n", + " 4 | .7601975 .0110199 -18.91 0.000 .7389028 .7821059\n", + " 5 | .793445 .0154531 -11.88 0.000 .7637282 .8243181\n", + " 6 | .7209769 .009222 -25.58 0.000 .7031268 .7392802\n", + " 7 | .72983 .0106562 -21.57 0.000 .7092403 .7510174\n", + " 8 | .6908203 .0120309 -21.24 0.000 .6676381 .7148075\n", + " 9 | .6910585 .0140465 -18.18 0.000 .664069 .7191448\n", " |\n", - " _cons | 140242.2 2690.91 617.65 0.000 135066.1 145616.7\n", + " _cons | 139389.9 2261.966 729.93 0.000 135026.3 143894.5\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -3.574182 .0386558 -3.649946 -3.498418\n", + " /lnalpha | -3.621674 .0344743 -3.689242 -3.554105\n", "-------------+----------------------------------------------------------------\n", - " alpha | .0280384 .0010838 .0259925 .0302452\n", + " alpha | .0267379 .0009218 .0249909 .028607\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Note: _cons estimates baseline incidence rate.\n" @@ -1537,7 +1616,7 @@ }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -1547,17 +1626,17 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t120000\n", - "\t125000\n", - "\t130000\n", - "\t135000\n", - "\t140000\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t120000\n", + "\t125000\n", + "\t130000\n", + "\t135000\n", + "\t140000\n", "\tavg\n", "\t\n", "\t\n", @@ -1583,17 +1662,17 @@ "\t<rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/>\n", "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", - "\t<line x1="55.44" y1="1370.56" x2="1205.31" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1088.75" x2="1205.31" y2="1088.75" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="806.95" x2="1205.31" y2="806.95" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="525.19" x2="1205.31" y2="525.19" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="243.39" x2="1205.31" y2="243.39" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 1342.84 L347.65 1272.74 L630.35 992.57 L913.06 655.88 L1195.81 229.73" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1227.34" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">120000</text>\n", - "\t<text x="1227.34" y="1109.94" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">125000</text>\n", - "\t<text x="1227.34" y="828.13" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">130000</text>\n", - "\t<text x="1227.34" y="546.38" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">135000</text>\n", - "\t<text x="1227.34" y="264.58" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">140000</text>\n", + "\t<line x1="55.44" y1="1299.67" x2="1205.31" y2="1299.67" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1032.22" x2="1205.31" y2="1032.22" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="764.73" x2="1205.31" y2="764.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="497.23" x2="1205.31" y2="497.23" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="229.73" x2="1205.31" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<path d=" M64.90 1370.61 L347.65 1188.54 L630.35 1020.05 L913.06 795.12 L1195.81 262.40" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1227.34" y="1320.86" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">120000</text>\n", + "\t<text x="1227.34" y="1053.41" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">125000</text>\n", + "\t<text x="1227.34" y="785.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">130000</text>\n", + "\t<text x="1227.34" y="518.41" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">135000</text>\n", + "\t<text x="1227.34" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">140000</text>\n", "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1205.31" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -1642,18 +1721,18 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 120492.2 2165.852 650.87 0.000 116321.1 124812.9\n", - " levels2 | 121736 2143.795 664.93 0.000 117605.9 126011.1\n", - " levels3 | 126707.4 2289.893 650.15 0.000 122297.8 131275.9\n", - " levels4 | 132681.5 2399.382 652.28 0.000 128061.2 137468.6\n", - " levels5 | 140242.2 2690.91 617.65 0.000 135066.1 145616.7\n", + " levels1 | 118675.2 1873.364 740.18 0.000 115059.7 122404.3\n", + " levels2 | 122077.6 1869.469 764.83 0.000 118468 125797.3\n", + " levels3 | 125227.2 2022.182 726.89 0.000 121325.9 129254\n", + " levels4 | 129432 1971.797 772.66 0.000 125624.5 133354.9\n", + " levels5 | 139390 2261.967 729.93 0.000 135026.4 143894.6\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | .8591722 .011219 -11.62 0.000 .8374625 .8814448`\n", + "` levels1 | .8513898 .0096039 -14.26 0.000 .832773 .8704228`\n", "\n", "\n", - "The lowest social distancing counties the 80th percentile of annual household income was around `$120,000`, whereas in the most social distancing counties it was `$140,000`, after adjusting for rurality and social distancing orders, a 16% (95% CI: 13%, 19%) difference. \n" + "The lowest social distancing counties the 80th percentile of annual household income was around `$120,000`, whereas in the most social distancing counties it was `$140,000`, after adjusting for rurality and social distancing orders, a 17% (95% CI: 15%, 20%) difference. \n" ] }, { @@ -1665,14 +1744,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "16.4\n", - "LL: 19.4\n", - "UL: 13.5\n" + "17.5\n", + "LL: 20.1\n", + "UL: 14.9\n" ] } ], "source": [ - "invert .8591722 .8374625 .8814448" + "invert .8513898 .832773 .8704228" ] }, { @@ -1698,51 +1777,52 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -238719.38 \n", - "Iteration 1: log pseudolikelihood = -28845.056 \n", - "Iteration 2: log pseudolikelihood = -14521.74 \n", - "Iteration 3: log pseudolikelihood = -10894.043 \n", - "Iteration 4: log pseudolikelihood = -10893.979 \n", - "Iteration 5: log pseudolikelihood = -10893.979 \n", + "Iteration 0: log pseudolikelihood = -290091.97 \n", + "Iteration 1: log pseudolikelihood = -36368.036 \n", + "Iteration 2: log pseudolikelihood = -18347.941 \n", + "Iteration 3: log pseudolikelihood = -13295.81 \n", + "Iteration 4: log pseudolikelihood = -13295.279 \n", + "Iteration 5: log pseudolikelihood = -13295.279 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -10197.12 \n", - "Iteration 1: log pseudolikelihood = -9339.4196 \n", - "Iteration 2: log pseudolikelihood = -8794.6573 \n", - "Iteration 3: log pseudolikelihood = -8586.4984 \n", - "Iteration 4: log pseudolikelihood = -8563.6236 \n", - "Iteration 5: log pseudolikelihood = -8563.5725 \n", - "Iteration 6: log pseudolikelihood = -8563.5725 \n", + "Iteration 0: log pseudolikelihood = -12572.459 \n", + "Iteration 1: log pseudolikelihood = -11473.257 \n", + "Iteration 2: log pseudolikelihood = -11337.08 \n", + "Iteration 3: log pseudolikelihood = -10886.708 \n", + "Iteration 4: log pseudolikelihood = -10555.215 \n", + "Iteration 5: log pseudolikelihood = -10531.961 \n", + "Iteration 6: log pseudolikelihood = -10531.538 \n", + "Iteration 7: log pseudolikelihood = -10531.537 \n", "\n", - "Negative binomial regression Number of obs = 2,039\n", - "Dispersion = mean Wald chi2(14) = 384105.40\n", - "Log pseudolikelihood = -8563.5725 Prob > chi2 = 0.0000\n", + "Negative binomial regression Number of obs = 2,520\n", + "Dispersion = mean Wald chi2(14) = 474857.54\n", + "Log pseudolikelihood = -10531.537 Prob > chi2 = 0.0000\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " schoollunch | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 41.4066 1.367529 112.74 0.000 38.8112 44.17555\n", - " levels2 | 40.94799 1.337411 113.66 0.000 38.40885 43.65499\n", - " levels3 | 37.43485 1.260305 107.60 0.000 35.04443 39.98832\n", - " levels4 | 34.79884 1.160151 106.47 0.000 32.59769 37.14863\n", - " levels5 | 33.23793 1.123619 103.64 0.000 31.10705 35.51478\n", - " homeorder | 1.280969 .0284229 11.16 0.000 1.226455 1.337906\n", + " levels1 | 40.67673 1.205293 125.06 0.000 38.38169 43.10901\n", + " levels2 | 37.48201 1.116649 121.64 0.000 35.35609 39.73576\n", + " levels3 | 35.28 1.065025 118.04 0.000 33.25315 37.4304\n", + " levels4 | 33.67241 1.010585 117.17 0.000 31.74882 35.71253\n", + " levels5 | 32.23657 .9852857 113.63 0.000 30.36215 34.22671\n", + " homeorder | 1.314872 .0259834 13.85 0.000 1.264919 1.366798\n", " |\n", " rucc |\n", - " 2 | 1.14593 .0323734 4.82 0.000 1.084204 1.21117\n", - " 3 | 1.107337 .031796 3.55 0.000 1.046739 1.171443\n", - " 4 | 1.187536 .0360939 5.66 0.000 1.118859 1.260429\n", - " 5 | 1.1906 .0511984 4.06 0.000 1.094366 1.295297\n", - " 6 | 1.281665 .0334481 9.51 0.000 1.217756 1.348927\n", - " 7 | 1.223703 .0340166 7.26 0.000 1.158816 1.292224\n", - " 8 | 1.219431 .0437915 5.52 0.000 1.136552 1.308353\n", - " 9 | 1.287903 .0460149 7.08 0.000 1.200801 1.381323\n", + " 2 | 1.137772 .0288759 5.09 0.000 1.08256 1.195799\n", + " 3 | 1.123658 .0287948 4.55 0.000 1.068615 1.181537\n", + " 4 | 1.199803 .0324922 6.73 0.000 1.13778 1.265206\n", + " 5 | 1.219814 .0468732 5.17 0.000 1.131318 1.315232\n", + " 6 | 1.288721 .0298801 10.94 0.000 1.231468 1.348636\n", + " 7 | 1.254022 .0310345 9.15 0.000 1.194647 1.316348\n", + " 8 | 1.279151 .0427903 7.36 0.000 1.197974 1.365829\n", + " 9 | 1.327901 .0434874 8.66 0.000 1.245345 1.41593\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -2.603665 .0428236 -2.687598 -2.519732\n", + " /lnalpha | -2.628985 .0381112 -2.703681 -2.554288\n", "-------------+----------------------------------------------------------------\n", - " alpha | .0740019 .003169 .0680442 .0804812\n", + " alpha | .0721517 .0027498 .0669586 .0777476\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Compare to tabular data:\n", @@ -1754,11 +1834,11 @@ "Highest |\n", "(5) | N(school~h) mean(school~h) sem(school~h)\n", "----------+-----------------------------------------------\n", - " 1 | 392 61.160261 .9604082\n", - " 2 | 399 60.248778 .8795932\n", - " 3 | 415 54.697143 .871953\n", - " 4 | 411 50.260251 .8165952\n", - " 5 | 422 47.847537 .7725881\n", + " 1 | 494 62.302429 .824396\n", + " 2 | 499 56.844878 .7677724\n", + " 3 | 485 53.521044 .8046446\n", + " 4 | 523 50.439806 .7411138\n", + " 5 | 519 47.833955 .7046043\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: you are responsible for interpretation of non-count dep. variable\n", @@ -1766,57 +1846,57 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -10893.979 \n", - "Iteration 1: log pseudolikelihood = -10893.979 \n", + "Iteration 0: log pseudolikelihood = -13295.279 \n", + "Iteration 1: log pseudolikelihood = -13295.279 \n", "\n", "Fitting constant-only model:\n", "\n", - "Iteration 0: log pseudolikelihood = -10217.846 \n", - "Iteration 1: log pseudolikelihood = -9584.6742 \n", - "Iteration 2: log pseudolikelihood = -9426.6102 \n", - "Iteration 3: log pseudolikelihood = -8984.7436 \n", - "Iteration 4: log pseudolikelihood = -8770.0756 \n", - "Iteration 5: log pseudolikelihood = -8769.7148 \n", - "Iteration 6: log pseudolikelihood = -8769.7148 \n", + "Iteration 0: log pseudolikelihood = -12599.584 \n", + "Iteration 1: log pseudolikelihood = -11803.044 \n", + "Iteration 2: log pseudolikelihood = -11687.587 \n", + "Iteration 3: log pseudolikelihood = -11124.475 \n", + "Iteration 4: log pseudolikelihood = -10805.793 \n", + "Iteration 5: log pseudolikelihood = -10804.028 \n", + "Iteration 6: log pseudolikelihood = -10804.028 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -8584.2404 \n", - "Iteration 1: log pseudolikelihood = -8563.6639 \n", - "Iteration 2: log pseudolikelihood = -8563.5726 \n", - "Iteration 3: log pseudolikelihood = -8563.5725 \n", + "Iteration 0: log pseudolikelihood = -10560.506 \n", + "Iteration 1: log pseudolikelihood = -10531.679 \n", + "Iteration 2: log pseudolikelihood = -10531.537 \n", + "Iteration 3: log pseudolikelihood = -10531.537 \n", "\n", - "Negative binomial regression Number of obs = 2,039\n", - " Wald chi2(13) = 412.14\n", + "Negative binomial regression Number of obs = 2,520\n", + " Wald chi2(13) = 555.41\n", "Dispersion = mean Prob > chi2 = 0.0000\n", - "Log pseudolikelihood = -8563.5725 Pseudo R2 = 0.0235\n", + "Log pseudolikelihood = -10531.537 Pseudo R2 = 0.0252\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " schoollunch | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 1.245763 .0271148 10.10 0.000 1.193737 1.300057\n", - " levels2 | 1.231966 .02658 9.67 0.000 1.180956 1.285179\n", - " levels3 | 1.126269 .0251381 5.33 0.000 1.078061 1.176632\n", - " levels4 | 1.046962 .0238353 2.02 0.044 1.001272 1.094736\n", + " levels1 | 1.26182 .0244192 12.02 0.000 1.214855 1.310599\n", + " levels2 | 1.162717 .0229728 7.63 0.000 1.118552 1.208626\n", + " levels3 | 1.094409 .0225473 4.38 0.000 1.051098 1.139506\n", + " levels4 | 1.044541 .0213772 2.13 0.033 1.003471 1.087291\n", " levels5 | 1 (omitted)\n", - " homeorder | 1.280969 .0284229 11.16 0.000 1.226455 1.337906\n", + " homeorder | 1.314872 .0259834 13.85 0.000 1.264919 1.366798\n", " |\n", " rucc |\n", - " 2 | 1.14593 .0323734 4.82 0.000 1.084204 1.21117\n", - " 3 | 1.107337 .031796 3.55 0.000 1.046739 1.171443\n", - " 4 | 1.187536 .0360939 5.66 0.000 1.118859 1.260429\n", - " 5 | 1.1906 .0511984 4.06 0.000 1.094366 1.295297\n", - " 6 | 1.281665 .0334481 9.51 0.000 1.217756 1.348927\n", - " 7 | 1.223703 .0340166 7.26 0.000 1.158816 1.292224\n", - " 8 | 1.219431 .0437915 5.52 0.000 1.136552 1.308353\n", - " 9 | 1.287903 .0460149 7.08 0.000 1.200801 1.381323\n", + " 2 | 1.137772 .0288759 5.09 0.000 1.08256 1.195799\n", + " 3 | 1.123658 .0287948 4.55 0.000 1.068615 1.181537\n", + " 4 | 1.199803 .0324922 6.73 0.000 1.13778 1.265206\n", + " 5 | 1.219814 .0468732 5.17 0.000 1.131318 1.315232\n", + " 6 | 1.288721 .0298801 10.94 0.000 1.231468 1.348636\n", + " 7 | 1.254022 .0310345 9.15 0.000 1.194647 1.316348\n", + " 8 | 1.279151 .0427903 7.36 0.000 1.197974 1.365829\n", + " 9 | 1.327901 .0434874 8.66 0.000 1.245345 1.41593\n", " |\n", - " _cons | 33.23793 1.123619 103.64 0.000 31.10705 35.51478\n", + " _cons | 32.23657 .9852857 113.63 0.000 30.36215 34.22671\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -2.603665 .0428236 -2.687598 -2.519732\n", + " /lnalpha | -2.628985 .0381112 -2.703681 -2.554288\n", "-------------+----------------------------------------------------------------\n", - " alpha | .0740019 .003169 .0680442 .0804812\n", + " alpha | .0721517 .0027498 .0669586 .0777476\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Note: _cons estimates baseline incidence rate.\n" @@ -1824,7 +1904,7 @@ }, { "data": { - "application/pdf": "JVBERi0xLjMKJbe+raoKMSAwIG9iago8PAovVHlwZSAvQ2F0YWxvZwovUGFnZXMgMiAwIFIKPj4KZW5kb2JqCjIgMCBvYmoKPDwKL1R5cGUgL1BhZ2VzCi9LaWRzIFsgNCAwIFIgXQovQ291bnQgMQo+PgplbmRvYmoKMyAwIG9iago8PAovUHJvZHVjZXIgKEhhcnUgRnJlZSBQREYgTGlicmFyeSAyLjQuMGRldikKPj4KZW5kb2JqCjQgMCBvYmoKPDwKL1R5cGUgL1BhZ2UKL01lZGlhQm94IFsgMCAwIDE1OC4zOTk5OSAxNTguMzk5OTkgXQovQ29udGVudHMgNSAwIFIKL1Jlc291cmNlcyA8PAovUHJvY1NldCBbIC9QREYgL1RleHQgL0ltYWdlQiAvSW1hZ2VDIC9JbWFnZUkgXQovRm9udCA8PAovRjEgNyAwIFIKPj4KPj4KL1BhcmVudCAyIDAgUgo+PgplbmRvYmoKNSAwIG9iago8PAovTGVuZ3RoIDYgMCBSCi9GaWx0ZXIgWyAvRmxhdGVEZWNvZGUgXQo+PgpzdHJlYW0NCnicpVTJctswDL3zK3BsD1UBENx6zKTLoZfWPvbi0SRexsskdpvfL0hatpORU0WVZkRAxMPDRmITgpcA2EQO4vKqC8Hj3OCVrZ9fDQICudjYpM+F9HhnbhSHKEnt83shtxvjbZNShKervv9NaykT1a/S3RtLTf1zXJJ1SkpKeimr6cQkpzZB6Qnyqx6JGMELBtgYDsRUlbUan2K9giLyGl0Hq9ognI/RnnFFG4JjzvYdrmqDcMG6dMYV7RnORpViRCrQlTaAxZJo4W1i8rpKQi7+LKkHF2IuVwi6xyIhqjuSkprLBmvD6FggOkWp4hMnkLoz0R6uztQP5mZqPn7RcJkZpvelw5ingIqQ43XegUsRYbox76y8h+nKfJ6aH4PBWgWKFe1HoLWvvqLj29FavyQFLTgC7ZPliuZL9CX0NW/Z04fs0iInVURb0qCPtiY0+zM/e62gbmhKO7uZKUrtXWdjiUHrImpT5c5keHBdqilww1E0RrElLOoJqgzbkbDK4wmDDT4z6lCcOLmvEFKSP2YpzyrxdlayIdJLWttDmw+P62iPynhaRopOaXWOTqzSx1pP6ZG1Kv/B6l1ML1jdmAk+ORTUniW9LIECpuJwsmuXszXcLveH2bZdbuef4Pvu6W5/gF/okPRDcNjBt+V80f10+WffOUyaam/Xol5krFc0Fcp9u9jt1uvf23Zx9vIXfvZ4MwplbmRzdHJlYW0KZW5kb2JqCjYgMCBvYmoKNTMxCmVuZG9iago3IDAgb2JqCjw8Ci9UeXBlIC9Gb250Ci9CYXNlRm9udCAvSGVsdmV0aWNhCi9TdWJ0eXBlIC9UeXBlMQovRW5jb2RpbmcgL1dpbkFuc2lFbmNvZGluZwo+PgplbmRvYmoKeHJlZgowIDgKMDAwMDAwMDAwMCA2NTUzNSBmDQowMDAwMDAwMDE1IDAwMDAwIG4NCjAwMDAwMDAwNjQgMDAwMDAgbg0KMDAwMDAwMDEyMyAwMDAwMCBuDQowMDAwMDAwMTg3IDAwMDAwIG4NCjAwMDAwMDAzNzUgMDAwMDAgbg0KMDAwMDAwMDk4NSAwMDAwMCBuDQowMDAwMDAxMDA0IDAwMDAwIG4NCnRyYWlsZXIKPDwKL1Jvb3QgMSAwIFIKL0luZm8gMyAwIFIKL1NpemUgOAo+PgpzdGFydHhyZWYKMTEwMQolJUVPRgo=", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -1834,17 +1914,17 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t34\n", - "\t36\n", - "\t38\n", - "\t40\n", - "\t42\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t32\n", + "\t34\n", + "\t36\n", + "\t38\n", + "\t40\n", "\tavg\n", "\t\n", "\t\n", @@ -1870,17 +1950,17 @@ "\t<rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/>\n", "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", - "\t<line x1="55.44" y1="1266.85" x2="1342.53" y2="1266.85" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1007.57" x2="1342.53" y2="1007.57" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="748.29" x2="1342.53" y2="748.29" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="489.01" x2="1342.53" y2="489.01" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="229.73" x2="1342.53" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 307.54 L381.95 372.34 L698.96 826.11 L1015.97 1163.15 L1333.03 1370.61" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1364.51" y="1288.04" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">34</text>\n", - "\t<text x="1364.51" y="1028.76" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">36</text>\n", - "\t<text x="1364.51" y="769.48" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">38</text>\n", - "\t<text x="1364.51" y="510.19" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">40</text>\n", - "\t<text x="1364.51" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">42</text>\n", + "\t<line x1="55.44" y1="1370.56" x2="1342.53" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1108.31" x2="1342.53" y2="1108.31" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="846.05" x2="1342.53" y2="846.05" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="583.75" x2="1342.53" y2="583.75" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="321.50" x2="1342.53" y2="321.50" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<path d=" M64.90 229.73 L381.95 649.39 L698.96 937.88 L1015.97 1147.66 L1333.03 1344.37" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">32</text>\n", + "\t<text x="1364.51" y="1129.49" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">34</text>\n", + "\t<text x="1364.51" y="867.24" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">36</text>\n", + "\t<text x="1364.51" y="604.94" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">38</text>\n", + "\t<text x="1364.51" y="342.69" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">40</text>\n", "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1342.53" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -1928,18 +2008,18 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 41.4066 1.367529 112.74 0.000 38.8112 44.17555\n", - " levels2 | 40.94799 1.337411 113.66 0.000 38.40885 43.65499\n", - " levels3 | 37.43485 1.260305 107.60 0.000 35.04443 39.98832\n", - " levels4 | 34.79884 1.160151 106.47 0.000 32.59769 37.14863\n", - " levels5 | 33.23793 1.123619 103.64 0.000 31.10705 35.51478\n", + " levels1 | 40.67673 1.205293 125.06 0.000 38.38169 43.10901\n", + " levels2 | 37.48201 1.116649 121.64 0.000 35.35609 39.73576\n", + " levels3 | 35.28 1.065025 118.04 0.000 33.25315 37.4304\n", + " levels4 | 33.67241 1.010585 117.17 0.000 31.74882 35.71253\n", + " levels5 | 32.23657 .9852857 113.63 0.000 30.36215 34.22671\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | 1.245763 .0271148 10.10 0.000 1.193737 1.300057`\n", + "` levels1 | 1.26182 .0244192 12.02 0.000 1.214855 1.310599`\n", "\n", "\n", - "In the lowest social distancing counties, 41% of schoolage children were eligible for free or reduced price lunches. By comparison, in the most social distancing counties 33% were eligible, after adjusting for rurality and social distancing orders, a 24% (95% CI: 19%, 30%) difference. \n" + "In the lowest social distancing counties, 41% of schoolage children were eligible for free or reduced price lunches. By comparison, in the most social distancing counties 32% were eligible, after adjusting for rurality and social distancing orders, a 26% (95% CI: 21%, 31%) difference. \n" ] }, { @@ -1965,42 +2045,42 @@ "----- RURALITY-ADJUSTED POISSON MODEL -----\n", "note: foodinsec has noninteger values\n", "\n", - "Iteration 0: log likelihood = -5808.946 \n", - "Iteration 1: log likelihood = -5807.0472 \n", - "Iteration 2: log likelihood = -5807.0471 \n", + "Iteration 0: log likelihood = -7184.3437 \n", + "Iteration 1: log likelihood = -7182.0944 \n", + "Iteration 2: log likelihood = -7182.0943 \n", "\n", - "Generalized linear models Number of obs = 2,120\n", - "Optimization : ML Residual df = 2,106\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", " Scale parameter = 1\n", - "Deviance = 2209.125243 (1/df) Deviance = 1.048967\n", - "Pearson = 2297.366836 (1/df) Pearson = 1.090867\n", + "Deviance = 2683.121099 (1/df) Deviance = 1.024483\n", + "Pearson = 2797.25662 (1/df) Pearson = 1.068063\n", "\n", "Variance function: V(u) = u [Poisson]\n", "Link function : g(u) = ln(u) [Log]\n", "\n", - " AIC = 5.491554\n", - "Log likelihood = -5807.047118 BIC = -13921.09\n", + " AIC = 5.46608\n", + "Log likelihood = -7182.094311 BIC = -17943.81\n", "\n", "------------------------------------------------------------------------------\n", " | OIM\n", " foodinsec | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 12.98755 .3612017 92.19 0.000 12.29855 13.71514\n", - " levels2 | 12.55802 .348123 91.28 0.000 11.89392 13.2592\n", - " levels3 | 11.60655 .3183067 89.39 0.000 10.99915 12.24749\n", - " levels4 | 11.0192 .3022595 87.48 0.000 10.44242 11.62783\n", - " levels5 | 10.22625 .2805256 84.75 0.000 9.690954 10.79112\n", - " homeorder | 1.08084 .0205391 4.09 0.000 1.041325 1.121855\n", + " levels1 | 12.93501 .3216935 102.93 0.000 12.31962 13.58114\n", + " levels2 | 11.84488 .2918237 100.33 0.000 11.28651 12.43088\n", + " levels3 | 11.64897 .2858669 100.05 0.000 11.10194 12.22295\n", + " levels4 | 10.89867 .2676669 97.26 0.000 10.38648 11.43612\n", + " levels5 | 10.17634 .2475631 95.37 0.000 9.702515 10.67331\n", + " homeorder | 1.088589 .0183214 5.04 0.000 1.053266 1.125097\n", " |\n", " rucc |\n", - " 2 | 1.091454 .0254729 3.75 0.000 1.042653 1.14254\n", - " 3 | 1.089282 .0261135 3.57 0.000 1.039285 1.141685\n", - " 4 | 1.149078 .0310828 5.14 0.000 1.089743 1.211643\n", - " 5 | 1.110912 .0403183 2.90 0.004 1.034635 1.192813\n", - " 6 | 1.166334 .0248874 7.21 0.000 1.118561 1.216146\n", - " 7 | 1.114854 .0261645 4.63 0.000 1.064734 1.167333\n", - " 8 | 1.181563 .0375657 5.25 0.000 1.110183 1.257533\n", - " 9 | 1.186191 .0389124 5.20 0.000 1.112324 1.264963\n", + " 2 | 1.10029 .0228933 4.59 0.000 1.056322 1.146087\n", + " 3 | 1.104049 .023559 4.64 0.000 1.058826 1.151203\n", + " 4 | 1.149369 .0274078 5.84 0.000 1.096887 1.204363\n", + " 5 | 1.131871 .0367085 3.82 0.000 1.062163 1.206155\n", + " 6 | 1.165615 .022003 8.12 0.000 1.123278 1.209548\n", + " 7 | 1.135392 .0237196 6.08 0.000 1.089841 1.182846\n", + " 8 | 1.219673 .0342862 7.06 0.000 1.154291 1.288758\n", + " 9 | 1.189152 .0350341 5.88 0.000 1.122431 1.259839\n", "------------------------------------------------------------------------------\n", "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", "Compare to tabular data:\n", @@ -2012,54 +2092,54 @@ "Highest |\n", "(5) | N(foodin~c) mean(foodin~c) sem(foodin~c)\n", "----------+-----------------------------------------------\n", - " 1 | 415 15.592289 .2265629\n", - " 2 | 418 15.035646 .1975179\n", - " 3 | 437 13.795423 .1893612\n", - " 4 | 416 12.998077 .1935268\n", - " 5 | 434 11.987327 .1526946\n", + " 1 | 519 15.759152 .1923148\n", + " 2 | 526 14.3327 .1790774\n", + " 3 | 513 14.082261 .174559\n", + " 4 | 540 13.073889 .1640634\n", + " 5 | 535 12.078692 .1441107\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: levels5 omitted because of collinearity\n", "note: foodinsec has noninteger values\n", "\n", - "Iteration 0: log likelihood = -5808.946 \n", - "Iteration 1: log likelihood = -5807.0472 \n", - "Iteration 2: log likelihood = -5807.0471 \n", + "Iteration 0: log likelihood = -7184.3437 \n", + "Iteration 1: log likelihood = -7182.0944 \n", + "Iteration 2: log likelihood = -7182.0943 \n", "\n", - "Generalized linear models Number of obs = 2,120\n", - "Optimization : ML Residual df = 2,106\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", " Scale parameter = 1\n", - "Deviance = 2209.125243 (1/df) Deviance = 1.048967\n", - "Pearson = 2297.366836 (1/df) Pearson = 1.090867\n", + "Deviance = 2683.121099 (1/df) Deviance = 1.024483\n", + "Pearson = 2797.25662 (1/df) Pearson = 1.068063\n", "\n", "Variance function: V(u) = u [Poisson]\n", "Link function : g(u) = ln(u) [Log]\n", "\n", - " AIC = 5.491554\n", - "Log likelihood = -5807.047118 BIC = -13921.09\n", + " AIC = 5.46608\n", + "Log likelihood = -7182.094311 BIC = -17943.81\n", "\n", "------------------------------------------------------------------------------\n", " | OIM\n", " foodinsec | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 1.27002 .0257634 11.78 0.000 1.220515 1.321533\n", - " levels2 | 1.228018 .0247122 10.21 0.000 1.180525 1.27742\n", - " levels3 | 1.134976 .0227503 6.32 0.000 1.09125 1.180453\n", - " levels4 | 1.07754 .0220282 3.65 0.000 1.035219 1.121591\n", + " levels1 | 1.271086 .0226617 13.45 0.000 1.227437 1.316288\n", + " levels2 | 1.163963 .0207984 8.50 0.000 1.123904 1.205449\n", + " levels3 | 1.144711 .0205843 7.52 0.000 1.105069 1.185774\n", + " levels4 | 1.070981 .0192196 3.82 0.000 1.033966 1.109321\n", " levels5 | 1 (omitted)\n", - " homeorder | 1.08084 .0205391 4.09 0.000 1.041325 1.121855\n", + " homeorder | 1.088589 .0183214 5.04 0.000 1.053266 1.125097\n", " |\n", " rucc |\n", - " 2 | 1.091454 .0254729 3.75 0.000 1.042653 1.14254\n", - " 3 | 1.089282 .0261135 3.57 0.000 1.039285 1.141685\n", - " 4 | 1.149078 .0310828 5.14 0.000 1.089743 1.211643\n", - " 5 | 1.110912 .0403183 2.90 0.004 1.034635 1.192813\n", - " 6 | 1.166334 .0248874 7.21 0.000 1.118561 1.216146\n", - " 7 | 1.114854 .0261645 4.63 0.000 1.064734 1.167333\n", - " 8 | 1.181563 .0375657 5.25 0.000 1.110183 1.257533\n", - " 9 | 1.186191 .0389124 5.20 0.000 1.112324 1.264963\n", + " 2 | 1.10029 .0228933 4.59 0.000 1.056322 1.146087\n", + " 3 | 1.104049 .023559 4.64 0.000 1.058826 1.151203\n", + " 4 | 1.149369 .0274078 5.84 0.000 1.096887 1.204363\n", + " 5 | 1.131871 .0367085 3.82 0.000 1.062163 1.206155\n", + " 6 | 1.165615 .022003 8.12 0.000 1.123278 1.209548\n", + " 7 | 1.135392 .0237196 6.08 0.000 1.089841 1.182846\n", + " 8 | 1.219673 .0342862 7.06 0.000 1.154291 1.288758\n", + " 9 | 1.189152 .0350341 5.88 0.000 1.122431 1.259839\n", " |\n", - " _cons | 10.22625 .2805256 84.75 0.000 9.690954 10.79112\n", + " _cons | 10.17634 .2475631 95.37 0.000 9.702515 10.67331\n", "------------------------------------------------------------------------------\n", "Note: _cons estimates baseline incidence rate.\n", "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" @@ -2067,7 +2147,7 @@ }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -2081,7 +2161,7 @@ "\t\n", "\t\n", "\t\n", - "\t\n", + "\t\n", "\t10\n", "\t11\n", "\t12\n", @@ -2115,7 +2195,7 @@ "\t<line x1="55.44" y1="990.30" x2="1342.53" y2="990.30" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="609.99" x2="1342.53" y2="609.99" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="229.73" x2="1342.53" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 229.73 L381.95 381.84 L698.96 762.15 L1015.97 990.30 L1333.03 1294.52" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<path d=" M64.90 267.80 L381.95 686.07 L698.96 762.15 L1015.97 1028.31 L1333.03 1294.52" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">10</text>\n", "\t<text x="1364.51" y="1011.48" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">11</text>\n", "\t<text x="1364.51" y="631.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">12</text>\n", @@ -2167,18 +2247,18 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 12.98755 .3612017 92.19 0.000 12.29855 13.71514\n", - " levels2 | 12.55802 .348123 91.28 0.000 11.89392 13.2592\n", - " levels3 | 11.60655 .3183067 89.39 0.000 10.99915 12.24749\n", - " levels4 | 11.0192 .3022595 87.48 0.000 10.44242 11.62783\n", - " levels5 | 10.22625 .2805256 84.75 0.000 9.690954 10.79112\n", + " levels1 | 12.93501 .3216935 102.93 0.000 12.31962 13.58114\n", + " levels2 | 11.84488 .2918237 100.33 0.000 11.28651 12.43088\n", + " levels3 | 11.64897 .2858669 100.05 0.000 11.10194 12.22295\n", + " levels4 | 10.89867 .2676669 97.26 0.000 10.38648 11.43612\n", + " levels5 | 10.17634 .2475631 95.37 0.000 9.702515 10.67331\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | 1.27002 .0257634 11.78 0.000 1.220515 1.321533`\n", + "` levels1 | 1.271086 .0226617 13.45 0.000 1.227437 1.316288`\n", "\n", "\n", - "The lowest social distancing counties had greater food insecurity, among 13.0% of residents. The most social distancing counties had 10.2%, after adjusting for rurality and social distancing orders, a 27% (95% CI: 22%, 32%) difference. \n", + "The lowest social distancing counties had greater food insecurity, among 12.9% of residents. The most social distancing counties had 10.2%, after adjusting for rurality and social distancing orders, a 27% (95% CI: 23%, 32%) difference. \n", "\n", "---" ] @@ -2204,50 +2284,50 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -323193.5 \n", - "Iteration 1: log pseudolikelihood = -19034.538 \n", - "Iteration 2: log pseudolikelihood = -13266.254 \n", - "Iteration 3: log pseudolikelihood = -13060.277 \n", - "Iteration 4: log pseudolikelihood = -13060.097 \n", - "Iteration 5: log pseudolikelihood = -13060.097 \n", + "Iteration 0: log pseudolikelihood = -397571.32 \n", + "Iteration 1: log pseudolikelihood = -24410.221 \n", + "Iteration 2: log pseudolikelihood = -16708.317 \n", + "Iteration 3: log pseudolikelihood = -16405.591 \n", + "Iteration 4: log pseudolikelihood = -16405.296 \n", + "Iteration 5: log pseudolikelihood = -16405.296 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -10987.976 \n", - "Iteration 1: log pseudolikelihood = -10172.876 \n", - "Iteration 2: log pseudolikelihood = -9662.7 \n", - "Iteration 3: log pseudolikelihood = -9608.1825 \n", - "Iteration 4: log pseudolikelihood = -9607.3544 \n", - "Iteration 5: log pseudolikelihood = -9607.3539 \n", + "Iteration 0: log pseudolikelihood = -13629.269 \n", + "Iteration 1: log pseudolikelihood = -12618.453 \n", + "Iteration 2: log pseudolikelihood = -12003.767 \n", + "Iteration 3: log pseudolikelihood = -11968.11 \n", + "Iteration 4: log pseudolikelihood = -11967.878 \n", + "Iteration 5: log pseudolikelihood = -11967.878 \n", "\n", - "Negative binomial regression Number of obs = 2,120\n", - "Dispersion = mean Wald chi2(14) = 593207.35\n", - "Log pseudolikelihood = -9607.3539 Prob > chi2 = 0.0000\n", + "Negative binomial regression Number of obs = 2,633\n", + "Dispersion = mean Wald chi2(14) = 725576.80\n", + "Log pseudolikelihood = -11967.878 Prob > chi2 = 0.0000\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " exercise | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 68.69511 2.008972 144.63 0.000 64.86832 72.74766\n", - " levels2 | 74.2871 2.075381 154.20 0.000 70.32878 78.46819\n", - " levels3 | 77.96532 2.065593 164.43 0.000 74.02015 82.12077\n", - " levels4 | 82.76289 2.142898 170.55 0.000 78.66768 87.07129\n", - " levels5 | 90.49707 2.206371 184.79 0.000 86.27436 94.92647\n", - " homeorder | .9328766 .0185927 -3.49 0.000 .8971381 .9700387\n", + " levels1 | 70.82756 1.902913 158.57 0.000 67.19442 74.65715\n", + " levels2 | 76.05735 1.841866 178.86 0.000 72.5317 79.75439\n", + " levels3 | 78.16616 1.906506 178.71 0.000 74.51738 81.9936\n", + " levels4 | 82.19944 1.871535 193.65 0.000 78.61194 85.95066\n", + " levels5 | 91.41497 2.000766 206.31 0.000 87.57646 95.42173\n", + " homeorder | .9263275 .0160495 -4.42 0.000 .8953991 .9583242\n", " |\n", " rucc |\n", - " 2 | .9403625 .0188896 -3.06 0.002 .9040589 .978124\n", - " 3 | .9400426 .0209179 -2.78 0.005 .8999254 .9819481\n", - " 4 | .9180416 .0189869 -4.13 0.000 .8815722 .9560198\n", - " 5 | 1.03397 .025211 1.37 0.171 .9857195 1.084582\n", - " 6 | .7913411 .0163763 -11.31 0.000 .7598864 .8240978\n", - " 7 | .8913811 .0202799 -5.05 0.000 .8525064 .9320285\n", - " 8 | .6113286 .0348599 -8.63 0.000 .5466842 .6836172\n", - " 9 | .6968747 .0363956 -6.92 0.000 .6290701 .7719877\n", + " 2 | .9315932 .0170231 -3.88 0.000 .8988189 .9655625\n", + " 3 | .9221747 .0189775 -3.94 0.000 .8857196 .9601303\n", + " 4 | .9217404 .0170166 -4.41 0.000 .8889847 .955703\n", + " 5 | 1.031934 .0218027 1.49 0.137 .9900745 1.075564\n", + " 6 | .7820271 .0142184 -13.52 0.000 .7546503 .8103971\n", + " 7 | .864374 .0185513 -6.79 0.000 .8287684 .9015094\n", + " 8 | .6065847 .0300327 -10.10 0.000 .5504875 .6683983\n", + " 9 | .6802969 .0328991 -7.97 0.000 .6187774 .7479328\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -2.267749 .0702183 -2.405374 -2.130124\n", + " /lnalpha | -2.222633 .0655469 -2.351102 -2.094163\n", "-------------+----------------------------------------------------------------\n", - " alpha | .103545 .0072708 .0902317 .1188226\n", + " alpha | .1083235 .0071003 .0952641 .1231733\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Compare to tabular data:\n", @@ -2259,11 +2339,11 @@ "Highest |\n", "(5) | N(exercise) mean(exercise) sem(exercise)\n", "----------+-----------------------------------------------\n", - " 1 | 415 55.966578 1.113067\n", - " 2 | 418 61.128548 .9513782\n", - " 3 | 437 64.817623 .9621726\n", - " 4 | 416 69.980657 .9676512\n", - " 5 | 434 77.43793 .8734737\n", + " 1 | 519 56.906538 .9544925\n", + " 2 | 526 61.787498 .9064765\n", + " 3 | 513 63.396324 .89163\n", + " 4 | 540 68.048161 .8530419\n", + " 5 | 535 76.76011 .8015522\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: you are responsible for interpretation of non-count dep. variable\n", @@ -2271,56 +2351,56 @@ "\n", "Fitting Poisson model:\n", "\n", - "Iteration 0: log pseudolikelihood = -13060.104 \n", - "Iteration 1: log pseudolikelihood = -13060.097 \n", - "Iteration 2: log pseudolikelihood = -13060.097 \n", + "Iteration 0: log pseudolikelihood = -16405.303 \n", + "Iteration 1: log pseudolikelihood = -16405.296 \n", + "Iteration 2: log pseudolikelihood = -16405.296 \n", "\n", "Fitting constant-only model:\n", "\n", - "Iteration 0: log pseudolikelihood = -11016.592 \n", - "Iteration 1: log pseudolikelihood = -10369.373 \n", - "Iteration 2: log pseudolikelihood = -9825.0325 \n", - "Iteration 3: log pseudolikelihood = -9824.4905 \n", - "Iteration 4: log pseudolikelihood = -9824.4905 \n", + "Iteration 0: log pseudolikelihood = -13662.944 \n", + "Iteration 1: log pseudolikelihood = -12842.373 \n", + "Iteration 2: log pseudolikelihood = -12215.778 \n", + "Iteration 3: log pseudolikelihood = -12215.652 \n", + "Iteration 4: log pseudolikelihood = -12215.652 \n", "\n", "Fitting full model:\n", "\n", - "Iteration 0: log pseudolikelihood = -9631.244 \n", - "Iteration 1: log pseudolikelihood = -9607.5319 \n", - "Iteration 2: log pseudolikelihood = -9607.3539 \n", - "Iteration 3: log pseudolikelihood = -9607.3539 \n", + "Iteration 0: log pseudolikelihood = -11993.273 \n", + "Iteration 1: log pseudolikelihood = -11968.037 \n", + "Iteration 2: log pseudolikelihood = -11967.878 \n", + "Iteration 3: log pseudolikelihood = -11967.878 \n", "\n", - "Negative binomial regression Number of obs = 2,120\n", - " Wald chi2(13) = 563.65\n", + "Negative binomial regression Number of obs = 2,633\n", + " Wald chi2(13) = 690.24\n", "Dispersion = mean Prob > chi2 = 0.0000\n", - "Log pseudolikelihood = -9607.3539 Pseudo R2 = 0.0221\n", + "Log pseudolikelihood = -11967.878 Pseudo R2 = 0.0203\n", "\n", "------------------------------------------------------------------------------\n", " | Robust\n", " exercise | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | .7590865 .0177023 -11.82 0.000 .7251717 .7945875\n", - " levels2 | .8208786 .0161365 -10.04 0.000 .7898531 .8531228\n", - " levels3 | .8615231 .0155678 -8.25 0.000 .8315449 .8925822\n", - " levels4 | .9145367 .0152978 -5.34 0.000 .8850396 .9450168\n", + " levels1 | .7747917 .0157072 -12.59 0.000 .7446098 .8061969\n", + " levels2 | .8320011 .0150512 -10.17 0.000 .803018 .8620301\n", + " levels3 | .8550695 .0146972 -9.11 0.000 .8267433 .8843663\n", + " levels4 | .8991901 .0137357 -6.96 0.000 .8726677 .9265186\n", " levels5 | 1 (omitted)\n", - " homeorder | .9328764 .0185927 -3.49 0.000 .8971379 .9700385\n", + " homeorder | .9263275 .0160495 -4.42 0.000 .8953991 .9583241\n", " |\n", " rucc |\n", - " 2 | .9403626 .0188896 -3.06 0.002 .9040589 .978124\n", - " 3 | .9400426 .0209179 -2.78 0.005 .8999254 .9819482\n", - " 4 | .9180418 .0189869 -4.13 0.000 .8815723 .95602\n", - " 5 | 1.03397 .025211 1.37 0.171 .9857196 1.084582\n", - " 6 | .7913412 .0163763 -11.31 0.000 .7598865 .824098\n", - " 7 | .8913813 .0202799 -5.05 0.000 .8525066 .9320287\n", - " 8 | .6113286 .0348599 -8.63 0.000 .5466842 .6836172\n", - " 9 | .6968747 .0363956 -6.92 0.000 .6290701 .7719877\n", + " 2 | .9315932 .0170231 -3.88 0.000 .8988189 .9655625\n", + " 3 | .9221747 .0189775 -3.94 0.000 .8857196 .9601303\n", + " 4 | .9217404 .0170166 -4.41 0.000 .8889847 .955703\n", + " 5 | 1.031934 .0218027 1.49 0.137 .9900745 1.075564\n", + " 6 | .7820271 .0142184 -13.52 0.000 .7546503 .8103971\n", + " 7 | .864374 .0185513 -6.79 0.000 .8287684 .9015094\n", + " 8 | .6065847 .0300327 -10.10 0.000 .5504875 .6683983\n", + " 9 | .6802969 .0328991 -7.97 0.000 .6187774 .7479328\n", " |\n", - " _cons | 90.49708 2.206371 184.79 0.000 86.27437 94.92647\n", + " _cons | 91.41497 2.000766 206.31 0.000 87.57646 95.42173\n", "-------------+----------------------------------------------------------------\n", - " /lnalpha | -2.267748 .0702183 -2.405374 -2.130123\n", + " /lnalpha | -2.222633 .0655469 -2.351102 -2.094163\n", "-------------+----------------------------------------------------------------\n", - " alpha | .1035451 .0072708 .0902318 .1188227\n", + " alpha | .1083235 .0071003 .0952641 .1231733\n", "------------------------------------------------------------------------------\n", "Note: Estimates are transformed only in the first equation.\n", "Note: _cons estimates baseline incidence rate.\n" @@ -2328,7 +2408,7 @@ }, { "data": { - "application/pdf": "JVBERi0xLjMKJbe+raoKMSAwIG9iago8PAovVHlwZSAvQ2F0YWxvZwovUGFnZXMgMiAwIFIKPj4KZW5kb2JqCjIgMCBvYmoKPDwKL1R5cGUgL1BhZ2VzCi9LaWRzIFsgNCAwIFIgXQovQ291bnQgMQo+PgplbmRvYmoKMyAwIG9iago8PAovUHJvZHVjZXIgKEhhcnUgRnJlZSBQREYgTGlicmFyeSAyLjQuMGRldikKPj4KZW5kb2JqCjQgMCBvYmoKPDwKL1R5cGUgL1BhZ2UKL01lZGlhQm94IFsgMCAwIDE1OC4zOTk5OSAxNTguMzk5OTkgXQovQ29udGVudHMgNSAwIFIKL1Jlc291cmNlcyA8PAovUHJvY1NldCBbIC9QREYgL1RleHQgL0ltYWdlQiAvSW1hZ2VDIC9JbWFnZUkgXQovRm9udCA8PAovRjEgNyAwIFIKPj4KPj4KL1BhcmVudCAyIDAgUgo+PgplbmRvYmoKNSAwIG9iago8PAovTGVuZ3RoIDYgMCBSCi9GaWx0ZXIgWyAvRmxhdGVEZWNvZGUgXQo+PgpzdHJlYW0NCnicpZRNb9swDIbv+hU8bod5JCXqY8diX4ddtvq4SxC4mYumwZJg3c8fJcVOUjhr5tmARdp6+VAiZWxC8C4ANpGDkzzqQLBdGbzw6dsng4BAEhub9Dqxtp25UR2iSzo/3yf2cm28bVKK8HQx9stYSxlUn4q7M5aa+uYwJCsKJYWe2jr11iTROUHxBPnWiESMID56WBsOxFSdB5085npBRZgCj7LqXaXzLPGoK941OiZxbtRV7yqdj5aOuuKd6WxUK0akIr3XArCz5HTjbWLyOrqEXOJZInD5sTYh6CdCtnnV5MrKHDmrHqOwA1JxXhn7xAk4WEmFi4oY2T/NTWveftR8mRnau1JizG1AxdCEg3gBYVW3a/Mq4Gto782H1ny9Wkwo2kdFLTPUkncsq+MMNhNRqOoZbPYOXVGnM/ap9G/RcqQ3OaRFTrlAUfREagv4EnTxa3WMWkVD15SCDk1TnFq7YY4lBic+N2S1hynXJzcsVQ9Ow9Fpjto9OS2aSKq02wFY7fnAYIPPRG3PkclTG+HK4g+rdGc78e9UsiHSc6ydwObjIwP24MzHMlIUxeo5HaluilqO6UCtzn9QvcT0jCpzOngM6FBrlnyMQAHrv+B2s+wXD/C+3+0Xj8v+cfUOvmyeut0evqMg6YNgv4HP/erH8FLyy6lzmHSpk1WL+ivjOPwCut/ddtnvumOIP31xdxYKZW5kc3RyZWFtCmVuZG9iago2IDAgb2JqCjUyOQplbmRvYmoKNyAwIG9iago8PAovVHlwZSAvRm9udAovQmFzZUZvbnQgL0hlbHZldGljYQovU3VidHlwZSAvVHlwZTEKL0VuY29kaW5nIC9XaW5BbnNpRW5jb2RpbmcKPj4KZW5kb2JqCnhyZWYKMCA4CjAwMDAwMDAwMDAgNjU1MzUgZg0KMDAwMDAwMDAxNSAwMDAwMCBuDQowMDAwMDAwMDY0IDAwMDAwIG4NCjAwMDAwMDAxMjMgMDAwMDAgbg0KMDAwMDAwMDE4NyAwMDAwMCBuDQowMDAwMDAwMzc1IDAwMDAwIG4NCjAwMDAwMDA5ODMgMDAwMDAgbg0KMDAwMDAwMTAwMiAwMDAwMCBuDQp0cmFpbGVyCjw8Ci9Sb290IDEgMCBSCi9JbmZvIDMgMCBSCi9TaXplIDgKPj4Kc3RhcnR4cmVmCjEwOTkKJSVFT0YK", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -2338,17 +2418,17 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t70\n", - "\t75\n", - "\t80\n", - "\t85\n", - "\t90\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t70\n", + "\t75\n", + "\t80\n", + "\t85\n", + "\t90\n", "\tavg\n", "\t\n", "\t\n", @@ -2374,17 +2454,17 @@ "\t<rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/>\n", "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", - "\t<line x1="55.44" y1="1302.54" x2="1342.53" y2="1302.54" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1040.89" x2="1342.53" y2="1040.89" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="779.23" x2="1342.53" y2="779.23" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="517.57" x2="1342.53" y2="517.57" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="255.87" x2="1342.53" y2="255.87" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 1370.61 L381.95 1077.52 L698.96 883.92 L1015.97 632.71 L1333.03 229.73" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1364.51" y="1323.73" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">70</text>\n", - "\t<text x="1364.51" y="1062.07" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">75</text>\n", - "\t<text x="1364.51" y="800.41" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">80</text>\n", - "\t<text x="1364.51" y="538.76" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">85</text>\n", - "\t<text x="1364.51" y="277.05" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">90</text>\n", + "\t<line x1="55.44" y1="1370.56" x2="1342.53" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1104.00" x2="1342.53" y2="1104.00" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="837.44" x2="1342.53" y2="837.44" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="570.93" x2="1342.53" y2="570.93" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="304.38" x2="1342.53" y2="304.38" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<path d=" M64.90 1327.94 L381.95 1045.39 L698.96 933.42 L1015.97 720.23 L1333.03 229.73" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">70</text>\n", + "\t<text x="1364.51" y="1125.18" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">75</text>\n", + "\t<text x="1364.51" y="858.62" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">80</text>\n", + "\t<text x="1364.51" y="592.12" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">85</text>\n", + "\t<text x="1364.51" y="325.56" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">90</text>\n", "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1342.53" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -2432,15 +2512,15 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 68.69511 2.008972 144.63 0.000 64.86832 72.74766\n", - " levels2 | 74.2871 2.075381 154.20 0.000 70.32878 78.46819\n", - " levels3 | 77.96532 2.065593 164.43 0.000 74.02015 82.12077\n", - " levels4 | 82.76289 2.142898 170.55 0.000 78.66768 87.07129\n", - " levels5 | 90.49707 2.206371 184.79 0.000 86.27436 94.92647\n", + " levels1 | 70.82756 1.902913 158.57 0.000 67.19442 74.65715\n", + " levels2 | 76.05735 1.841866 178.86 0.000 72.5317 79.75439\n", + " levels3 | 78.16616 1.906506 178.71 0.000 74.51738 81.9936\n", + " levels4 | 82.19944 1.871535 193.65 0.000 78.61194 85.95066\n", + " levels5 | 91.41497 2.000766 206.31 0.000 87.57646 95.42173\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | .7590865 .0177023 -11.82 0.000 .7251717 .7945875`\n", + "` levels1 | .7747917 .0157072 -12.59 0.000 .7446098 .8061969`\n", "\n", "\n", "In the lowest social distancing counties, 69% of residents had access to physical spaces for exercise, whereas in the most social distancing counties 90% had access, after adjusting for rurality and social distancing orders, a 32% (95% CI: 26%, 40%) difference. " @@ -2455,14 +2535,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "31.7\n", - "LL: 37.9\n", - "UL: 25.9\n" + "29.1\n", + "LL: 34.3\n", + "UL: 24\n" ] } ], "source": [ - "invert .7590865 .7251717 .7945875" + "invert .7747917 .7446098 .8061969" ] }, { @@ -2486,43 +2566,43 @@ "----- RURALITY-ADJUSTED POISSON MODEL -----\n", "note: overcrowding has noninteger values\n", "\n", - "Iteration 0: log likelihood = -3949.1959 \n", - "Iteration 1: log likelihood = -3938.7057 \n", - "Iteration 2: log likelihood = -3938.6942 \n", - "Iteration 3: log likelihood = -3938.6942 \n", + "Iteration 0: log likelihood = -4756.5858 \n", + "Iteration 1: log likelihood = -4744.9748 \n", + "Iteration 2: log likelihood = -4744.9632 \n", + "Iteration 3: log likelihood = -4744.9632 \n", "\n", - "Generalized linear models Number of obs = 2,120\n", - "Optimization : ML Residual df = 2,106\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", " Scale parameter = 1\n", - "Deviance = 2312.719078 (1/df) Deviance = 1.098157\n", - "Pearson = 3076.380884 (1/df) Pearson = 1.46077\n", + "Deviance = 2650.197886 (1/df) Deviance = 1.011912\n", + "Pearson = 3524.251156 (1/df) Pearson = 1.345648\n", "\n", "Variance function: V(u) = u [Poisson]\n", "Link function : g(u) = ln(u) [Log]\n", "\n", - " AIC = 3.728957\n", - "Log likelihood = -3938.694205 BIC = -13817.5\n", + " AIC = 3.61486\n", + "Log likelihood = -4744.963158 BIC = -17976.73\n", "\n", "------------------------------------------------------------------------------\n", " | OIM\n", "overcrowding | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 2.340737 .179192 11.11 0.000 2.014605 2.719663\n", - " levels2 | 2.256858 .1718827 10.69 0.000 1.943912 2.620184\n", - " levels3 | 2.106022 .1585213 9.89 0.000 1.817158 2.440806\n", - " levels4 | 2.077067 .155689 9.75 0.000 1.793278 2.405766\n", - " levels5 | 2.010469 .1498266 9.37 0.000 1.737253 2.326653\n", - " homeorder | 1.19005 .0650093 3.19 0.001 1.069218 1.324537\n", + " levels1 | 2.367699 .1589751 12.84 0.000 2.075745 2.700717\n", + " levels2 | 2.162979 .1437291 11.61 0.000 1.898848 2.46385\n", + " levels3 | 1.876126 .1261733 9.36 0.000 1.644436 2.14046\n", + " levels4 | 1.934634 .1285511 9.93 0.000 1.698396 2.203733\n", + " levels5 | 2.068613 .1338108 11.24 0.000 1.822293 2.348229\n", + " homeorder | 1.167809 .055477 3.27 0.001 1.063985 1.281765\n", " |\n", " rucc |\n", - " 2 | 1.025268 .0622453 0.41 0.681 .9102482 1.154821\n", - " 3 | .9320338 .0598093 -1.10 0.273 .821882 1.056949\n", - " 4 | .9217681 .0686173 -1.09 0.274 .7966313 1.066562\n", - " 5 | 1.153511 .1071488 1.54 0.124 .9615106 1.38385\n", - " 6 | 1.024312 .0576045 0.43 0.669 .9174087 1.143671\n", - " 7 | .9856186 .0614556 -0.23 0.816 .872237 1.113739\n", - " 8 | .9112646 .0822027 -1.03 0.303 .7635894 1.0875\n", - " 9 | .8039023 .0792778 -2.21 0.027 .6626147 .9753164\n", + " 2 | 1.029311 .055311 0.54 0.591 .9264173 1.143634\n", + " 3 | .9498813 .0537281 -0.91 0.363 .8502035 1.061245\n", + " 4 | .9483685 .0611599 -0.82 0.411 .8357637 1.076145\n", + " 5 | 1.15092 .0952304 1.70 0.089 .97862 1.353555\n", + " 6 | 1.04307 .051264 0.86 0.391 .947282 1.148544\n", + " 7 | .9874639 .0545025 -0.23 0.819 .8862162 1.100279\n", + " 8 | 1.001858 .0774844 0.02 0.981 .8609413 1.165839\n", + " 9 | .817196 .0710091 -2.32 0.020 .6892269 .9689251\n", "------------------------------------------------------------------------------\n", "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", "Compare to tabular data:\n", @@ -2534,55 +2614,55 @@ "Highest |\n", "(5) | N(overcr~g) mean(overcr~g) sem(overcr~g)\n", "----------+-----------------------------------------------\n", - " 1 | 415 2.6399799 .1043603\n", - " 2 | 418 2.5857132 .0877558\n", - " 3 | 437 2.4125517 .0883216\n", - " 4 | 416 2.3948265 .09872\n", - " 5 | 434 2.3430135 .0888598\n", + " 1 | 519 2.6832882 .0894385\n", + " 2 | 526 2.4762081 .0830075\n", + " 3 | 513 2.1469685 .0588481\n", + " 4 | 540 2.2162362 .0747709\n", + " 5 | 535 2.388474 .0858484\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", "note: levels5 omitted because of collinearity\n", "note: overcrowding has noninteger values\n", "\n", - "Iteration 0: log likelihood = -3949.1959 \n", - "Iteration 1: log likelihood = -3938.7057 \n", - "Iteration 2: log likelihood = -3938.6942 \n", - "Iteration 3: log likelihood = -3938.6942 \n", + "Iteration 0: log likelihood = -4756.5858 \n", + "Iteration 1: log likelihood = -4744.9748 \n", + "Iteration 2: log likelihood = -4744.9632 \n", + "Iteration 3: log likelihood = -4744.9632 \n", "\n", - "Generalized linear models Number of obs = 2,120\n", - "Optimization : ML Residual df = 2,106\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", " Scale parameter = 1\n", - "Deviance = 2312.719078 (1/df) Deviance = 1.098157\n", - "Pearson = 3076.380884 (1/df) Pearson = 1.46077\n", + "Deviance = 2650.197886 (1/df) Deviance = 1.011912\n", + "Pearson = 3524.251156 (1/df) Pearson = 1.345648\n", "\n", "Variance function: V(u) = u [Poisson]\n", "Link function : g(u) = ln(u) [Log]\n", "\n", - " AIC = 3.728957\n", - "Log likelihood = -3938.694205 BIC = -13817.5\n", + " AIC = 3.61486\n", + "Log likelihood = -4744.963158 BIC = -17976.73\n", "\n", "------------------------------------------------------------------------------\n", " | OIM\n", "overcrowding | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 1.164274 .0640694 2.76 0.006 1.045235 1.29687\n", - " levels2 | 1.122553 .061054 2.13 0.034 1.009047 1.248828\n", - " levels3 | 1.047528 .0564633 0.86 0.389 .9425072 1.164251\n", - " levels4 | 1.033126 .0561102 0.60 0.548 .9288029 1.149166\n", + " levels1 | 1.144583 .0535491 2.89 0.004 1.044297 1.2545\n", + " levels2 | 1.045618 .0488222 0.96 0.339 .9541759 1.145823\n", + " levels3 | .9069486 .0439847 -2.01 0.044 .8247106 .9973871\n", + " levels4 | .9352325 .0440449 -1.42 0.155 .8527704 1.025669\n", " levels5 | 1 (omitted)\n", - " homeorder | 1.19005 .0650093 3.19 0.001 1.069218 1.324537\n", + " homeorder | 1.167809 .055477 3.27 0.001 1.063985 1.281765\n", " |\n", " rucc |\n", - " 2 | 1.025268 .0622453 0.41 0.681 .9102482 1.154821\n", - " 3 | .9320338 .0598093 -1.10 0.273 .821882 1.056949\n", - " 4 | .9217681 .0686173 -1.09 0.274 .7966313 1.066562\n", - " 5 | 1.153511 .1071488 1.54 0.124 .9615106 1.38385\n", - " 6 | 1.024312 .0576045 0.43 0.669 .9174087 1.143671\n", - " 7 | .9856186 .0614556 -0.23 0.816 .872237 1.113739\n", - " 8 | .9112646 .0822027 -1.03 0.303 .7635894 1.0875\n", - " 9 | .8039023 .0792778 -2.21 0.027 .6626147 .9753164\n", + " 2 | 1.029311 .055311 0.54 0.591 .9264173 1.143634\n", + " 3 | .9498813 .0537281 -0.91 0.363 .8502035 1.061245\n", + " 4 | .9483685 .0611599 -0.82 0.411 .8357637 1.076145\n", + " 5 | 1.15092 .0952304 1.70 0.089 .97862 1.353555\n", + " 6 | 1.04307 .051264 0.86 0.391 .947282 1.148544\n", + " 7 | .9874639 .0545025 -0.23 0.819 .8862162 1.100279\n", + " 8 | 1.001858 .0774844 0.02 0.981 .8609413 1.165839\n", + " 9 | .817196 .0710091 -2.32 0.020 .6892269 .9689251\n", " |\n", - " _cons | 2.010469 .1498266 9.37 0.000 1.737253 2.326653\n", + " _cons | 2.068613 .1338108 11.24 0.000 1.822293 2.348229\n", "------------------------------------------------------------------------------\n", "Note: _cons estimates baseline incidence rate.\n", "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" @@ -2590,7 +2670,7 @@ }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -2601,14 +2681,18 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", "\t\n", - "\t\n", - "\t2\n", - "\t2.1\n", - "\t2.2\n", - "\t2.3\n", + "\t\n", + "\t1.9\n", + "\t2\n", + "\t2.1\n", + "\t2.2\n", + "\t2.3\n", + "\t2.4\n", "\tavg\n", "\t\n", "\t\n", @@ -2635,14 +2719,18 @@ "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", "\t<line x1="55.44" y1="1370.56" x2="1324.12" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="990.30" x2="1324.12" y2="990.30" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="609.99" x2="1324.12" y2="609.99" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1142.41" x2="1324.12" y2="1142.41" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="914.22" x2="1324.12" y2="914.22" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="686.07" x2="1324.12" y2="686.07" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="457.88" x2="1324.12" y2="457.88" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="229.73" x2="1324.12" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 229.73 L377.35 229.73 L689.75 990.30 L1002.16 990.30 L1314.56 1370.61" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1346.10" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2</text>\n", - "\t<text x="1346.10" y="1011.48" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.1</text>\n", - "\t<text x="1346.10" y="631.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.2</text>\n", - "\t<text x="1346.10" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.3</text>\n", + "\t<path d=" M64.90 229.73 L377.35 686.07 L689.75 1370.61 L1002.16 1370.61 L1314.56 914.26" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1346.10" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">1.9</text>\n", + "\t<text x="1346.10" y="1163.59" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2</text>\n", + "\t<text x="1346.10" y="935.40" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.1</text>\n", + "\t<text x="1346.10" y="707.25" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.2</text>\n", + "\t<text x="1346.10" y="479.06" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.3</text>\n", + "\t<text x="1346.10" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">2.4</text>\n", "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", "\t<line x1="55.44" y1="1380.06" x2="1324.12" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", @@ -2690,17 +2778,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 2.340737 .179192 11.11 0.000 2.014605 2.719663\n", - " levels2 | 2.256858 .1718827 10.69 0.000 1.943912 2.620184\n", - " levels3 | 2.106022 .1585213 9.89 0.000 1.817158 2.440806\n", - " levels4 | 2.077067 .155689 9.75 0.000 1.793278 2.405766\n", - " levels5 | 2.010469 .1498266 9.37 0.000 1.737253 2.326653\n", + " levels1 | 2.367699 .1589751 12.84 0.000 2.075745 2.700717\n", + " levels2 | 2.162979 .1437291 11.61 0.000 1.898848 2.46385\n", + " levels3 | 1.876126 .1261733 9.36 0.000 1.644436 2.14046\n", + " levels4 | 1.934634 .1285511 9.93 0.000 1.698396 2.203733\n", + " levels5 | 2.068613 .1338108 11.24 0.000 1.822293 2.348229\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | 1.164274 .0640694 2.76 0.006 1.045235 1.29687`\n", + "` levels1 | 1.144583 .0535491 2.89 0.004 1.044297 1.2545`\n", "\n", - "The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders. " + "The lowest social distancing counties had 14% (95% CI: 4.4%, 25%) less overcrowding, after adjusting for rurality and social distancing orders. " ] }, { @@ -2708,8 +2796,11 @@ "metadata": {}, "source": [ "---\n", + "# Sociodemographics\n", "\n", - "# Exploratory analyses" + "## Youth\n", + "\n", + "Less than age 18" ] }, { @@ -2721,27 +2812,112 @@ "name": "stdout", "output_type": "stream", "text": [ + "----- RURALITY-ADJUSTED POISSON MODEL -----\n", + "note: youth has noninteger values\n", + "\n", + "Iteration 0: log likelihood = -7047.4698 \n", + "Iteration 1: log likelihood = -7047.2764 \n", + "Iteration 2: log likelihood = -7047.2764 \n", "\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", + " Scale parameter = 1\n", + "Deviance = 1084.966828 (1/df) Deviance = .4142676\n", + "Pearson = 1091.163655 (1/df) Pearson = .4166337\n", "\n", + "Variance function: V(u) = u [Poisson]\n", + "Link function : g(u) = ln(u) [Log]\n", "\n", - "----------------------------------------------------------\n", + " AIC = 5.363674\n", + "Log likelihood = -7047.276406 BIC = -19541.96\n", + "\n", + "------------------------------------------------------------------------------\n", + " | OIM\n", + " youth | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", + "-------------+----------------------------------------------------------------\n", + " levels1 | 25.54333 .3027171 273.42 0.000 24.95686 26.14359\n", + " levels2 | 24.97196 .2904891 276.61 0.000 24.40905 25.54785\n", + " levels3 | 24.65616 .2854101 276.88 0.000 24.10306 25.22195\n", + " levels4 | 24.76482 .2843834 279.48 0.000 24.21366 25.32852\n", + " levels5 | 23.59149 .2675883 278.67 0.000 23.07281 24.12183\n", + " homeorder | .9355478 .0075003 -8.31 0.000 .9209623 .9503643\n", + " |\n", + " rucc |\n", + " 2 | .9687538 .009452 -3.25 0.001 .9504042 .9874577\n", + " 3 | .9474401 .0095882 -5.34 0.000 .9288327 .9664203\n", + " 4 | .9468317 .0109397 -4.73 0.000 .9256313 .9685176\n", + " 5 | .9792884 .015412 -1.33 0.184 .9495426 1.009966\n", + " 6 | .9535839 .0085333 -5.31 0.000 .9370048 .9704565\n", + " 7 | .9514734 .009464 -5.00 0.000 .933104 .9702044\n", + " 8 | .9060676 .0129889 -6.88 0.000 .8809641 .9318865\n", + " 9 | .8965563 .0133864 -7.31 0.000 .8706997 .9231807\n", + "------------------------------------------------------------------------------\n", + "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", + "Compare to tabular data:\n", + "\n", + "-------------------------------------------------\n", "Distancin |\n", "g: Lowest |\n", "(1) to |\n", "Highest |\n", - "(5) | N(drivea~p) mean(drivea~p) sem(drivea~p)\n", - "----------+-----------------------------------------------\n", - " 1 | 415 81.620573 .294145\n", - " 2 | 418 82.205365 .2133359\n", - " 3 | 437 81.202296 .2154223\n", - " 4 | 416 80.164788 .2826221\n", - " 5 | 434 77.456791 .4113914\n", - "----------------------------------------------------------\n" + "(5) | N(youth) mean(youth) sem(youth)\n", + "----------+--------------------------------------\n", + " 1 | 519 23.013114 .1388531\n", + " 2 | 526 22.562249 .1310086\n", + " 3 | 513 22.287634 .1227711\n", + " 4 | 540 22.396111 .1464031\n", + " 5 | 535 21.385909 .1390452\n", + "-------------------------------------------------\n", + "----- PERCENT DIFFERENCE MODEL -----\n", + "note: levels5 omitted because of collinearity\n", + "note: youth has noninteger values\n", + "\n", + "Iteration 0: log likelihood = -7047.4698 \n", + "Iteration 1: log likelihood = -7047.2764 \n", + "Iteration 2: log likelihood = -7047.2764 \n", + "\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", + " Scale parameter = 1\n", + "Deviance = 1084.966828 (1/df) Deviance = .4142676\n", + "Pearson = 1091.163655 (1/df) Pearson = .4166337\n", + "\n", + "Variance function: V(u) = u [Poisson]\n", + "Link function : g(u) = ln(u) [Log]\n", + "\n", + " AIC = 5.363674\n", + "Log likelihood = -7047.276406 BIC = -19541.96\n", + "\n", + "------------------------------------------------------------------------------\n", + " | OIM\n", + " youth | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", + "-------------+----------------------------------------------------------------\n", + " levels1 | 1.082735 .0095077 9.05 0.000 1.06426 1.101531\n", + " levels2 | 1.058516 .0091489 6.58 0.000 1.040735 1.0766\n", + " levels3 | 1.045129 .0090803 5.08 0.000 1.027483 1.063079\n", + " levels4 | 1.049735 .0089242 5.71 0.000 1.032389 1.067373\n", + " levels5 | 1 (omitted)\n", + " homeorder | .9355478 .0075003 -8.31 0.000 .9209623 .9503643\n", + " |\n", + " rucc |\n", + " 2 | .9687538 .009452 -3.25 0.001 .9504042 .9874577\n", + " 3 | .9474401 .0095882 -5.34 0.000 .9288327 .9664203\n", + " 4 | .9468317 .0109397 -4.73 0.000 .9256313 .9685176\n", + " 5 | .9792884 .015412 -1.33 0.184 .9495426 1.009966\n", + " 6 | .9535839 .0085333 -5.31 0.000 .9370048 .9704565\n", + " 7 | .9514734 .009464 -5.00 0.000 .933104 .9702044\n", + " 8 | .9060676 .0129889 -6.88 0.000 .8809641 .9318865\n", + " 9 | .8965563 .0133864 -7.31 0.000 .8706997 .9231807\n", + " |\n", + " _cons | 23.59149 .2675883 278.67 0.000 23.07281 24.12183\n", + "------------------------------------------------------------------------------\n", + "Note: _cons estimates baseline incidence rate.\n", + "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" ] }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -2751,35 +2927,31 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t77\n", - "\t78\n", - "\t79\n", - "\t80\n", - "\t81\n", - "\t82\n", - "\t% of Drivers\n", - "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t23.5\n", + "\t24\n", + "\t24.5\n", + "\t25\n", + "\t25.5\n", + "\tavg\n", + "\t\n", "\t\n", "\t1\n", - "\t\n", - "\t2\n", - "\t\n", - "\t3\n", - "\t\n", - "\t4\n", - "\t\n", - "\t5\n", - "\tDistancing: Lowest (1) to Highest (5)\n", - "\t\n", - "\t\n", - "\tCommuting Alone by Vehicle\n", + "\t\n", + "\t2\n", + "\t\n", + "\t3\n", + "\t\n", + "\t4\n", + "\t\n", + "\t5\n", + "\tSocial Distancing: Lowest (1) to Highest (5)\n", + "\tyouth\n", "\n" ], "text/html": [ @@ -2791,35 +2963,31 @@ "\t<rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/>\n", "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", - "\t<line x1="55.44" y1="1370.56" x2="1342.53" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="1136.07" x2="1342.53" y2="1136.07" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="901.59" x2="1342.53" y2="901.59" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="667.11" x2="1342.53" y2="667.11" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="432.63" x2="1342.53" y2="432.63" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="198.15" x2="1342.53" y2="198.15" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 287.15 L381.95 150.03 L698.96 385.21 L1015.97 628.50 L1333.03 1263.49" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">77</text>\n", - "\t<text x="1364.51" y="1157.26" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">78</text>\n", - "\t<text x="1364.51" y="922.78" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">79</text>\n", - "\t<text x="1364.51" y="688.30" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">80</text>\n", - "\t<text x="1364.51" y="453.81" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">81</text>\n", - "\t<text x="1364.51" y="219.33" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">82</text>\n", - "\t<text x="1499.40" y="760.27" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,760.27)" text-anchor="middle">% of Drivers</text>\n", - "\t<line x1="55.44" y1="1380.06" x2="1342.53" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<line x1="55.44" y1="1370.56" x2="1289.81" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1085.34" x2="1289.81" y2="1085.34" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="800.17" x2="1289.81" y2="800.17" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="514.95" x2="1289.81" y2="514.95" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="229.73" x2="1289.81" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<path d=" M64.90 229.73 L368.74 514.95 L672.58 686.07 L976.42 629.05 L1280.26 1313.53" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1311.79" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">23.5</text>\n", + "\t<text x="1311.79" y="1106.52" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">24</text>\n", + "\t<text x="1311.79" y="821.35" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">24.5</text>\n", + "\t<text x="1311.79" y="536.13" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">25</text>\n", + "\t<text x="1311.79" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">25.5</text>\n", + "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", + "\t<line x1="55.44" y1="1380.06" x2="1289.81" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", "\t<text x="64.95" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">1</text>\n", - "\t<line x1="381.95" y1="1358.08" x2="381.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", - "\t<text x="381.95" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">2</text>\n", - "\t<line x1="699.01" y1="1358.08" x2="699.01" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", - "\t<text x="699.01" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">3</text>\n", - "\t<line x1="1016.02" y1="1358.08" x2="1016.02" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", - "\t<text x="1016.02" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">4</text>\n", - "\t<line x1="1333.03" y1="1358.08" x2="1333.03" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", - "\t<text x="1333.03" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">5</text>\n", - "\t<text x="699.01" y="1499.40" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">Distancing: Lowest (1) to Highest (5)</text>\n", - "\t<rect x="453.78" y="55.49" width="888.75" height="84.99" style="fill:#7B92A8"/>\n", - "\t<rect x="455.36" y="57.07" width="885.58" height="81.82" style="fill:none;stroke:#7B92A8;stroke-width:3.17"/>\n", - "\t<text x="1333.03" y="121.12" style="font-family:'Helvetica';font-size:65.99px;fill:#FFFFff" text-anchor="end">Commuting Alone by Vehicle</text>\n", + "\t<line x1="368.79" y1="1358.08" x2="368.79" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="368.79" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">2</text>\n", + "\t<line x1="672.63" y1="1358.08" x2="672.63" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="672.63" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">3</text>\n", + "\t<line x1="976.47" y1="1358.08" x2="976.47" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="976.47" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">4</text>\n", + "\t<line x1="1280.31" y1="1358.08" x2="1280.31" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="1280.31" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">5</text>\n", + "\t<text x="672.63" y="1499.40" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">Social Distancing: Lowest (1) to Highest (5)</text>\n", + "\t<text x="90.29" y="156.84" style="font-family:'Helvetica';font-size:95.04px;fill:#000000">youth</text>\n", "</svg>\n", "\">\n" ], @@ -2841,24 +3009,40 @@ } ], "source": [ - "frame change default\n", - "foreach var of varlist drivealone_p {\n", - " table iso5, c(count `var' mean `var' sem `var')\n", - " frame put `var' iso5, into(`var')\n", - " frame change `var'\n", - " collapse (mean) `var', by(iso5)\n", - " la var `var' \"% of Drivers\"\n", - " line `var' iso5, note(\"Commuting Alone by Vehicle\") \n", - " frame change default\n", - " frame drop `var'\n", - "}" + "modelpoisson youth" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "---\n" + "### Interpretation\n", + "

\n", + "\n", + "Relative effect measures
\n", + "` ^\n", + " levels1 | 25.54333 .3027171 273.42 0.000 24.95686 26.14359\n", + " levels2 | 24.97196 .2904891 276.61 0.000 24.40905 25.54785\n", + " levels3 | 24.65616 .2854101 276.88 0.000 24.10306 25.22195\n", + " levels4 | 24.76482 .2843834 279.48 0.000 24.21366 25.32852\n", + " levels5 | 23.59149 .2675883 278.67 0.000 23.07281 24.12183\n", + "`\n", + "

\n", + "Percent difference
\n", + "` levels1 | 1.082735 .0095077 9.05 0.000 1.06426 1.101531`\n", + "\n", + "Counties with the least restriction of movement had 8.2% more children (95% CI: 6.4%, 10%) than areas that most greatly had their movement reduced." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n", + "\n", + "## Elderly\n", + "\n", + "### Interpretation" ] }, { @@ -2870,27 +3054,112 @@ "name": "stdout", "output_type": "stream", "text": [ + "----- RURALITY-ADJUSTED POISSON MODEL -----\n", + "note: elderly has noninteger values\n", "\n", + "Iteration 0: log likelihood = -7218.0267 \n", + "Iteration 1: log likelihood = -7216.9383 \n", + "Iteration 2: log likelihood = -7216.9383 \n", "\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", + " Scale parameter = 1\n", + "Deviance = 2003.292916 (1/df) Deviance = .7649076\n", + "Pearson = 2072.529905 (1/df) Pearson = .791344\n", "\n", - "----------------------------------------------\n", + "Variance function: V(u) = u [Poisson]\n", + "Link function : g(u) = ln(u) [Log]\n", + "\n", + " AIC = 5.492547\n", + "Log likelihood = -7216.938253 BIC = -18623.63\n", + "\n", + "------------------------------------------------------------------------------\n", + " | OIM\n", + " elderly | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", + "-------------+----------------------------------------------------------------\n", + " levels1 | 14.16821 .2681426 140.07 0.000 13.65229 14.70363\n", + " levels2 | 14.60455 .2709946 144.50 0.000 14.08295 15.14547\n", + " levels3 | 14.94807 .2750933 146.96 0.000 14.4185 15.49708\n", + " levels4 | 15.00888 .2744157 148.15 0.000 14.48055 15.55647\n", + " levels5 | 15.22254 .2732011 151.71 0.000 14.69638 15.76753\n", + " homeorder | 1.035171 .0129279 2.77 0.006 1.010141 1.060822\n", + " |\n", + " rucc |\n", + " 2 | 1.124375 .0177546 7.42 0.000 1.09011 1.159718\n", + " 3 | 1.149063 .0186141 8.58 0.000 1.113153 1.186131\n", + " 4 | 1.182938 .0214397 9.27 0.000 1.141655 1.225715\n", + " 5 | 1.096782 .0282309 3.59 0.000 1.042823 1.153533\n", + " 6 | 1.268995 .018046 16.75 0.000 1.234114 1.304862\n", + " 7 | 1.284184 .0200334 16.03 0.000 1.245514 1.324056\n", + " 8 | 1.391649 .028851 15.94 0.000 1.336236 1.449361\n", + " 9 | 1.430253 .0305757 16.74 0.000 1.371563 1.491453\n", + "------------------------------------------------------------------------------\n", + "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", + "Compare to tabular data:\n", + "\n", + "-------------------------------------------------------\n", "Distancin |\n", "g: Lowest |\n", "(1) to |\n", "Highest |\n", - "(5) | N(rucc) mean(rucc) sem(rucc)\n", - "----------+-----------------------------------\n", - " 1 | 415 5.4409637 .1084386\n", - " 2 | 418 4.7703347 .1129045\n", - " 3 | 437 4.3913045 .1183801\n", - " 4 | 416 3.9711537 .1178481\n", - " 5 | 434 3.5714285 .1197383\n", - "----------------------------------------------\n" + "(5) | N(elderly) mean(elderly) sem(elderly)\n", + "----------+--------------------------------------------\n", + " 1 | 519 17.897361 .1736063\n", + " 2 | 526 18.07345 .1721838\n", + " 3 | 513 18.483562 .162852\n", + " 4 | 540 18.246843 .1874652\n", + " 5 | 535 18.115094 .2083171\n", + "-------------------------------------------------------\n", + "----- PERCENT DIFFERENCE MODEL -----\n", + "note: levels5 omitted because of collinearity\n", + "note: elderly has noninteger values\n", + "\n", + "Iteration 0: log likelihood = -7218.0267 \n", + "Iteration 1: log likelihood = -7216.9383 \n", + "Iteration 2: log likelihood = -7216.9383 \n", + "\n", + "Generalized linear models Number of obs = 2,633\n", + "Optimization : ML Residual df = 2,619\n", + " Scale parameter = 1\n", + "Deviance = 2003.292916 (1/df) Deviance = .7649076\n", + "Pearson = 2072.529905 (1/df) Pearson = .791344\n", + "\n", + "Variance function: V(u) = u [Poisson]\n", + "Link function : g(u) = ln(u) [Log]\n", + "\n", + " AIC = 5.492547\n", + "Log likelihood = -7216.938253 BIC = -18623.63\n", + "\n", + "------------------------------------------------------------------------------\n", + " | OIM\n", + " elderly | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", + "-------------+----------------------------------------------------------------\n", + " levels1 | .930739 .0124421 -5.37 0.000 .9066697 .9554473\n", + " levels2 | .9594032 .0125665 -3.16 0.002 .9350867 .9843519\n", + " levels3 | .9819697 .0128292 -1.39 0.164 .9571442 1.007439\n", + " levels4 | .9859642 .0126631 -1.10 0.271 .9614548 1.011098\n", + " levels5 | 1 (omitted)\n", + " homeorder | 1.035171 .0129279 2.77 0.006 1.010141 1.060822\n", + " |\n", + " rucc |\n", + " 2 | 1.124375 .0177546 7.42 0.000 1.09011 1.159718\n", + " 3 | 1.149063 .0186141 8.58 0.000 1.113153 1.186131\n", + " 4 | 1.182938 .0214397 9.27 0.000 1.141655 1.225715\n", + " 5 | 1.096782 .0282309 3.59 0.000 1.042823 1.153533\n", + " 6 | 1.268995 .018046 16.75 0.000 1.234114 1.304862\n", + " 7 | 1.284184 .0200334 16.03 0.000 1.245514 1.324056\n", + " 8 | 1.391649 .028851 15.94 0.000 1.336236 1.449361\n", + " 9 | 1.430253 .0305757 16.74 0.000 1.371563 1.491453\n", + " |\n", + " _cons | 15.22254 .2732011 151.71 0.000 14.69638 15.76753\n", + "------------------------------------------------------------------------------\n", + "Note: _cons estimates baseline incidence rate.\n", + "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" ] }, { "data": { - "application/pdf": "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", + "application/pdf": "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", "image/svg+xml": [ "\n", "\n", @@ -2900,31 +3169,33 @@ "\t\n", "\t\n", "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t\n", - "\t3\n", - "\t4\n", - "\t5\n", - "\t6\n", - "\tMeidan RUCC\n", - "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t14.2\n", + "\t14.4\n", + "\t14.6\n", + "\t14.8\n", + "\t15\n", + "\t15.2\n", + "\tavg\n", + "\t\n", "\t\n", "\t1\n", - "\t\n", - "\t2\n", - "\t\n", - "\t3\n", - "\t\n", - "\t4\n", - "\t\n", - "\t5\n", - "\tDistancing: Lowest (1) to Highest (5)\n", - "\t\n", - "\t\n", - "\tUrban-Rural\n", + "\t\n", + "\t2\n", + "\t\n", + "\t3\n", + "\t\n", + "\t4\n", + "\t\n", + "\t5\n", + "\tSocial Distancing: Lowest (1) to Highest (5)\n", + "\telderly\n", "\n" ], "text/html": [ @@ -2936,17 +3207,616 @@ "\t<rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/>\n", "\t<rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/>\n", "\t<rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/>\n", - "\t<line x1="55.44" y1="1370.56" x2="1376.84" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="963.72" x2="1376.84" y2="963.72" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="556.88" x2="1376.84" y2="556.88" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<line x1="55.44" y1="149.98" x2="1376.84" y2="149.98" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 150.03 L390.52 150.03 L716.14 963.72 L1041.71 1370.61 L1367.33 1370.61" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", - "\t<text x="1398.81" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">3</text>\n", - "\t<text x="1398.81" y="984.90" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">4</text>\n", - "\t<text x="1398.81" y="578.06" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">5</text>\n", - "\t<text x="1398.81" y="171.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">6</text>\n", - "\t<text x="1499.40" y="760.27" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,760.27)" text-anchor="middle">Meidan RUCC</text>\n", - "\t<line x1="55.44" y1="1380.06" x2="1376.84" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<line x1="55.44" y1="1370.56" x2="1289.81" y2="1370.56" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="1142.41" x2="1289.81" y2="1142.41" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="914.22" x2="1289.81" y2="914.22" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="686.07" x2="1289.81" y2="686.07" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="457.88" x2="1289.81" y2="457.88" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<line x1="55.44" y1="229.73" x2="1289.81" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", + "\t<path d=" M64.90 1370.61 L368.74 914.26 L672.58 572.02 L976.42 457.92 L1280.26 229.73" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<text x="1311.79" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">14.2</text>\n", + "\t<text x="1311.79" y="1163.59" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">14.4</text>\n", + "\t<text x="1311.79" y="935.40" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">14.6</text>\n", + "\t<text x="1311.79" y="707.25" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">14.8</text>\n", + "\t<text x="1311.79" y="479.06" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">15</text>\n", + "\t<text x="1311.79" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">15.2</text>\n", + "\t<text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text>\n", + "\t<line x1="55.44" y1="1380.06" x2="1289.81" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="64.95" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">1</text>\n", + "\t<line x1="368.79" y1="1358.08" x2="368.79" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="368.79" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">2</text>\n", + "\t<line x1="672.63" y1="1358.08" x2="672.63" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="672.63" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">3</text>\n", + "\t<line x1="976.47" y1="1358.08" x2="976.47" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="976.47" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">4</text>\n", + "\t<line x1="1280.31" y1="1358.08" x2="1280.31" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/>\n", + "\t<text x="1280.31" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">5</text>\n", + "\t<text x="672.63" y="1499.40" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">Social Distancing: Lowest (1) to Highest (5)</text>\n", + "\t<text x="90.29" y="156.84" style="font-family:'Helvetica';font-size:95.04px;fill:#000000">elderly</text>\n", + "</svg>\n", + "\">\n" + ], + "text/plain": [ + "This front-end cannot display the desired image type." + ] + }, + "metadata": { + "image/svg+xml": { + "height": 600, + "width": 600 + }, + "text/html": { + "height": 600, + "width": 600 + } + }, + "output_type": "display_data" + } + ], + "source": [ + "modelpoisson elderly" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Interpretation\n", + "

\n", + "\n", + "Relative effect measures
\n", + "` ^\n", + " levels1 | 14.16821 .2681426 140.07 0.000 13.65229 14.70363\n", + " levels2 | 14.60455 .2709946 144.50 0.000 14.08295 15.14547\n", + " levels3 | 14.94807 .2750933 146.96 0.000 14.4185 15.49708\n", + " levels4 | 15.00888 .2744157 148.15 0.000 14.48055 15.55647\n", + " levels5 | 15.22254 .2732011 151.71 0.000 14.69638 15.76753\n", + "`\n", + "

\n", + "Percent difference
\n", + "` levels1 | .930739 .0124421 -5.37 0.000 .9066697 .9554473`\n", + "\n", + "Counties that did the best at restricting movement had 7.4% (95% CI: 4.7%, 10%) more elderly people, compared to the lowest tier of movement restriction." + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "7.4\n", + "LL: 10.3\n", + "UL: 4.7\n" + ] + } + ], + "source": [ + "invert .930739 .9066697 .9554473" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n", + "\n", + "## Segregation\n", + "\n", + "### Interpretation" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "----- RURALITY-ADJUSTED NEGBIN MODEL -----\n", + "note: you are responsible for interpretation of non-count dep. variable\n", + "\n", + "Fitting Poisson model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -137517.44 \n", + "Iteration 1: log pseudolikelihood = -14752.896 \n", + "Iteration 2: log pseudolikelihood = -13227.944 \n", + "Iteration 3: log pseudolikelihood = -13227.003 \n", + "Iteration 4: log pseudolikelihood = -13227.003 \n", + "\n", + "Fitting full model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -11513.071 \n", + "Iteration 1: log pseudolikelihood = -10469.271 \n", + "Iteration 2: log pseudolikelihood = -10203.36 \n", + "Iteration 3: log pseudolikelihood = -10195.34 \n", + "Iteration 4: log pseudolikelihood = -10195.337 \n", + "Iteration 5: log pseudolikelihood = -10195.337 \n", + "\n", + "Negative binomial regression Number of obs = 2,581\n", + "Dispersion = mean Wald chi2(14) = 216209.08\n", + "Log pseudolikelihood = -10195.337 Prob > chi2 = 0.0000\n", + "\n", + "-------------------------------------------------------------------------------\n", + " | Robust\n", + "segregatio~nw | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", + "--------------+----------------------------------------------------------------\n", + " levels1 | 29.07819 1.106746 88.54 0.000 26.98794 31.33033\n", + " levels2 | 29.62044 1.078855 93.03 0.000 27.57963 31.81226\n", + " levels3 | 30.05066 1.117545 91.50 0.000 27.93823 32.32281\n", + " levels4 | 30.79493 1.091354 96.71 0.000 28.72851 33.00999\n", + " levels5 | 31.866 1.11741 98.72 0.000 29.74948 34.1331\n", + " homeorder | 1.075855 .0295754 2.66 0.008 1.019423 1.135412\n", + " |\n", + " rucc |\n", + " 2 | 1.023216 .0261958 0.90 0.370 .9731398 1.075869\n", + " 3 | 1.018929 .028167 0.68 0.498 .9651917 1.075658\n", + " 4 | 1.020022 .0288734 0.70 0.484 .9649719 1.078212\n", + " 5 | 1.028869 .0393519 0.74 0.457 .9545604 1.108962\n", + " 6 | .9387745 .0235422 -2.52 0.012 .8937482 .9860692\n", + " 7 | .9473715 .0303327 -1.69 0.091 .8897474 1.008728\n", + " 8 | .7978725 .0400495 -4.50 0.000 .7231146 .8803591\n", + " 9 | .8521205 .0532593 -2.56 0.010 .7538747 .9631698\n", + "--------------+----------------------------------------------------------------\n", + " /lnalpha | -1.962348 .0443816 -2.049334 -1.875361\n", + "--------------+----------------------------------------------------------------\n", + " alpha | .1405281 .0062369 .1288207 .1532996\n", + "-------------------------------------------------------------------------------\n", + "Note: Estimates are transformed only in the first equation.\n", + "Compare to tabular data:\n", + "\n", + "----------------------------------------------------------\n", + "Distancin |\n", + "g: Lowest |\n", + "(1) to |\n", + "Highest |\n", + "(5) | N(segreg~w) mean(segreg~w) sem(segreg~w)\n", + "----------+-----------------------------------------------\n", + " 1 | 506 29.687411 .5589038\n", + " 2 | 513 30.582416 .5315276\n", + " 3 | 498 31.042838 .5548652\n", + " 4 | 535 32.147839 .5309895\n", + " 5 | 529 33.599273 .5652224\n", + "----------------------------------------------------------\n", + "----- PERCENT DIFFERENCE MODEL -----\n", + "note: you are responsible for interpretation of non-count dep. variable\n", + "note: levels5 omitted because of collinearity\n", + "\n", + "Fitting Poisson model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -13227.003 \n", + "Iteration 1: log pseudolikelihood = -13227.003 \n", + "\n", + "Fitting constant-only model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -11521.003 \n", + "Iteration 1: log pseudolikelihood = -10513.242 \n", + "Iteration 2: log pseudolikelihood = -10247.006 \n", + "Iteration 3: log pseudolikelihood = -10241.9 \n", + "Iteration 4: log pseudolikelihood = -10241.899 \n", + "\n", + "Fitting full model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -10196.302 \n", + "Iteration 1: log pseudolikelihood = -10195.338 \n", + "Iteration 2: log pseudolikelihood = -10195.337 \n", + "\n", + "Negative binomial regression Number of obs = 2,581\n", + " Wald chi2(13) = 87.17\n", + "Dispersion = mean Prob > chi2 = 0.0000\n", + "Log pseudolikelihood = -10195.337 Pseudo R2 = 0.0045\n", + "\n", + "-------------------------------------------------------------------------------\n", + " | Robust\n", + "segregatio~nw | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", + "--------------+----------------------------------------------------------------\n", + " levels1 | .9125145 .0241028 -3.47 0.001 .8664759 .9609994\n", + " levels2 | .9295311 .0233124 -2.91 0.004 .8849443 .9763643\n", + " levels3 | .9430319 .0234616 -2.36 0.018 .8981512 .9901554\n", + " levels4 | .9663884 .0228272 -1.45 0.148 .9226678 1.012181\n", + " levels5 | 1 (omitted)\n", + " homeorder | 1.075855 .0295754 2.66 0.008 1.019423 1.135412\n", + " |\n", + " rucc |\n", + " 2 | 1.023216 .0261958 0.90 0.370 .9731399 1.075869\n", + " 3 | 1.018929 .028167 0.68 0.498 .9651917 1.075658\n", + " 4 | 1.020022 .0288734 0.70 0.484 .9649719 1.078212\n", + " 5 | 1.028869 .0393519 0.74 0.457 .9545604 1.108962\n", + " 6 | .9387745 .0235422 -2.52 0.012 .8937482 .9860692\n", + " 7 | .9473715 .0303327 -1.69 0.091 .8897475 1.008728\n", + " 8 | .7978726 .0400495 -4.50 0.000 .7231146 .8803592\n", + " 9 | .8521205 .0532593 -2.56 0.010 .7538748 .9631698\n", + " |\n", + " _cons | 31.866 1.11741 98.72 0.000 29.74948 34.1331\n", + "--------------+----------------------------------------------------------------\n", + " /lnalpha | -1.962348 .0443816 -2.049334 -1.875362\n", + "--------------+----------------------------------------------------------------\n", + " alpha | .1405281 .0062369 .1288207 .1532995\n", + "-------------------------------------------------------------------------------\n", + "Note: Estimates are transformed only in the first equation.\n", + "Note: _cons estimates baseline incidence rate.\n" + ] + }, + { + "data": { + "application/pdf": "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", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + "\tStata Graph - Graph\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t29\n", + "\t30\n", + "\t31\n", + "\t32\n", + "\tavg\n", + "\t\n", + "\t\n", + "\t1\n", + "\t\n", + "\t2\n", + "\t\n", + "\t3\n", + "\t\n", + "\t4\n", + "\t\n", + "\t5\n", + "\tSocial Distancing: Lowest (1) to Highest (5)\n", + "\tsegregation_wnw\n", + "\n" + ], + "text/html": [ + " \n" + ], + "text/plain": [ + "This front-end cannot display the desired image type." + ] + }, + "metadata": { + "image/svg+xml": { + "height": 600, + "width": 600 + }, + "text/html": { + "height": 600, + "width": 600 + } + }, + "output_type": "display_data" + } + ], + "source": [ + "modelrun segregation_wnw" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Interpretation\n", + "

\n", + "\n", + "Relative effect measures
\n", + "` ^\n", + " levels1 | 29.07819 1.106746 88.54 0.000 26.98794 31.33033\n", + " levels2 | 29.62044 1.078855 93.03 0.000 27.57963 31.81226\n", + " levels3 | 30.05066 1.117545 91.50 0.000 27.93823 32.32281\n", + " levels4 | 30.79493 1.091354 96.71 0.000 28.72851 33.00999\n", + " levels5 | 31.866 1.11741 98.72 0.000 29.74948 34.1331\n", + "`\n", + "

\n", + "Percent difference
\n", + "` levels1 | .9125145 .0241028 -3.47 0.001 .8664759 .9609994`\n", + "\n", + "The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders. " + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "9.6\n", + "LL: 15.4\n", + "UL: 4.1\n" + ] + } + ], + "source": [ + "invert .9125145 .8664759 .9609994" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n", + "\n", + "# Exploratory analyses" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "\n", + "\n", + "----------------------------------------------------------\n", + "Distancin |\n", + "g: Lowest |\n", + "(1) to |\n", + "Highest |\n", + "(5) | N(drivea~p) mean(drivea~p) sem(drivea~p)\n", + "----------+-----------------------------------------------\n", + " 1 | 519 82.089322 .2316678\n", + " 2 | 526 81.973571 .1901521\n", + " 3 | 513 81.401447 .1971413\n", + " 4 | 540 80.95994 .1918464\n", + " 5 | 535 77.732723 .3661999\n", + "----------------------------------------------------------\n" + ] + }, + { + "data": { + "application/pdf": "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", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + "\tStata Graph - Graph\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t78\n", + "\t79\n", + "\t80\n", + "\t81\n", + "\t82\n", + "\t% of Drivers\n", + "\t\n", + "\t\n", + "\t1\n", + "\t\n", + "\t2\n", + "\t\n", + "\t3\n", + "\t\n", + "\t4\n", + "\t\n", + "\t5\n", + "\tDistancing: Lowest (1) to Highest (5)\n", + "\t\n", + "\t\n", + "\tCommuting Alone by Vehicle\n", + "\n" + ], + "text/html": [ + " \n" + ], + "text/plain": [ + "This front-end cannot display the desired image type." + ] + }, + "metadata": { + "image/svg+xml": { + "height": 600, + "width": 600 + }, + "text/html": { + "height": 600, + "width": 600 + } + }, + "output_type": "display_data" + } + ], + "source": [ + "frame change default\n", + "foreach var of varlist drivealone_p {\n", + " table iso5, c(count `var' mean `var' sem `var')\n", + " frame put `var' iso5, into(`var')\n", + " frame change `var'\n", + " collapse (mean) `var', by(iso5)\n", + " la var `var' \"% of Drivers\"\n", + " line `var' iso5, note(\"Commuting Alone by Vehicle\") \n", + " frame change default\n", + " frame drop `var'\n", + "}" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "\n", + "\n", + "----------------------------------------------\n", + "Distancin |\n", + "g: Lowest |\n", + "(1) to |\n", + "Highest |\n", + "(5) | N(rucc) mean(rucc) sem(rucc)\n", + "----------+-----------------------------------\n", + " 1 | 519 5.2524085 .0987557\n", + " 2 | 526 4.63308 .1022326\n", + " 3 | 513 4.594542 .1055764\n", + " 4 | 540 4.062963 .104029\n", + " 5 | 535 3.6093459 .1101034\n", + "----------------------------------------------\n" + ] + }, + { + "data": { + "application/pdf": "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", + "image/svg+xml": [ + "\n", + "\n", + "\n", + "\n", + "\tStata Graph - Graph\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t\n", + "\t3\n", + "\t4\n", + "\t5\n", + "\t6\n", + "\tMeidan RUCC\n", + "\t\n", + "\t\n", + "\t1\n", + "\t\n", + "\t2\n", + "\t\n", + "\t3\n", + "\t\n", + "\t4\n", + "\t\n", + "\t5\n", + "\tDistancing: Lowest (1) to Highest (5)\n", + "\t\n", + "\t\n", + "\tUrban-Rural\n", + "\n" + ], + "text/html": [ + " @@ -13349,20 +13851,27 @@

Primary Care Providers
-
-

Mental Health Providers

For people sequestered at home, mental health providers may play an emerging role. These data are from CMS National Provider Identification (NPI) via RWJF. It is expressed in rates per 100,000 population.

+

Interpretation



+

Relative effect measures
+^ + levels1 | 49.92198 2.325957 83.93 0.000 45.56514 54.69541 + levels2 | 54.89074 2.441119 90.06 0.000 50.30882 59.88996 + levels3 | 57.89873 2.709556 86.73 0.000 52.82437 63.46053 + levels4 | 61.49219 2.753973 91.97 0.000 56.32462 67.13387 + levels5 | 73.62823 3.207945 98.67 0.000 67.60175 80.19195 +

+Percent difference
+levels1 | .6780277 .0239497 -11.00 0.000 .6326751 .7266314

-
In [11]:
+
In [5]:
-
// Basic descriptive on commuting alone
-frame change default
-summ mhproviders_rate, d
+
invert .6780277 .6326751 .7266314
@@ -13379,23 +13888,9 @@

Mental Health Providers -
-
-
-       Mental Health Providers per 100,000 population
--------------------------------------------------------------
-      Percentiles      Smallest
- 1%      7.51089              0
- 5%     13.71109          4.153
-10%     22.33016         4.2155       Obs               2,450
-25%     48.25257        4.41326       Sum of Wgt.       2,450
-
-50%     108.2791                      Mean           150.8881
-                        Largest       Std. Dev.      149.7092
-75%     203.5753       1202.062
-90%     321.5745       1355.829       Variance       22412.86
-95%     430.9503       1419.854       Skewness       2.616105
-99%     745.7847        1420.66       Kurtosis       14.55825
+
47.5
+LL: 58.1
+UL: 37.6

@@ -13404,134 +13899,30 @@

Mental Health Providers

-
-
-
In [12]:
-
-
-
// Comparing mental health provider rate to social distancing
-foreach var of varlist mhproviders_rate {
-    frame change default
-        table iso5, c(count `var' mean `var' sem `var')
-            frame put `var' iso5, into(`var')
-                frame change `var'
-                    collapse (mean) `var', by(iso5)
-                        la var mhproviders_rate "Mental Health Proviers per 100k population"
-                            line `var' iso5, note("Mental Health Providers")           
-}
-
- -
-
-
- -
-
- - -
- -
- - -
-
-
-----------------------------------------------------------
-Social    |
-Distancin |
-g: Lowest |
-(1) to    |
-Highest   |
-(5)       |    N(mhprov~e)  mean(mhprov~e)   sem(mhprov~e)
-----------+-----------------------------------------------
-        1 |            468        129.8275        7.605745
-        2 |            486       130.68528        6.615416
-        3 |            502       143.27894        5.757351
-        4 |            497       153.20643         6.05538
-        5 |            497        195.8431        7.293431
-----------------------------------------------------------
-
-
-
- -
- -
- - - -
- - -
- -
+
+
+
+

The counties showing the smallest declines in mobility had 50 primary care providers per 100,000, whereas the most social distancing counties had 74 per 100,000 after adjusting for rurality and stay-at-home orders, a 47% (95% CI: 38%, 58%) difference.

+
-
-
-

Health Insurance

Getting hospitalized for Covid-19 is sure to be expensive. After hearing that even some front line nurses in pulmonary infectious disease units in North Carolina do not have health insurance, we wanted to see if there was any association between health insurance status and social distancing.

+

Percent uninsured

Percent of without health insurance below Medicare elgibility (age 65).

-
In [13]:
+
In [6]:
// Comparing percent uninsured to social distancing
-frame change default
-foreach var of varlist uninsured_p {
-    table iso5, c(count `var' mean `var' sem `var')
-        frame put `var' iso5, into(`var')
-            frame change `var'
-                collapse (mean) `var', by(iso5)
-                    la var `var' "% Uninsured"
-                        line `var' iso5 , note("Uninsured Adults Under 65") 
-}
+modelrun uninsured_p
@@ -13548,23 +13939,128 @@

Health Insurance -
+
----- RURALITY-ADJUSTED NEGBIN MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -25368.195  
+Iteration 1:   log pseudolikelihood = -7847.5733  
+Iteration 2:   log pseudolikelihood = -7718.5574  
+Iteration 3:   log pseudolikelihood = -7718.2864  
+Iteration 4:   log pseudolikelihood = -7718.2864  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -8982.7909  
+Iteration 1:   log pseudolikelihood = -7464.8863  
+Iteration 2:   log pseudolikelihood = -7462.9819  
+Iteration 3:   log pseudolikelihood =  -7462.981  
+Iteration 4:   log pseudolikelihood =  -7462.981  
 
+Negative binomial regression                    Number of obs     =      2,633
+Dispersion           = mean                     Wald chi2(14)     =   94084.62
+Log pseudolikelihood =  -7462.981               Prob > chi2       =     0.0000
+
+------------------------------------------------------------------------------
+             |               Robust
+ uninsured_p |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   10.74092   .3917303    65.09   0.000     9.999938     11.5368
+     levels2 |   9.464582   .3546025    59.99   0.000     8.794479    10.18574
+     levels3 |   8.480079   .3192442    56.78   0.000     7.876899    9.129449
+     levels4 |   8.000203   .3006194    55.34   0.000     7.432174    8.611645
+     levels5 |   7.044127   .2647529    51.94   0.000     6.543873    7.582624
+   homeorder |   1.156171   .0313348     5.35   0.000     1.096358    1.219246
+             |
+        rucc |
+          2  |   1.036269   .0311154     1.19   0.235     .9770437    1.099084
+          3  |    1.02913   .0307182     0.96   0.336     .9706509    1.091133
+          4  |   1.050335   .0360446     1.43   0.152     .9820124    1.123411
+          5  |    1.05925   .0478817     1.27   0.203     .9694404    1.157379
+          6  |   1.170248   .0327939     5.61   0.000     1.107706    1.236321
+          7  |   1.123108   .0352113     3.70   0.000     1.056173    1.194286
+          8  |   1.237528   .0477472     5.52   0.000     1.147396     1.33474
+          9  |     1.1618   .0488127     3.57   0.000     1.069962    1.261521
+-------------+----------------------------------------------------------------
+    /lnalpha |  -2.682096   .0589782                     -2.797691   -2.566501
+-------------+----------------------------------------------------------------
+       alpha |   .0684196   .0040353                      .0609506    .0768038
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Compare to tabular data:
 
 ----------------------------------------------------------
-Social    |
 Distancin |
 g: Lowest |
 (1) to    |
 Highest   |
 (5)       |    N(uninsu~p)  mean(uninsu~p)   sem(uninsu~p)
 ----------+-----------------------------------------------
-        1 |            499       13.308561        .1971504
-        2 |            497       12.334842        .2039521
-        3 |            506       10.995156        .2120537
-        4 |            503       9.4989416        .1942909
-        5 |            502       7.7708062        .1648914
+        1 |            519       13.379365        .2038213
+        2 |            526       11.729307        .2069097
+        3 |            513       10.520381        .2010726
+        4 |            540       9.8374245        .1876119
+        5 |            535       8.6297361        .1775762
 ----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+note: levels5 omitted because of collinearity
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -7718.2865  
+Iteration 1:   log pseudolikelihood = -7718.2864  
+
+Fitting constant-only model:
+
+Iteration 0:   log pseudolikelihood = -9016.9274  
+Iteration 1:   log pseudolikelihood =  -7680.503  
+Iteration 2:   log pseudolikelihood = -7675.9149  
+Iteration 3:   log pseudolikelihood = -7675.8619  
+Iteration 4:   log pseudolikelihood = -7675.8619  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -7480.4687  
+Iteration 1:   log pseudolikelihood = -7463.1407  
+Iteration 2:   log pseudolikelihood =  -7462.981  
+Iteration 3:   log pseudolikelihood =  -7462.981  
+
+Negative binomial regression                    Number of obs     =      2,633
+                                                Wald chi2(13)     =     482.26
+Dispersion           = mean                     Prob > chi2       =     0.0000
+Log pseudolikelihood =  -7462.981               Pseudo R2         =     0.0277
+
+------------------------------------------------------------------------------
+             |               Robust
+ uninsured_p |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   1.524804   .0395603    16.26   0.000     1.449206    1.604347
+     levels2 |   1.343613   .0368115    10.78   0.000     1.273367    1.417735
+     levels3 |   1.203851   .0339756     6.57   0.000     1.139068    1.272318
+     levels4 |   1.135727   .0316513     4.57   0.000     1.075355    1.199488
+     levels5 |          1  (omitted)
+   homeorder |   1.156171   .0313348     5.35   0.000     1.096358    1.219246
+             |
+        rucc |
+          2  |   1.036269   .0311154     1.19   0.235     .9770437    1.099084
+          3  |    1.02913   .0307182     0.96   0.336     .9706509    1.091133
+          4  |   1.050335   .0360446     1.43   0.152     .9820124    1.123411
+          5  |    1.05925   .0478817     1.27   0.203     .9694404    1.157379
+          6  |   1.170248   .0327939     5.61   0.000     1.107706    1.236321
+          7  |   1.123108   .0352113     3.70   0.000     1.056173    1.194286
+          8  |   1.237528   .0477472     5.52   0.000     1.147396     1.33474
+          9  |     1.1618   .0488127     3.57   0.000     1.069962    1.261521
+             |
+       _cons |   7.044127   .2647529    51.94   0.000     6.543873    7.582624
+-------------+----------------------------------------------------------------
+    /lnalpha |  -2.682096   .0589782                     -2.797691   -2.566501
+-------------+----------------------------------------------------------------
+       alpha |   .0684196   .0040353                      .0609506    .0768038
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Note: _cons estimates baseline incidence rate.

@@ -13584,20 +14080,18 @@

Health InsuranceHealth Insurance @@ -13627,7 +14119,19 @@

Health Insurance
-

Alarmingly, counties with lower social distancing also had more uninsured residents!

+

Interpretation



+

Relative effect measures
+^ + levels1 | 10.74092 .3917303 65.09 0.000 9.999938 11.5368 + levels2 | 9.464582 .3546025 59.99 0.000 8.794479 10.18574 + levels3 | 8.480079 .3192442 56.78 0.000 7.876899 9.129449 + levels4 | 8.000203 .3006194 55.34 0.000 7.432174 8.611645 + levels5 | 7.044127 .2647529 51.94 0.000 6.543873 7.582624 +

+Percent difference
+levels1 | 1.524804 .0395603 16.26 0.000 1.449206 1.604347

+

Counties with lower social distancing also had a higher proportion of people without health insurance. The lowest social distancing counties had 10.7% uninsured adults, whereas the most social distancing counties had only 7.0% uninsured after adjusting for rurality and social distancing orders, a 52% (95% CI: 45%, 60%) difference.

+
@@ -13635,15 +14139,14 @@

Health Insurance
-
-

Flu Vaccination

We had a hypothesis that counties that were more involved in preventative behaviors would be more likely to self-isolate more thoroughly. To test this, we examined whether earlier flu vaccination rates impacted how much the county was likely to slow down in the current coronavirus outbreak. This is quantified as the percent of annual Medicare enrollees having an annual flu vaccination, as reported by the Robert Wood Johnson Foundation. Since the flu vaccine is free to all Medicare beneficiaries, and this is the elderly age group with the most influenza mortality, this is a convenient metric to test a priori how conscientious the population was, on average.

+

Flu Vaccination

We had a hypothesis that counties that were more involved in preventative behaviors would be more likely to self-isolate more thoroughly. To test this, we examined whether earlier flu vaccination rates impacted how much the county was likely to slow down in the current coronavirus outbreak. This is quantified as the percent of annual Medicare enrollees having an annual flu vaccination, as reported by the Robert Wood Johnson Foundation. Since the flu vaccine is free to all Medicare beneficiaries, and this is the elderly age group with the most influenza mortality, this is a convenient metric to test a priori how conscientious the population was, on average.

-
In [14]:
+
In [7]:
// Basic descriptive on background influenza vaccine
@@ -13673,16 +14176,16 @@ 
Flu VaccinationFlu VaccinationFlu Vaccination
 

-
In [15]:

+
In [8]:

 

     

 
// Comparing background flu vaccination with current social distancing
-table iso5, c(count fluvaccine mean fluvaccine sem fluvaccine)
-
-frame put fluvaccine iso5, into(flu)
-    frame change flu
-        collapse (mean) fluvaccine, by(iso5)
-            la var fluvaccine "Medicare Beneficiary Flu Vaccination %"
-                line fluvaccine iso5, note("Elderly Flu Vaccination")
+modelpoisson fluvaccine
 

 
     

@@ -13802,24 +14299,107 @@ 
Flu Vaccination
-
+
----- RURALITY-ADJUSTED POISSON MODEL -----
+
+Iteration 0:   log likelihood = -9326.8988  
+Iteration 1:   log likelihood = -9325.1183  
+Iteration 2:   log likelihood = -9325.1183  
+
+Generalized linear models                         Number of obs   =      2,630
+Optimization     : ML                             Residual df     =      2,616
+                                                  Scale parameter =          1
+Deviance         =  4029.442144                   (1/df) Deviance =   1.540307
+Pearson          =  3820.490076                   (1/df) Pearson  =   1.460432
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =   7.101991
+Log likelihood   = -9325.118318                   BIC             =  -16570.88
+
+------------------------------------------------------------------------------
+             |                 OIM
+  fluvaccine |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   47.99562   .7799153   238.23   0.000      46.4911    49.54882
+     levels2 |    48.2899   .7663704   244.31   0.000     46.81096    49.81556
+     levels3 |   49.35107   .7754712   248.13   0.000     47.85434    50.89462
+     levels4 |   49.66889   .7746225   250.41   0.000     48.17363    51.21056
+     levels5 |   50.96012   .7778188   257.55   0.000      49.4582    52.50765
+   homeorder |   .9198825   .0101901    -7.54   0.000     .9001255    .9400732
+             |
+        rucc |
+          2  |   .9804055    .012666    -1.53   0.126     .9558922    1.005547
+          3  |   .9893659   .0131858    -0.80   0.422     .9638569     1.01555
+          4  |   .9732812   .0148226    -1.78   0.075     .9446587    1.002771
+          5  |   .9212544   .0199685    -3.78   0.000     .8829365    .9612352
+          6  |   .8735529   .0106074   -11.13   0.000     .8530082    .8945924
+          7  |   .8206082   .0113219   -14.33   0.000     .7987151    .8431015
+          8  |   .8344541   .0164987    -9.15   0.000     .8027359    .8674256
+          9  |   .7862014   .0165937   -11.40   0.000     .7543418    .8194065
+------------------------------------------------------------------------------
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+Compare to tabular data:
 
 ----------------------------------------------------------
-Social    |
 Distancin |
 g: Lowest |
 (1) to    |
 Highest   |
 (5)       |    N(fluvac~e)  mean(fluvac~e)   sem(fluvac~e)
 ----------+-----------------------------------------------
-        1 |            498       39.339359        .3682052
-        2 |            497       41.714287        .3583028
-        3 |            506       42.600792        .3745272
-        4 |            503       42.842941        .3787197
-        5 |            502       44.537849        .3723575
+        1 |            517       40.218567        .3511693
+        2 |            526       41.165398        .3810684
+        3 |            513       42.111111        .3742107
+        4 |            539       42.922077        .3666867
+        5 |            535       44.278503        .3587537
 ----------------------------------------------------------
-

-

+----- PERCENT DIFFERENCE MODEL -----
+note: levels5 omitted because of collinearity
+
+Iteration 0:   log likelihood = -9326.8988  
+Iteration 1:   log likelihood = -9325.1183  
+Iteration 2:   log likelihood = -9325.1183  
+
+Generalized linear models                         Number of obs   =      2,630
+Optimization     : ML                             Residual df     =      2,616
+                                                  Scale parameter =          1
+Deviance         =  4029.442144                   (1/df) Deviance =   1.540307
+Pearson          =  3820.490076                   (1/df) Pearson  =   1.460432
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =   7.101991
+Log likelihood   = -9325.118318                   BIC             =  -16570.88
+
+------------------------------------------------------------------------------
+             |                 OIM
+  fluvaccine |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   .9418269   .0112491    -5.02   0.000     .9200352    .9641349
+     levels2 |   .9476017   .0110092    -4.63   0.000      .926268    .9694268
+     levels3 |   .9684253   .0112147    -2.77   0.006     .9466925     .990657
+     levels4 |   .9746619   .0110035    -2.27   0.023     .9533323    .9964687
+     levels5 |          1  (omitted)
+   homeorder |   .9198825   .0101901    -7.54   0.000     .9001255    .9400732
+             |
+        rucc |
+          2  |   .9804055    .012666    -1.53   0.126     .9558922    1.005547
+          3  |   .9893659   .0131858    -0.80   0.422     .9638569     1.01555
+          4  |   .9732812   .0148226    -1.78   0.075     .9446587    1.002771
+          5  |   .9212544   .0199685    -3.78   0.000     .8829365    .9612352
+          6  |   .8735529   .0106074   -11.13   0.000     .8530082    .8945924
+          7  |   .8206082   .0113219   -14.33   0.000     .7987151    .8431015
+          8  |   .8344541   .0164987    -9.15   0.000     .8027359    .8674256
+          9  |   .7862014   .0165937   -11.40   0.000     .7543418    .8194065
+             |
+       _cons |   50.96012   .7778188   257.55   0.000      49.4582    52.50765
+------------------------------------------------------------------------------
+Note: _cons estimates baseline incidence rate.
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+
+
@@ -13837,16 +14417,16 @@

Flu VaccinationFlu Vaccination @@ -13869,18 +14447,54 @@

Flu Vaccination +

+
-
+
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 47.99562 .7799153 238.23 0.000 46.4911 49.54882 + levels2 | 48.2899 .7663704 244.31 0.000 46.81096 49.81556 + levels3 | 49.35107 .7754712 248.13 0.000 47.85434 50.89462 + levels4 | 49.66889 .7746225 250.41 0.000 48.17363 51.21056 + levels5 | 50.96012 .7778188 257.55 0.000 49.4582 52.50765 +

+Percent difference
+levels1 | .9418269 .0112491 -5.02 0.000 .9200352 .9641349

+

The lowest social distancing counties had 48.0% flu vaccine coverage among Medicare beneficiaries, whereas the most social distancing counties had 51.0% after adjusting for rurality and social distancing orders, a 6.2% (95% CI: 3.7%, 8.7%) difference.

+
+
+
+
+
+
In [9]:
+
+
+
invert .9418269 .9200352    .9641349
+
-
-
+
+
+
+ +
+
+
+
+
+
6.2
+LL: 8.7
+UL: 3.7
@@ -13893,26 +14507,18 @@

Flu Vaccination
-

Income

There is a trend emerging. So, since the places with more social distancing seem to have more health resources, perhaps there are trends in financial means? Income data are from the American Community Survey 5-year estimates via RWJF.

- +

Economic

There is a trend emerging. So, since the places with more social distancing seem to have more health resources, perhaps there are trends in financial means? We explored two baseline economic metrics in relation to social distancing, one representing the overall wealth of the community and one proxy for poverty: 80th percentile of annual household income in dollars and the percent of school-age children eligible for subsidized or free lunches.

+

Household income

-
In [16]:
+
In [10]:
-
// Comparing income inequality to social distancing
-frame change default
-foreach var of varlist income80 {
-    table iso5, c(count `var' mean `var' sem `var')
-        frame put `var' iso5, into(`var')
-            frame change `var'
-                collapse (median) `var', by(iso5)
-                    la var `var' "80th Percentile Income ($)"
-                        line `var' iso5, note("Annual Income")          
-}
+
// Comparing 80th percentile income to social distancing
+modelrun income80
@@ -13929,23 +14535,130 @@

Income

Th

-
+
----- RURALITY-ADJUSTED NEGBIN MODEL -----
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -2.396e+09  
+Iteration 1:   log pseudolikelihood = -2.765e+08  (backed up)
+Iteration 2:   log pseudolikelihood = -2.287e+08  
+Iteration 3:   log pseudolikelihood = -7797007.7  
+Iteration 4:   log pseudolikelihood =   -3712358  
+Iteration 5:   log pseudolikelihood = -3655049.8  
+Iteration 6:   log pseudolikelihood =   -3655041  
+Iteration 7:   log pseudolikelihood =   -3655041  
 
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -32813.382  
+Iteration 1:   log pseudolikelihood = -29141.104  
+Iteration 2:   log pseudolikelihood = -29125.526  
+Iteration 3:   log pseudolikelihood = -29125.351  
+Iteration 4:   log pseudolikelihood = -29125.351  
+
+Negative binomial regression                    Number of obs     =      2,633
+Dispersion           = mean                     Wald chi2(14)     =   1.42e+07
+Log pseudolikelihood = -29125.351               Prob > chi2       =     0.0000
+
+------------------------------------------------------------------------------
+             |               Robust
+    income80 |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   118675.2   1873.364   740.18   0.000     115059.7    122404.3
+     levels2 |   122077.6   1869.469   764.83   0.000       118468    125797.3
+     levels3 |   125227.2   2022.182   726.89   0.000     121325.9      129254
+     levels4 |     129432   1971.797   772.66   0.000     125624.5    133354.9
+     levels5 |     139390   2261.967   729.93   0.000     135026.4    143894.6
+   homeorder |   .9368546    .008621    -7.09   0.000     .9201093    .9539046
+             |
+        rucc |
+          2  |   .8432181   .0117573   -12.23   0.000     .8204863    .8665797
+          3  |   .8122305   .0110667   -15.26   0.000     .7908273    .8342129
+          4  |   .7601974   .0110199   -18.91   0.000     .7389027    .7821058
+          5  |    .793445   .0154531   -11.88   0.000     .7637282    .8243182
+          6  |   .7209769    .009222   -25.58   0.000     .7031268    .7392802
+          7  |     .72983   .0106562   -21.57   0.000     .7092403    .7510174
+          8  |   .6908203   .0120309   -21.24   0.000     .6676381    .7148074
+          9  |   .6910585   .0140465   -18.18   0.000     .6640691    .7191448
+-------------+----------------------------------------------------------------
+    /lnalpha |   -3.62168   .0344743                     -3.689249   -3.554112
+-------------+----------------------------------------------------------------
+       alpha |   .0267377   .0009218                      .0249908    .0286068
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Compare to tabular data:
 
 ----------------------------------------------------------
-Social    |
 Distancin |
 g: Lowest |
 (1) to    |
 Highest   |
 (5)       |    N(income80)  mean(income80)   sem(income80)
 ----------+-----------------------------------------------
-        1 |            499       86193.695        636.8371
-        2 |            497       88885.761        652.4932
-        3 |            506       93321.575        671.9309
-        4 |            503       98127.835        856.1122
-        5 |            502       112882.41        1424.063
+        1 |            519       85968.287        590.9798
+        2 |            526       90843.228        733.2388
+        3 |            513       93619.183        872.0948
+        4 |            540       98925.328        890.1288
+        5 |            535       110406.09        1298.318
 ----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: levels5 omitted because of collinearity
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -3655043.1  
+Iteration 1:   log pseudolikelihood =   -3655041  
+Iteration 2:   log pseudolikelihood =   -3655041  
+
+Fitting constant-only model:
+
+Iteration 0:   log pseudolikelihood = -32840.589  
+Iteration 1:   log pseudolikelihood =   -29881.7  
+Iteration 2:   log pseudolikelihood = -29881.116  
+Iteration 3:   log pseudolikelihood = -29881.116  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -29307.258  
+Iteration 1:   log pseudolikelihood = -29157.452  
+Iteration 2:   log pseudolikelihood = -29125.374  
+Iteration 3:   log pseudolikelihood = -29125.351  
+Iteration 4:   log pseudolikelihood = -29125.351  
+
+Negative binomial regression                    Number of obs     =      2,633
+                                                Wald chi2(13)     =    1399.51
+Dispersion           = mean                     Prob > chi2       =     0.0000
+Log pseudolikelihood = -29125.351               Pseudo R2         =     0.0253
+
+------------------------------------------------------------------------------
+             |               Robust
+    income80 |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   .8513898   .0096039   -14.26   0.000      .832773    .8704228
+     levels2 |   .8757993   .0097536   -11.91   0.000     .8568898    .8951261
+     levels3 |   .8983949    .010434    -9.23   0.000     .8781757    .9190797
+     levels4 |   .9285604   .0105342    -6.53   0.000     .9081416    .9494384
+     levels5 |          1  (omitted)
+   homeorder |   .9368545    .008621    -7.09   0.000     .9201092    .9539046
+             |
+        rucc |
+          2  |   .8432181   .0117573   -12.23   0.000     .8204863    .8665797
+          3  |   .8122305   .0110667   -15.26   0.000     .7908273     .834213
+          4  |   .7601975   .0110199   -18.91   0.000     .7389028    .7821059
+          5  |    .793445   .0154531   -11.88   0.000     .7637282    .8243181
+          6  |   .7209769    .009222   -25.58   0.000     .7031268    .7392802
+          7  |     .72983   .0106562   -21.57   0.000     .7092403    .7510174
+          8  |   .6908203   .0120309   -21.24   0.000     .6676381    .7148075
+          9  |   .6910585   .0140465   -18.18   0.000      .664069    .7191448
+             |
+       _cons |   139389.9   2261.966   729.93   0.000     135026.3    143894.5
+-------------+----------------------------------------------------------------
+    /lnalpha |  -3.621674   .0344743                     -3.689242   -3.554105
+-------------+----------------------------------------------------------------
+       alpha |   .0267379   .0009218                      .0249909     .028607
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Note: _cons estimates baseline incidence rate.
@@ -13965,18 +14678,18 @@

Income

Th <rect x="0" y="0" width="1584" height="1584" style="fill:#C6D3DF;stroke:none"/> <rect x="0.00" y="0.05" width="1584.00" height="1583.95" style="fill:#C6D3DF"/> <rect x="1.58" y="1.63" width="1580.83" height="1580.78" style="fill:none;stroke:#C6D3DF;stroke-width:3.17"/> - <line x1="55.44" y1="1349.96" x2="1205.31" y2="1349.96" style="stroke:#FFFFFF;stroke-width:4.75"/> - <line x1="55.44" y1="1049.99" x2="1205.31" y2="1049.99" style="stroke:#FFFFFF;stroke-width:4.75"/> - <line x1="55.44" y1="749.97" x2="1205.31" y2="749.97" style="stroke:#FFFFFF;stroke-width:4.75"/> - <line x1="55.44" y1="450.00" x2="1205.31" y2="450.00" style="stroke:#FFFFFF;stroke-width:4.75"/> - <line x1="55.44" y1="149.98" x2="1205.31" y2="149.98" style="stroke:#FFFFFF;stroke-width:4.75"/> - <path d=" M64.90 1370.61 L347.65 1214.78 L630.35 932.38 L913.06 692.01 L1195.81 185.67" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/> - <text x="1227.34" y="1371.15" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">85000</text> - <text x="1227.34" y="1071.18" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">90000</text> - <text x="1227.34" y="771.16" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">95000</text> - <text x="1227.34" y="471.19" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">100000</text> - <text x="1227.34" y="171.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">105000</text> - <text x="1499.40" y="760.27" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,760.27)" text-anchor="middle">80th Percentile Income ($)</text> + <line x1="55.44" y1="1299.67" x2="1205.31" y2="1299.67" style="stroke:#FFFFFF;stroke-width:4.75"/> + <line x1="55.44" y1="1032.22" x2="1205.31" y2="1032.22" style="stroke:#FFFFFF;stroke-width:4.75"/> + <line x1="55.44" y1="764.73" x2="1205.31" y2="764.73" style="stroke:#FFFFFF;stroke-width:4.75"/> + <line x1="55.44" y1="497.23" x2="1205.31" y2="497.23" style="stroke:#FFFFFF;stroke-width:4.75"/> + <line x1="55.44" y1="229.73" x2="1205.31" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/> + <path d=" M64.90 1370.61 L347.65 1188.54 L630.35 1020.05 L913.06 795.12 L1195.81 262.40" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/> + <text x="1227.34" y="1320.86" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">120000</text> + <text x="1227.34" y="1053.41" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">125000</text> + <text x="1227.34" y="785.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">130000</text> + <text x="1227.34" y="518.41" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">135000</text> + <text x="1227.34" y="250.91" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">140000</text> + <text x="1499.40" y="800.17" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" transform="rotate(-90 1499.40,800.17)" text-anchor="middle">avg</text> <line x1="55.44" y1="1380.06" x2="1205.31" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/> <line x1="64.95" y1="1358.08" x2="64.95" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/> <text x="64.95" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">1</text> @@ -13989,9 +14702,7 @@

Income

Th <line x1="1195.81" y1="1358.08" x2="1195.81" y2="1380.06" style="stroke:#000000;stroke-width:3.17"/> <text x="1195.81" y="1433.37" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">5</text> <text x="630.40" y="1499.40" style="font-family:'Helvetica';font-size:60.49px;fill:#000000" text-anchor="middle">Social Distancing: Lowest (1) to Highest (5)</text> - <rect x="740.69" y="55.49" width="464.62" height="84.99" style="fill:#7B92A8"/> - <rect x="742.28" y="57.07" width="461.45" height="81.82" style="fill:none;stroke:#7B92A8;stroke-width:3.17"/> - <text x="1195.81" y="121.12" style="font-family:'Helvetica';font-size:65.99px;fill:#FFFFff" text-anchor="end">Annual Income</text> + <text x="90.29" y="156.84" style="font-family:'Helvetica';font-size:95.04px;fill:#000000">income80</text> </svg> "> @@ -14002,31 +14713,306 @@

Income

Th

+
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 118675.2 1873.364 740.18 0.000 115059.7 122404.3 + levels2 | 122077.6 1869.469 764.83 0.000 118468 125797.3 + levels3 | 125227.2 2022.182 726.89 0.000 121325.9 129254 + levels4 | 129432 1971.797 772.66 0.000 125624.5 133354.9 + levels5 | 139390 2261.967 729.93 0.000 135026.4 143894.6 +

+Percent difference
+levels1 | .8513898 .0096039 -14.26 0.000 .832773 .8704228

+

The lowest social distancing counties the 80th percentile of annual household income was around $120,000, whereas in the most social distancing counties it was $140,000, after adjusting for rurality and social distancing orders, a 17% (95% CI: 15%, 20%) difference.

+ +
+
+
+
+
+
In [11]:
+
+
+
invert .8513898 .832773 .8704228
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
17.5
+LL: 20.1
+UL: 14.9
+
+
+
+ +
+
+
-

Income Inequality

So places with more social distancing tend to be wealthier. What about income inequality? Is this merely a matter of have and have-nots? Or does the overall structure of society tell us something. There are many compelling ways to measure income inequality. We are using the ratio of the 80th percentile of income to the 20th percentile. Loosely interpreted, the closer to 4.0, the less inequality there is. Income data are from the American Community Survey 5-year estimates via RWJF.

+

Subsidized lunches

+
+
+
+
+
+
In [12]:
+
+
+
modelrun schoollunch
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
----- RURALITY-ADJUSTED NEGBIN MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -290091.97  
+Iteration 1:   log pseudolikelihood = -36368.036  
+Iteration 2:   log pseudolikelihood = -18347.941  
+Iteration 3:   log pseudolikelihood =  -13295.81  
+Iteration 4:   log pseudolikelihood = -13295.279  
+Iteration 5:   log pseudolikelihood = -13295.279  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -12572.459  
+Iteration 1:   log pseudolikelihood = -11473.257  
+Iteration 2:   log pseudolikelihood =  -11337.08  
+Iteration 3:   log pseudolikelihood = -10886.708  
+Iteration 4:   log pseudolikelihood = -10555.215  
+Iteration 5:   log pseudolikelihood = -10531.961  
+Iteration 6:   log pseudolikelihood = -10531.538  
+Iteration 7:   log pseudolikelihood = -10531.537  
+
+Negative binomial regression                    Number of obs     =      2,520
+Dispersion           = mean                     Wald chi2(14)     =  474857.54
+Log pseudolikelihood = -10531.537               Prob > chi2       =     0.0000
+
+------------------------------------------------------------------------------
+             |               Robust
+ schoollunch |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   40.67673   1.205293   125.06   0.000     38.38169    43.10901
+     levels2 |   37.48201   1.116649   121.64   0.000     35.35609    39.73576
+     levels3 |      35.28   1.065025   118.04   0.000     33.25315     37.4304
+     levels4 |   33.67241   1.010585   117.17   0.000     31.74882    35.71253
+     levels5 |   32.23657   .9852857   113.63   0.000     30.36215    34.22671
+   homeorder |   1.314872   .0259834    13.85   0.000     1.264919    1.366798
+             |
+        rucc |
+          2  |   1.137772   .0288759     5.09   0.000      1.08256    1.195799
+          3  |   1.123658   .0287948     4.55   0.000     1.068615    1.181537
+          4  |   1.199803   .0324922     6.73   0.000      1.13778    1.265206
+          5  |   1.219814   .0468732     5.17   0.000     1.131318    1.315232
+          6  |   1.288721   .0298801    10.94   0.000     1.231468    1.348636
+          7  |   1.254022   .0310345     9.15   0.000     1.194647    1.316348
+          8  |   1.279151   .0427903     7.36   0.000     1.197974    1.365829
+          9  |   1.327901   .0434874     8.66   0.000     1.245345     1.41593
+-------------+----------------------------------------------------------------
+    /lnalpha |  -2.628985   .0381112                     -2.703681   -2.554288
+-------------+----------------------------------------------------------------
+       alpha |   .0721517   .0027498                      .0669586    .0777476
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Compare to tabular data:
+
+----------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(school~h)  mean(school~h)   sem(school~h)
+----------+-----------------------------------------------
+        1 |            494       62.302429         .824396
+        2 |            499       56.844878        .7677724
+        3 |            485       53.521044        .8046446
+        4 |            523       50.439806        .7411138
+        5 |            519       47.833955        .7046043
+----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+note: levels5 omitted because of collinearity
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -13295.279  
+Iteration 1:   log pseudolikelihood = -13295.279  
+
+Fitting constant-only model:
+
+Iteration 0:   log pseudolikelihood = -12599.584  
+Iteration 1:   log pseudolikelihood = -11803.044  
+Iteration 2:   log pseudolikelihood = -11687.587  
+Iteration 3:   log pseudolikelihood = -11124.475  
+Iteration 4:   log pseudolikelihood = -10805.793  
+Iteration 5:   log pseudolikelihood = -10804.028  
+Iteration 6:   log pseudolikelihood = -10804.028  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -10560.506  
+Iteration 1:   log pseudolikelihood = -10531.679  
+Iteration 2:   log pseudolikelihood = -10531.537  
+Iteration 3:   log pseudolikelihood = -10531.537  
+
+Negative binomial regression                    Number of obs     =      2,520
+                                                Wald chi2(13)     =     555.41
+Dispersion           = mean                     Prob > chi2       =     0.0000
+Log pseudolikelihood = -10531.537               Pseudo R2         =     0.0252
+
+------------------------------------------------------------------------------
+             |               Robust
+ schoollunch |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |    1.26182   .0244192    12.02   0.000     1.214855    1.310599
+     levels2 |   1.162717   .0229728     7.63   0.000     1.118552    1.208626
+     levels3 |   1.094409   .0225473     4.38   0.000     1.051098    1.139506
+     levels4 |   1.044541   .0213772     2.13   0.033     1.003471    1.087291
+     levels5 |          1  (omitted)
+   homeorder |   1.314872   .0259834    13.85   0.000     1.264919    1.366798
+             |
+        rucc |
+          2  |   1.137772   .0288759     5.09   0.000      1.08256    1.195799
+          3  |   1.123658   .0287948     4.55   0.000     1.068615    1.181537
+          4  |   1.199803   .0324922     6.73   0.000      1.13778    1.265206
+          5  |   1.219814   .0468732     5.17   0.000     1.131318    1.315232
+          6  |   1.288721   .0298801    10.94   0.000     1.231468    1.348636
+          7  |   1.254022   .0310345     9.15   0.000     1.194647    1.316348
+          8  |   1.279151   .0427903     7.36   0.000     1.197974    1.365829
+          9  |   1.327901   .0434874     8.66   0.000     1.245345     1.41593
+             |
+       _cons |   32.23657   .9852857   113.63   0.000     30.36215    34.22671
+-------------+----------------------------------------------------------------
+    /lnalpha |  -2.628985   .0381112                     -2.703681   -2.554288
+-------------+----------------------------------------------------------------
+       alpha |   .0721517   .0027498                      .0669586    .0777476
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Note: _cons estimates baseline incidence rate.
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 40.67673 1.205293 125.06 0.000 38.38169 43.10901 + levels2 | 37.48201 1.116649 121.64 0.000 35.35609 39.73576 + levels3 | 35.28 1.065025 118.04 0.000 33.25315 37.4304 + levels4 | 33.67241 1.010585 117.17 0.000 31.74882 35.71253 + levels5 | 32.23657 .9852857 113.63 0.000 30.36215 34.22671 +

+Percent difference
+levels1 | 1.26182 .0244192 12.02 0.000 1.214855 1.310599

+

In the lowest social distancing counties, 41% of schoolage children were eligible for free or reduced price lunches. By comparison, in the most social distancing counties 32% were eligible, after adjusting for rurality and social distancing orders, a 26% (95% CI: 21%, 31%) difference.

+
+
+
+
+
+
+
+

Structural

Three lifestyle metrics were selected to provide a diverse snapshot of baseline structural factors that could influence defiance of prolonged stay-at-home orders. The percent of people experiencing food insecurity was derived from Map the Meal Gap project, based on responses from the Current Population Survey and a cost-of-food index. Access to exercise opportunities was the percent of population with adequate access to locations for physical activity. The percent of households with overcrowding was based on the Comprehensive Housing Affordability Strategy measurements.

+

Food insecurity

-
In [17]:
+
In [13]:
-
// Comparing income inequality to social distancing
-frame change default
-foreach var of varlist incomeratio {
-    table iso5, c(count `var' mean `var' sem `var')
-        frame put `var' iso5, into(`var')
-            frame change `var'
-                collapse (mean) `var', by(iso5)
-                    la var `var' "Income inequality (80:20)"
-                        line `var' iso5, note("Income Inequality")     
-}
+
modelpoisson foodinsec
@@ -14043,23 +15029,107 @@

Income Inequality -
+
----- RURALITY-ADJUSTED POISSON MODEL -----
+note: foodinsec has noninteger values
+
+Iteration 0:   log likelihood = -7184.3437  
+Iteration 1:   log likelihood = -7182.0944  
+Iteration 2:   log likelihood = -7182.0943  
 
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  2683.121099                   (1/df) Deviance =   1.024483
+Pearson          =   2797.25662                   (1/df) Pearson  =   1.068063
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =    5.46608
+Log likelihood   = -7182.094311                   BIC             =  -17943.81
+
+------------------------------------------------------------------------------
+             |                 OIM
+   foodinsec |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   12.93501   .3216935   102.93   0.000     12.31962    13.58114
+     levels2 |   11.84488   .2918237   100.33   0.000     11.28651    12.43088
+     levels3 |   11.64897   .2858669   100.05   0.000     11.10194    12.22295
+     levels4 |   10.89867   .2676669    97.26   0.000     10.38648    11.43612
+     levels5 |   10.17634   .2475631    95.37   0.000     9.702515    10.67331
+   homeorder |   1.088589   .0183214     5.04   0.000     1.053266    1.125097
+             |
+        rucc |
+          2  |    1.10029   .0228933     4.59   0.000     1.056322    1.146087
+          3  |   1.104049    .023559     4.64   0.000     1.058826    1.151203
+          4  |   1.149369   .0274078     5.84   0.000     1.096887    1.204363
+          5  |   1.131871   .0367085     3.82   0.000     1.062163    1.206155
+          6  |   1.165615    .022003     8.12   0.000     1.123278    1.209548
+          7  |   1.135392   .0237196     6.08   0.000     1.089841    1.182846
+          8  |   1.219673   .0342862     7.06   0.000     1.154291    1.288758
+          9  |   1.189152   .0350341     5.88   0.000     1.122431    1.259839
+------------------------------------------------------------------------------
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+Compare to tabular data:
 
 ----------------------------------------------------------
-Social    |
 Distancin |
 g: Lowest |
 (1) to    |
 Highest   |
-(5)       |    N(income~o)  mean(income~o)   sem(income~o)
+(5)       |    N(foodin~c)  mean(foodin~c)   sem(foodin~c)
 ----------+-----------------------------------------------
-        1 |            499       4.8361004        .0371413
-        2 |            497        4.628354        .0297515
-        3 |            506       4.4730893        .0277805
-        4 |            503       4.3748814        .0270911
-        5 |            502       4.3956305        .0311182
+        1 |            519       15.759152        .1923148
+        2 |            526         14.3327        .1790774
+        3 |            513       14.082261         .174559
+        4 |            540       13.073889        .1640634
+        5 |            535       12.078692        .1441107
 ----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: levels5 omitted because of collinearity
+note: foodinsec has noninteger values
+
+Iteration 0:   log likelihood = -7184.3437  
+Iteration 1:   log likelihood = -7182.0944  
+Iteration 2:   log likelihood = -7182.0943  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  2683.121099                   (1/df) Deviance =   1.024483
+Pearson          =   2797.25662                   (1/df) Pearson  =   1.068063
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =    5.46608
+Log likelihood   = -7182.094311                   BIC             =  -17943.81
+
+------------------------------------------------------------------------------
+             |                 OIM
+   foodinsec |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   1.271086   .0226617    13.45   0.000     1.227437    1.316288
+     levels2 |   1.163963   .0207984     8.50   0.000     1.123904    1.205449
+     levels3 |   1.144711   .0205843     7.52   0.000     1.105069    1.185774
+     levels4 |   1.070981   .0192196     3.82   0.000     1.033966    1.109321
+     levels5 |          1  (omitted)
+   homeorder |   1.088589   .0183214     5.04   0.000     1.053266    1.125097
+             |
+        rucc |
+          2  |    1.10029   .0228933     4.59   0.000     1.056322    1.146087
+          3  |   1.104049    .023559     4.64   0.000     1.058826    1.151203
+          4  |   1.149369   .0274078     5.84   0.000     1.096887    1.204363
+          5  |   1.131871   .0367085     3.82   0.000     1.062163    1.206155
+          6  |   1.165615    .022003     8.12   0.000     1.123278    1.209548
+          7  |   1.135392   .0237196     6.08   0.000     1.089841    1.182846
+          8  |   1.219673   .0342862     7.06   0.000     1.154291    1.288758
+          9  |   1.189152   .0350341     5.88   0.000     1.122431    1.259839
+             |
+       _cons |   10.17634   .2475631    95.37   0.000     9.702515    10.67331
+------------------------------------------------------------------------------
+Note: _cons estimates baseline incidence rate.
+(Standard errors scaled using square root of Pearson X2-based dispersion.)

@@ -14079,33 +15149,29 @@

Income Inequality @@ -14116,6 +15182,1941 @@

Income Inequality
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 12.93501 .3216935 102.93 0.000 12.31962 13.58114 + levels2 | 11.84488 .2918237 100.33 0.000 11.28651 12.43088 + levels3 | 11.64897 .2858669 100.05 0.000 11.10194 12.22295 + levels4 | 10.89867 .2676669 97.26 0.000 10.38648 11.43612 + levels5 | 10.17634 .2475631 95.37 0.000 9.702515 10.67331 +

+Percent difference
+levels1 | 1.271086 .0226617 13.45 0.000 1.227437 1.316288

+

The lowest social distancing counties had greater food insecurity, among 12.9% of residents. The most social distancing counties had 10.2%, after adjusting for rurality and social distancing orders, a 27% (95% CI: 23%, 32%) difference.

+
+ +
+
+

+
+
+
+

Exercise opportunities

+
+
+
+
+
+
In [14]:
+
+
+
modelrun exercise
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
----- RURALITY-ADJUSTED NEGBIN MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -397571.32  
+Iteration 1:   log pseudolikelihood = -24410.221  
+Iteration 2:   log pseudolikelihood = -16708.317  
+Iteration 3:   log pseudolikelihood = -16405.591  
+Iteration 4:   log pseudolikelihood = -16405.296  
+Iteration 5:   log pseudolikelihood = -16405.296  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -13629.269  
+Iteration 1:   log pseudolikelihood = -12618.453  
+Iteration 2:   log pseudolikelihood = -12003.767  
+Iteration 3:   log pseudolikelihood =  -11968.11  
+Iteration 4:   log pseudolikelihood = -11967.878  
+Iteration 5:   log pseudolikelihood = -11967.878  
+
+Negative binomial regression                    Number of obs     =      2,633
+Dispersion           = mean                     Wald chi2(14)     =  725576.80
+Log pseudolikelihood = -11967.878               Prob > chi2       =     0.0000
+
+------------------------------------------------------------------------------
+             |               Robust
+    exercise |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   70.82756   1.902913   158.57   0.000     67.19442    74.65715
+     levels2 |   76.05735   1.841866   178.86   0.000      72.5317    79.75439
+     levels3 |   78.16616   1.906506   178.71   0.000     74.51738     81.9936
+     levels4 |   82.19944   1.871535   193.65   0.000     78.61194    85.95066
+     levels5 |   91.41497   2.000766   206.31   0.000     87.57646    95.42173
+   homeorder |   .9263275   .0160495    -4.42   0.000     .8953991    .9583242
+             |
+        rucc |
+          2  |   .9315932   .0170231    -3.88   0.000     .8988189    .9655625
+          3  |   .9221747   .0189775    -3.94   0.000     .8857196    .9601303
+          4  |   .9217404   .0170166    -4.41   0.000     .8889847     .955703
+          5  |   1.031934   .0218027     1.49   0.137     .9900745    1.075564
+          6  |   .7820271   .0142184   -13.52   0.000     .7546503    .8103971
+          7  |    .864374   .0185513    -6.79   0.000     .8287684    .9015094
+          8  |   .6065847   .0300327   -10.10   0.000     .5504875    .6683983
+          9  |   .6802969   .0328991    -7.97   0.000     .6187774    .7479328
+-------------+----------------------------------------------------------------
+    /lnalpha |  -2.222633   .0655469                     -2.351102   -2.094163
+-------------+----------------------------------------------------------------
+       alpha |   .1083235   .0071003                      .0952641    .1231733
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Compare to tabular data:
+
+----------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(exercise)  mean(exercise)   sem(exercise)
+----------+-----------------------------------------------
+        1 |            519       56.906538        .9544925
+        2 |            526       61.787498        .9064765
+        3 |            513       63.396324          .89163
+        4 |            540       68.048161        .8530419
+        5 |            535        76.76011        .8015522
+----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+note: levels5 omitted because of collinearity
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -16405.303  
+Iteration 1:   log pseudolikelihood = -16405.296  
+Iteration 2:   log pseudolikelihood = -16405.296  
+
+Fitting constant-only model:
+
+Iteration 0:   log pseudolikelihood = -13662.944  
+Iteration 1:   log pseudolikelihood = -12842.373  
+Iteration 2:   log pseudolikelihood = -12215.778  
+Iteration 3:   log pseudolikelihood = -12215.652  
+Iteration 4:   log pseudolikelihood = -12215.652  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -11993.273  
+Iteration 1:   log pseudolikelihood = -11968.037  
+Iteration 2:   log pseudolikelihood = -11967.878  
+Iteration 3:   log pseudolikelihood = -11967.878  
+
+Negative binomial regression                    Number of obs     =      2,633
+                                                Wald chi2(13)     =     690.24
+Dispersion           = mean                     Prob > chi2       =     0.0000
+Log pseudolikelihood = -11967.878               Pseudo R2         =     0.0203
+
+------------------------------------------------------------------------------
+             |               Robust
+    exercise |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   .7747917   .0157072   -12.59   0.000     .7446098    .8061969
+     levels2 |   .8320011   .0150512   -10.17   0.000      .803018    .8620301
+     levels3 |   .8550695   .0146972    -9.11   0.000     .8267433    .8843663
+     levels4 |   .8991901   .0137357    -6.96   0.000     .8726677    .9265186
+     levels5 |          1  (omitted)
+   homeorder |   .9263275   .0160495    -4.42   0.000     .8953991    .9583241
+             |
+        rucc |
+          2  |   .9315932   .0170231    -3.88   0.000     .8988189    .9655625
+          3  |   .9221747   .0189775    -3.94   0.000     .8857196    .9601303
+          4  |   .9217404   .0170166    -4.41   0.000     .8889847     .955703
+          5  |   1.031934   .0218027     1.49   0.137     .9900745    1.075564
+          6  |   .7820271   .0142184   -13.52   0.000     .7546503    .8103971
+          7  |    .864374   .0185513    -6.79   0.000     .8287684    .9015094
+          8  |   .6065847   .0300327   -10.10   0.000     .5504875    .6683983
+          9  |   .6802969   .0328991    -7.97   0.000     .6187774    .7479328
+             |
+       _cons |   91.41497   2.000766   206.31   0.000     87.57646    95.42173
+-------------+----------------------------------------------------------------
+    /lnalpha |  -2.222633   .0655469                     -2.351102   -2.094163
+-------------+----------------------------------------------------------------
+       alpha |   .1083235   .0071003                      .0952641    .1231733
+------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Note: _cons estimates baseline incidence rate.
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 70.82756 1.902913 158.57 0.000 67.19442 74.65715 + levels2 | 76.05735 1.841866 178.86 0.000 72.5317 79.75439 + levels3 | 78.16616 1.906506 178.71 0.000 74.51738 81.9936 + levels4 | 82.19944 1.871535 193.65 0.000 78.61194 85.95066 + levels5 | 91.41497 2.000766 206.31 0.000 87.57646 95.42173 +

+Percent difference
+levels1 | .7747917 .0157072 -12.59 0.000 .7446098 .8061969

+

In the lowest social distancing counties, 69% of residents had access to physical spaces for exercise, whereas in the most social distancing counties 90% had access, after adjusting for rurality and social distancing orders, a 32% (95% CI: 26%, 40%) difference.

+ +
+
+
+
+
+
In [15]:
+
+
+
invert .7747917 .7446098 .8061969
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
29.1
+LL: 34.3
+UL: 24
+
+
+
+ +
+
+ +
+
+
+
+
+

Overcrowding

+
+
+
+
+
+
In [16]:
+
+
+
modelpoisson overcrowding
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
----- RURALITY-ADJUSTED POISSON MODEL -----
+note: overcrowding has noninteger values
+
+Iteration 0:   log likelihood = -4756.5858  
+Iteration 1:   log likelihood = -4744.9748  
+Iteration 2:   log likelihood = -4744.9632  
+Iteration 3:   log likelihood = -4744.9632  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  2650.197886                   (1/df) Deviance =   1.011912
+Pearson          =  3524.251156                   (1/df) Pearson  =   1.345648
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =    3.61486
+Log likelihood   = -4744.963158                   BIC             =  -17976.73
+
+------------------------------------------------------------------------------
+             |                 OIM
+overcrowding |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   2.367699   .1589751    12.84   0.000     2.075745    2.700717
+     levels2 |   2.162979   .1437291    11.61   0.000     1.898848     2.46385
+     levels3 |   1.876126   .1261733     9.36   0.000     1.644436     2.14046
+     levels4 |   1.934634   .1285511     9.93   0.000     1.698396    2.203733
+     levels5 |   2.068613   .1338108    11.24   0.000     1.822293    2.348229
+   homeorder |   1.167809    .055477     3.27   0.001     1.063985    1.281765
+             |
+        rucc |
+          2  |   1.029311    .055311     0.54   0.591     .9264173    1.143634
+          3  |   .9498813   .0537281    -0.91   0.363     .8502035    1.061245
+          4  |   .9483685   .0611599    -0.82   0.411     .8357637    1.076145
+          5  |    1.15092   .0952304     1.70   0.089       .97862    1.353555
+          6  |    1.04307    .051264     0.86   0.391      .947282    1.148544
+          7  |   .9874639   .0545025    -0.23   0.819     .8862162    1.100279
+          8  |   1.001858   .0774844     0.02   0.981     .8609413    1.165839
+          9  |    .817196   .0710091    -2.32   0.020     .6892269    .9689251
+------------------------------------------------------------------------------
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+Compare to tabular data:
+
+----------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(overcr~g)  mean(overcr~g)   sem(overcr~g)
+----------+-----------------------------------------------
+        1 |            519       2.6832882        .0894385
+        2 |            526       2.4762081        .0830075
+        3 |            513       2.1469685        .0588481
+        4 |            540       2.2162362        .0747709
+        5 |            535        2.388474        .0858484
+----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: levels5 omitted because of collinearity
+note: overcrowding has noninteger values
+
+Iteration 0:   log likelihood = -4756.5858  
+Iteration 1:   log likelihood = -4744.9748  
+Iteration 2:   log likelihood = -4744.9632  
+Iteration 3:   log likelihood = -4744.9632  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  2650.197886                   (1/df) Deviance =   1.011912
+Pearson          =  3524.251156                   (1/df) Pearson  =   1.345648
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =    3.61486
+Log likelihood   = -4744.963158                   BIC             =  -17976.73
+
+------------------------------------------------------------------------------
+             |                 OIM
+overcrowding |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   1.144583   .0535491     2.89   0.004     1.044297      1.2545
+     levels2 |   1.045618   .0488222     0.96   0.339     .9541759    1.145823
+     levels3 |   .9069486   .0439847    -2.01   0.044     .8247106    .9973871
+     levels4 |   .9352325   .0440449    -1.42   0.155     .8527704    1.025669
+     levels5 |          1  (omitted)
+   homeorder |   1.167809    .055477     3.27   0.001     1.063985    1.281765
+             |
+        rucc |
+          2  |   1.029311    .055311     0.54   0.591     .9264173    1.143634
+          3  |   .9498813   .0537281    -0.91   0.363     .8502035    1.061245
+          4  |   .9483685   .0611599    -0.82   0.411     .8357637    1.076145
+          5  |    1.15092   .0952304     1.70   0.089       .97862    1.353555
+          6  |    1.04307    .051264     0.86   0.391      .947282    1.148544
+          7  |   .9874639   .0545025    -0.23   0.819     .8862162    1.100279
+          8  |   1.001858   .0774844     0.02   0.981     .8609413    1.165839
+          9  |    .817196   .0710091    -2.32   0.020     .6892269    .9689251
+             |
+       _cons |   2.068613   .1338108    11.24   0.000     1.822293    2.348229
+------------------------------------------------------------------------------
+Note: _cons estimates baseline incidence rate.
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 2.367699 .1589751 12.84 0.000 2.075745 2.700717 + levels2 | 2.162979 .1437291 11.61 0.000 1.898848 2.46385 + levels3 | 1.876126 .1261733 9.36 0.000 1.644436 2.14046 + levels4 | 1.934634 .1285511 9.93 0.000 1.698396 2.203733 + levels5 | 2.068613 .1338108 11.24 0.000 1.822293 2.348229 +

+Percent difference
+levels1 | 1.144583 .0535491 2.89 0.004 1.044297 1.2545

+

The lowest social distancing counties had 14% (95% CI: 4.4%, 25%) less overcrowding, after adjusting for rurality and social distancing orders.

+ +
+
+
+
+
+
+
+

Sociodemographics

Youth

Less than age 18

+ +
+
+
+
+
+
In [17]:
+
+
+
modelpoisson youth
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
----- RURALITY-ADJUSTED POISSON MODEL -----
+note: youth has noninteger values
+
+Iteration 0:   log likelihood = -7047.4698  
+Iteration 1:   log likelihood = -7047.2764  
+Iteration 2:   log likelihood = -7047.2764  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  1084.966828                   (1/df) Deviance =   .4142676
+Pearson          =  1091.163655                   (1/df) Pearson  =   .4166337
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =   5.363674
+Log likelihood   = -7047.276406                   BIC             =  -19541.96
+
+------------------------------------------------------------------------------
+             |                 OIM
+       youth |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   25.54333   .3027171   273.42   0.000     24.95686    26.14359
+     levels2 |   24.97196   .2904891   276.61   0.000     24.40905    25.54785
+     levels3 |   24.65616   .2854101   276.88   0.000     24.10306    25.22195
+     levels4 |   24.76482   .2843834   279.48   0.000     24.21366    25.32852
+     levels5 |   23.59149   .2675883   278.67   0.000     23.07281    24.12183
+   homeorder |   .9355478   .0075003    -8.31   0.000     .9209623    .9503643
+             |
+        rucc |
+          2  |   .9687538    .009452    -3.25   0.001     .9504042    .9874577
+          3  |   .9474401   .0095882    -5.34   0.000     .9288327    .9664203
+          4  |   .9468317   .0109397    -4.73   0.000     .9256313    .9685176
+          5  |   .9792884    .015412    -1.33   0.184     .9495426    1.009966
+          6  |   .9535839   .0085333    -5.31   0.000     .9370048    .9704565
+          7  |   .9514734    .009464    -5.00   0.000      .933104    .9702044
+          8  |   .9060676   .0129889    -6.88   0.000     .8809641    .9318865
+          9  |   .8965563   .0133864    -7.31   0.000     .8706997    .9231807
+------------------------------------------------------------------------------
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+Compare to tabular data:
+
+-------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(youth)  mean(youth)   sem(youth)
+----------+--------------------------------------
+        1 |         519    23.013114     .1388531
+        2 |         526    22.562249     .1310086
+        3 |         513    22.287634     .1227711
+        4 |         540    22.396111     .1464031
+        5 |         535    21.385909     .1390452
+-------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: levels5 omitted because of collinearity
+note: youth has noninteger values
+
+Iteration 0:   log likelihood = -7047.4698  
+Iteration 1:   log likelihood = -7047.2764  
+Iteration 2:   log likelihood = -7047.2764  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  1084.966828                   (1/df) Deviance =   .4142676
+Pearson          =  1091.163655                   (1/df) Pearson  =   .4166337
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =   5.363674
+Log likelihood   = -7047.276406                   BIC             =  -19541.96
+
+------------------------------------------------------------------------------
+             |                 OIM
+       youth |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   1.082735   .0095077     9.05   0.000      1.06426    1.101531
+     levels2 |   1.058516   .0091489     6.58   0.000     1.040735      1.0766
+     levels3 |   1.045129   .0090803     5.08   0.000     1.027483    1.063079
+     levels4 |   1.049735   .0089242     5.71   0.000     1.032389    1.067373
+     levels5 |          1  (omitted)
+   homeorder |   .9355478   .0075003    -8.31   0.000     .9209623    .9503643
+             |
+        rucc |
+          2  |   .9687538    .009452    -3.25   0.001     .9504042    .9874577
+          3  |   .9474401   .0095882    -5.34   0.000     .9288327    .9664203
+          4  |   .9468317   .0109397    -4.73   0.000     .9256313    .9685176
+          5  |   .9792884    .015412    -1.33   0.184     .9495426    1.009966
+          6  |   .9535839   .0085333    -5.31   0.000     .9370048    .9704565
+          7  |   .9514734    .009464    -5.00   0.000      .933104    .9702044
+          8  |   .9060676   .0129889    -6.88   0.000     .8809641    .9318865
+          9  |   .8965563   .0133864    -7.31   0.000     .8706997    .9231807
+             |
+       _cons |   23.59149   .2675883   278.67   0.000     23.07281    24.12183
+------------------------------------------------------------------------------
+Note: _cons estimates baseline incidence rate.
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 25.54333 .3027171 273.42 0.000 24.95686 26.14359 + levels2 | 24.97196 .2904891 276.61 0.000 24.40905 25.54785 + levels3 | 24.65616 .2854101 276.88 0.000 24.10306 25.22195 + levels4 | 24.76482 .2843834 279.48 0.000 24.21366 25.32852 + levels5 | 23.59149 .2675883 278.67 0.000 23.07281 24.12183 +

+Percent difference
+levels1 | 1.082735 .0095077 9.05 0.000 1.06426 1.101531

+

Counties with the least restriction of movement had 8.2% more children (95% CI: 6.4%, 10%) than areas that most greatly had their movement reduced.

+ +
+
+
+
+
+
+
+

Elderly

Interpretation

+
+
+
+
+
+
In [18]:
+
+
+
modelpoisson elderly
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
----- RURALITY-ADJUSTED POISSON MODEL -----
+note: elderly has noninteger values
+
+Iteration 0:   log likelihood = -7218.0267  
+Iteration 1:   log likelihood = -7216.9383  
+Iteration 2:   log likelihood = -7216.9383  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  2003.292916                   (1/df) Deviance =   .7649076
+Pearson          =  2072.529905                   (1/df) Pearson  =    .791344
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =   5.492547
+Log likelihood   = -7216.938253                   BIC             =  -18623.63
+
+------------------------------------------------------------------------------
+             |                 OIM
+     elderly |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |   14.16821   .2681426   140.07   0.000     13.65229    14.70363
+     levels2 |   14.60455   .2709946   144.50   0.000     14.08295    15.14547
+     levels3 |   14.94807   .2750933   146.96   0.000      14.4185    15.49708
+     levels4 |   15.00888   .2744157   148.15   0.000     14.48055    15.55647
+     levels5 |   15.22254   .2732011   151.71   0.000     14.69638    15.76753
+   homeorder |   1.035171   .0129279     2.77   0.006     1.010141    1.060822
+             |
+        rucc |
+          2  |   1.124375   .0177546     7.42   0.000      1.09011    1.159718
+          3  |   1.149063   .0186141     8.58   0.000     1.113153    1.186131
+          4  |   1.182938   .0214397     9.27   0.000     1.141655    1.225715
+          5  |   1.096782   .0282309     3.59   0.000     1.042823    1.153533
+          6  |   1.268995    .018046    16.75   0.000     1.234114    1.304862
+          7  |   1.284184   .0200334    16.03   0.000     1.245514    1.324056
+          8  |   1.391649    .028851    15.94   0.000     1.336236    1.449361
+          9  |   1.430253   .0305757    16.74   0.000     1.371563    1.491453
+------------------------------------------------------------------------------
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+Compare to tabular data:
+
+-------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(elderly)  mean(elderly)   sem(elderly)
+----------+--------------------------------------------
+        1 |           519      17.897361       .1736063
+        2 |           526       18.07345       .1721838
+        3 |           513      18.483562        .162852
+        4 |           540      18.246843       .1874652
+        5 |           535      18.115094       .2083171
+-------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: levels5 omitted because of collinearity
+note: elderly has noninteger values
+
+Iteration 0:   log likelihood = -7218.0267  
+Iteration 1:   log likelihood = -7216.9383  
+Iteration 2:   log likelihood = -7216.9383  
+
+Generalized linear models                         Number of obs   =      2,633
+Optimization     : ML                             Residual df     =      2,619
+                                                  Scale parameter =          1
+Deviance         =  2003.292916                   (1/df) Deviance =   .7649076
+Pearson          =  2072.529905                   (1/df) Pearson  =    .791344
+
+Variance function: V(u) = u                       [Poisson]
+Link function    : g(u) = ln(u)                   [Log]
+
+                                                  AIC             =   5.492547
+Log likelihood   = -7216.938253                   BIC             =  -18623.63
+
+------------------------------------------------------------------------------
+             |                 OIM
+     elderly |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+-------------+----------------------------------------------------------------
+     levels1 |    .930739   .0124421    -5.37   0.000     .9066697    .9554473
+     levels2 |   .9594032   .0125665    -3.16   0.002     .9350867    .9843519
+     levels3 |   .9819697   .0128292    -1.39   0.164     .9571442    1.007439
+     levels4 |   .9859642   .0126631    -1.10   0.271     .9614548    1.011098
+     levels5 |          1  (omitted)
+   homeorder |   1.035171   .0129279     2.77   0.006     1.010141    1.060822
+             |
+        rucc |
+          2  |   1.124375   .0177546     7.42   0.000      1.09011    1.159718
+          3  |   1.149063   .0186141     8.58   0.000     1.113153    1.186131
+          4  |   1.182938   .0214397     9.27   0.000     1.141655    1.225715
+          5  |   1.096782   .0282309     3.59   0.000     1.042823    1.153533
+          6  |   1.268995    .018046    16.75   0.000     1.234114    1.304862
+          7  |   1.284184   .0200334    16.03   0.000     1.245514    1.324056
+          8  |   1.391649    .028851    15.94   0.000     1.336236    1.449361
+          9  |   1.430253   .0305757    16.74   0.000     1.371563    1.491453
+             |
+       _cons |   15.22254   .2732011   151.71   0.000     14.69638    15.76753
+------------------------------------------------------------------------------
+Note: _cons estimates baseline incidence rate.
+(Standard errors scaled using square root of Pearson X2-based dispersion.)
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 14.16821 .2681426 140.07 0.000 13.65229 14.70363 + levels2 | 14.60455 .2709946 144.50 0.000 14.08295 15.14547 + levels3 | 14.94807 .2750933 146.96 0.000 14.4185 15.49708 + levels4 | 15.00888 .2744157 148.15 0.000 14.48055 15.55647 + levels5 | 15.22254 .2732011 151.71 0.000 14.69638 15.76753 +

+Percent difference
+levels1 | .930739 .0124421 -5.37 0.000 .9066697 .9554473

+

Counties that did the best at restricting movement had 7.4% (95% CI: 4.7%, 10%) more elderly people, compared to the lowest tier of movement restriction.

+ +
+
+
+
+
+
In [19]:
+
+
+
invert .930739 .9066697 .9554473
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
7.4
+LL: 10.3
+UL: 4.7
+
+
+
+ +
+
+ +
+
+
+
+
+

Segregation

Interpretation

+
+
+
+
+
+
In [20]:
+
+
+
modelrun segregation_wnw
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
----- RURALITY-ADJUSTED NEGBIN MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -137517.44  
+Iteration 1:   log pseudolikelihood = -14752.896  
+Iteration 2:   log pseudolikelihood = -13227.944  
+Iteration 3:   log pseudolikelihood = -13227.003  
+Iteration 4:   log pseudolikelihood = -13227.003  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -11513.071  
+Iteration 1:   log pseudolikelihood = -10469.271  
+Iteration 2:   log pseudolikelihood =  -10203.36  
+Iteration 3:   log pseudolikelihood =  -10195.34  
+Iteration 4:   log pseudolikelihood = -10195.337  
+Iteration 5:   log pseudolikelihood = -10195.337  
+
+Negative binomial regression                    Number of obs     =      2,581
+Dispersion           = mean                     Wald chi2(14)     =  216209.08
+Log pseudolikelihood = -10195.337               Prob > chi2       =     0.0000
+
+-------------------------------------------------------------------------------
+              |               Robust
+segregatio~nw |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+--------------+----------------------------------------------------------------
+      levels1 |   29.07819   1.106746    88.54   0.000     26.98794    31.33033
+      levels2 |   29.62044   1.078855    93.03   0.000     27.57963    31.81226
+      levels3 |   30.05066   1.117545    91.50   0.000     27.93823    32.32281
+      levels4 |   30.79493   1.091354    96.71   0.000     28.72851    33.00999
+      levels5 |     31.866    1.11741    98.72   0.000     29.74948     34.1331
+    homeorder |   1.075855   .0295754     2.66   0.008     1.019423    1.135412
+              |
+         rucc |
+           2  |   1.023216   .0261958     0.90   0.370     .9731398    1.075869
+           3  |   1.018929    .028167     0.68   0.498     .9651917    1.075658
+           4  |   1.020022   .0288734     0.70   0.484     .9649719    1.078212
+           5  |   1.028869   .0393519     0.74   0.457     .9545604    1.108962
+           6  |   .9387745   .0235422    -2.52   0.012     .8937482    .9860692
+           7  |   .9473715   .0303327    -1.69   0.091     .8897474    1.008728
+           8  |   .7978725   .0400495    -4.50   0.000     .7231146    .8803591
+           9  |   .8521205   .0532593    -2.56   0.010     .7538747    .9631698
+--------------+----------------------------------------------------------------
+     /lnalpha |  -1.962348   .0443816                     -2.049334   -1.875361
+--------------+----------------------------------------------------------------
+        alpha |   .1405281   .0062369                      .1288207    .1532996
+-------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Compare to tabular data:
+
+----------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(segreg~w)  mean(segreg~w)   sem(segreg~w)
+----------+-----------------------------------------------
+        1 |            506       29.687411        .5589038
+        2 |            513       30.582416        .5315276
+        3 |            498       31.042838        .5548652
+        4 |            535       32.147839        .5309895
+        5 |            529       33.599273        .5652224
+----------------------------------------------------------
+----- PERCENT DIFFERENCE MODEL -----
+note: you are responsible for interpretation of non-count dep. variable
+note: levels5 omitted because of collinearity
+
+Fitting Poisson model:
+
+Iteration 0:   log pseudolikelihood = -13227.003  
+Iteration 1:   log pseudolikelihood = -13227.003  
+
+Fitting constant-only model:
+
+Iteration 0:   log pseudolikelihood = -11521.003  
+Iteration 1:   log pseudolikelihood = -10513.242  
+Iteration 2:   log pseudolikelihood = -10247.006  
+Iteration 3:   log pseudolikelihood =   -10241.9  
+Iteration 4:   log pseudolikelihood = -10241.899  
+
+Fitting full model:
+
+Iteration 0:   log pseudolikelihood = -10196.302  
+Iteration 1:   log pseudolikelihood = -10195.338  
+Iteration 2:   log pseudolikelihood = -10195.337  
+
+Negative binomial regression                    Number of obs     =      2,581
+                                                Wald chi2(13)     =      87.17
+Dispersion           = mean                     Prob > chi2       =     0.0000
+Log pseudolikelihood = -10195.337               Pseudo R2         =     0.0045
+
+-------------------------------------------------------------------------------
+              |               Robust
+segregatio~nw |        IRR   Std. Err.      z    P>|z|     [95% Conf. Interval]
+--------------+----------------------------------------------------------------
+      levels1 |   .9125145   .0241028    -3.47   0.001     .8664759    .9609994
+      levels2 |   .9295311   .0233124    -2.91   0.004     .8849443    .9763643
+      levels3 |   .9430319   .0234616    -2.36   0.018     .8981512    .9901554
+      levels4 |   .9663884   .0228272    -1.45   0.148     .9226678    1.012181
+      levels5 |          1  (omitted)
+    homeorder |   1.075855   .0295754     2.66   0.008     1.019423    1.135412
+              |
+         rucc |
+           2  |   1.023216   .0261958     0.90   0.370     .9731399    1.075869
+           3  |   1.018929    .028167     0.68   0.498     .9651917    1.075658
+           4  |   1.020022   .0288734     0.70   0.484     .9649719    1.078212
+           5  |   1.028869   .0393519     0.74   0.457     .9545604    1.108962
+           6  |   .9387745   .0235422    -2.52   0.012     .8937482    .9860692
+           7  |   .9473715   .0303327    -1.69   0.091     .8897475    1.008728
+           8  |   .7978726   .0400495    -4.50   0.000     .7231146    .8803592
+           9  |   .8521205   .0532593    -2.56   0.010     .7538748    .9631698
+              |
+        _cons |     31.866    1.11741    98.72   0.000     29.74948     34.1331
+--------------+----------------------------------------------------------------
+     /lnalpha |  -1.962348   .0443816                     -2.049334   -1.875362
+--------------+----------------------------------------------------------------
+        alpha |   .1405281   .0062369                      .1288207    .1532995
+-------------------------------------------------------------------------------
+Note: Estimates are transformed only in the first equation.
+Note: _cons estimates baseline incidence rate.
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+

Interpretation



+

Relative effect measures
+^ + levels1 | 29.07819 1.106746 88.54 0.000 26.98794 31.33033 + levels2 | 29.62044 1.078855 93.03 0.000 27.57963 31.81226 + levels3 | 30.05066 1.117545 91.50 0.000 27.93823 32.32281 + levels4 | 30.79493 1.091354 96.71 0.000 28.72851 33.00999 + levels5 | 31.866 1.11741 98.72 0.000 29.74948 34.1331 +

+Percent difference
+levels1 | .9125145 .0241028 -3.47 0.001 .8664759 .9609994

+

The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders.

+ +
+
+
+
+
+
In [21]:
+
+
+
invert .9125145  .8664759  .9609994
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
9.6
+LL: 15.4
+UL: 4.1
+
+
+
+ +
+
+ +
+
+
+
+
+

Exploratory analyses

+
+
+
+
+
+
In [22]:
+
+
+
frame change default
+foreach var of varlist drivealone_p {
+    table iso5, c(count `var' mean `var' sem `var')
+        frame put `var' iso5, into(`var')
+            frame change `var'
+                collapse (mean) `var', by(iso5)
+                    la var `var' "% of Drivers"
+                        line `var' iso5, note("Commuting Alone by Vehicle")  
+                            frame change default
+                                frame drop `var'
+}
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
+
+
+----------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(drivea~p)  mean(drivea~p)   sem(drivea~p)
+----------+-----------------------------------------------
+        1 |            519       82.089322        .2316678
+        2 |            526       81.973571        .1901521
+        3 |            513       81.401447        .1971413
+        4 |            540        80.95994        .1918464
+        5 |            535       77.732723        .3661999
+----------------------------------------------------------
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+
+ +
+
+
+
+
+
In [23]:
+
+
+
frame change default
+foreach var of varlist rucc {
+    table iso5, c(count `var' mean `var' sem `var')
+        frame put `var' iso5, into(`var')
+            frame change `var'
+                collapse (median) `var', by(iso5)
+                    la var `var' "Meidan RUCC"
+                        line `var' iso5, note("Urban-Rural")   
+                            frame change default
+                                frame drop `var'
+}
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
+
+
+----------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(rucc)  mean(rucc)   sem(rucc)
+----------+-----------------------------------
+        1 |        519   5.2524085    .0987557
+        2 |        526     4.63308    .1022326
+        3 |        513    4.594542    .1055764
+        4 |        540    4.062963     .104029
+        5 |        535   3.6093459    .1101034
+----------------------------------------------
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
In [24]:
+
+
+
frame change default
+foreach var of varlist longcommute_p {
+    table iso5, c(count `var' mean `var' sem `var')
+        frame put `var' iso5, into(`var')
+            frame change `var'
+                collapse (mean) `var', by(iso5)
+                    la var `var' "% of Solo Commuters Driving 30+ mins"
+                        line `var' iso5, note("Long Solo Commute")     
+                            frame change default
+                                frame drop `var'
+}
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
+
+
+----------------------------------------------------------
+Distancin |
+g: Lowest |
+(1) to    |
+Highest   |
+(5)       |    N(longco~p)  mean(longco~p)   sem(longco~p)
+----------+-----------------------------------------------
+        1 |            519       29.061464        .5501718
+        2 |            526       31.741635         .530849
+        3 |            513       32.373489         .503766
+        4 |            540       32.046667        .5223995
+        5 |            535        33.41215         .499872
+----------------------------------------------------------
+
+
+
+ +
+ +
+ + + +
+ + +
+ +
+ +
+
+ +
+
+
+
+
+

External validation with Google data

Correlations between March 1 to April 11 in Google and DL data by county-day

+ +
+
+
+
+
+
In [25]:
+
+
+
// Import Descartes Labs data
+clear
+import delimited "https://raw.githubusercontent.com/descarteslabs/DL-COVID-19/master/DL-us-mobility-daterow.csv", encoding(ISO-8859-9) stringcols(6) 
+
+    di "Drop state aggregates:"
+        drop if admin2==""
+    di "Drop improbable outliers (n=490 or <0.4% of observations N=130,685):"
+        drop if m50_index>200
+
+    * Format date
+        gen date2=date(date,"YMD")
+            format date2 %td
+                drop date
+                    rename date2 date
+
+    * Note data start and end dates for graphs
+        su date 
+            local latest: disp %td r(max)
+                di "`latest'"
+            local earliest: disp %td r(min)
+                di "`earliest'"
+
+    * Rename variables for consistency
+        rename admin1 state
+        rename admin2 county
+
+
+save dl_x_valid, replace
+
+// Process Google app check-in data
+clear
+import delimited "/Users/nabarun/Documents/GitHub/covid/fips-google-mobility-daily-as-of-04-20-20.csv", stringcols(1) numericcols(5 6 8)  
+
+    * Format date
+        gen date=date(report_date, "YMD")
+            format date %td
+                order date, first
+                    drop report_date
+
+    * Note data start and end dates for graphs
+        su date 
+            local latest: disp %td r(max)
+                di "Latest: " "`latest'"
+            local earliest: disp %td r(min)
+                di "Earliest: " "`earliest'"
+
+    save google_x_valid, replace
+    
+
+// Merge by date and county, retain
+    merge 1:1 date fips using dl_x_valid
+
+    tab _merge
+        keep if _merge==3
+            drop _merge country_code admin_level
+
+    * Variable cleanup
+    destring retail_and_recreation_percent_ch grocery_and_pharmacy_percent_cha workplaces_percent_change_from_b, replace force
+    rename retail_and_recreation_percent_ch retailrec
+    rename grocery_and_pharmacy_percent_cha grocery
+    rename workplaces_percent_change_from_b work
+    rename residential_percent_change_from_ home
+    rename parks_percent_change_from_baseli parks
+        la var parks "Parks"
+    rename transit_stations_percent_change_ transit
+                la var transit "Transit"
+    
+    gen m50i = m50_index-100
+            la var m50i "Re-centered at zero"
+
+    * Correlation matrix on complete case data (n=20,891)
+    correlate m50i retailrec grocery work home parks transit
+        matrix C = r(C)
+     
+    * Pairwise correlations
+    correlate m50i retailrec
+    correlate m50i grocery
+    correlate m50i work
+    correlate m50i home
+    correlate m50i parks
+    correlate m50i transit
+
+frame change default
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
+
+(9 vars, 133,285 obs)
+
+Drop state aggregates:
+
+(2,600 observations deleted)
+
+Drop improbable outliers (n=490 or <0.4% of observations N=130,685):
+
+(490 observations deleted)
+
+
+
+
+
+
+    Variable |        Obs        Mean    Std. Dev.       Min        Max
+-------------+---------------------------------------------------------
+        date |    130,195    21999.54    14.42399      21975      22024
+
+
+19apr2020
+
+
+01mar2020
+
+
+
+file dl_x_valid.dta saved
+
+
+(8 vars, 152,039 obs)
+
+
+
+
+
+
+    Variable |        Obs        Mean    Std. Dev.       Min        Max
+-------------+---------------------------------------------------------
+        date |    152,039     21988.1    16.42271      21960      22016
+
+
+Latest: 11apr2020
+
+
+Earliest: 15feb2020
+
+file google_x_valid.dta saved
+
+
+    Result                           # of obs.
+    -----------------------------------------
+    not matched                        70,280
+        from master                    46,062  (_merge==1)
+        from using                     24,218  (_merge==2)
+
+    matched                           105,977  (_merge==3)
+    -----------------------------------------
+
+
+                 _merge |      Freq.     Percent        Cum.
+------------------------+-----------------------------------
+        master only (1) |     46,062       26.13       26.13
+         using only (2) |     24,218       13.74       39.87
+            matched (3) |    105,977       60.13      100.00
+------------------------+-----------------------------------
+                  Total |    176,257      100.00
+
+(70,280 observations deleted)
+
+
+retail_and_recreation_percent_ch: contains nonnumeric characters; replaced as in
+> t
+(13296 missing values generated)
+grocery_and_pharmacy_percent_cha: contains nonnumeric characters; replaced as in
+> t
+(16918 missing values generated)
+workplaces_percent_change_from_b: contains nonnumeric characters; replaced as by
+> te
+(4687 missing values generated)
+
+
+
+
+
+
+
+
+
+
+
+(obs=20,880)
+
+             |     m50i retail~c  grocery     work     home    parks  transit
+-------------+---------------------------------------------------------------
+        m50i |   1.0000
+   retailrec |   0.8886   1.0000
+     grocery |   0.7168   0.7586   1.0000
+        work |   0.8442   0.9074   0.6948   1.0000
+        home |  -0.8197  -0.8908  -0.6763  -0.9734   1.0000
+       parks |   0.2293   0.3274   0.2975   0.2521  -0.2896   1.0000
+     transit |   0.7759   0.8348   0.6688   0.8106  -0.8111   0.3484   1.0000
+
+
+
+(obs=92,681)
+
+             |     m50i retail~c
+-------------+------------------
+        m50i |   1.0000
+   retailrec |   0.7649   1.0000
+
+
+(obs=89,059)
+
+             |     m50i  grocery
+-------------+------------------
+        m50i |   1.0000
+     grocery |   0.5353   1.0000
+
+
+(obs=101,290)
+
+             |     m50i     work
+-------------+------------------
+        m50i |   1.0000
+        work |   0.7471   1.0000
+
+
+(obs=49,666)
+
+             |     m50i     home
+-------------+------------------
+        m50i |   1.0000
+        home |  -0.8297   1.0000
+
+
+(obs=27,182)
+
+             |     m50i    parks
+-------------+------------------
+        m50i |   1.0000
+       parks |   0.2424   1.0000
+
+
+(obs=41,299)
+
+             |     m50i  transit
+-------------+------------------
+        m50i |   1.0000
+     transit |   0.7088   1.0000
+
+
+
+
+
+ +
+
+ +
+
+
+
+
+

Figure Data

+
+
+
+
+
+
In [28]:
+
+
+
frame change results
+list
+
+ +
+
+
+ +
+
+ + +
+ +
+ + +
+
+
+
+     +----------------------------------------------------------+
+     |           strat   level        avg         LL         UL |
+     |----------------------------------------------------------|
+  1. |        pcp_rate       1       49.9       45.6       54.7 |
+  2. |        pcp_rate       2       54.9       50.3       59.9 |
+  3. |        pcp_rate       3       57.9       52.8       63.5 |
+  4. |        pcp_rate       4       61.5       56.3       67.1 |
+  5. |        pcp_rate       5       73.6       67.6       80.2 |
+     |----------------------------------------------------------|
+  6. |     uninsured_p       1       10.7         10       11.5 |
+  7. |     uninsured_p       2        9.5        8.8       10.2 |
+  8. |     uninsured_p       3        8.5        7.9        9.1 |
+  9. |     uninsured_p       4          8        7.4        8.6 |
+ 10. |     uninsured_p       5          7        6.5        7.6 |
+     |----------------------------------------------------------|
+ 11. |      fluvaccine       1         48       46.5       49.5 |
+ 12. |      fluvaccine       2       48.3       46.8       49.8 |
+ 13. |      fluvaccine       3       49.4       47.9       50.9 |
+ 14. |      fluvaccine       4       49.7       48.2       51.2 |
+ 15. |      fluvaccine       5         51       49.5       52.5 |
+     |----------------------------------------------------------|
+ 16. |        income80       1   118675.2   115059.7   122404.3 |
+ 17. |        income80       2   122077.6     118468   125797.3 |
+ 18. |        income80       3   125227.2   121325.9     129254 |
+ 19. |        income80       4     129432   125624.5   133354.9 |
+ 20. |        income80       5     139390   135026.4   143894.6 |
+     |----------------------------------------------------------|
+ 21. |     schoollunch       1       40.7       38.4       43.1 |
+ 22. |     schoollunch       2       37.5       35.4       39.7 |
+ 23. |     schoollunch       3       35.3       33.3       37.4 |
+ 24. |     schoollunch       4       33.7       31.7       35.7 |
+ 25. |     schoollunch       5       32.2       30.4       34.2 |
+     |----------------------------------------------------------|
+ 26. |       foodinsec       1       12.9       12.3       13.6 |
+ 27. |       foodinsec       2       11.8       11.3       12.4 |
+ 28. |       foodinsec       3       11.6       11.1       12.2 |
+ 29. |       foodinsec       4       10.9       10.4       11.4 |
+ 30. |       foodinsec       5       10.2        9.7       10.7 |
+     |----------------------------------------------------------|
+ 31. |        exercise       1       70.8       67.2       74.7 |
+ 32. |        exercise       2       76.1       72.5       79.8 |
+ 33. |        exercise       3       78.2       74.5         82 |
+ 34. |        exercise       4       82.2       78.6         86 |
+ 35. |        exercise       5       91.4       87.6       95.4 |
+     |----------------------------------------------------------|
+ 36. |    overcrowding       1        2.4        2.1        2.7 |
+ 37. |    overcrowding       2        2.2        1.9        2.5 |
+ 38. |    overcrowding       3        1.9        1.6        2.1 |
+ 39. |    overcrowding       4        1.9        1.7        2.2 |
+ 40. |    overcrowding       5        2.1        1.8        2.3 |
+     |----------------------------------------------------------|
+ 41. |           youth       1       25.5         25       26.1 |
+ 42. |           youth       2         25       24.4       25.5 |
+ 43. |           youth       3       24.7       24.1       25.2 |
+ 44. |           youth       4       24.8       24.2       25.3 |
+ 45. |           youth       5       23.6       23.1       24.1 |
+     |----------------------------------------------------------|
+ 46. |         elderly       1       14.2       13.7       14.7 |
+ 47. |         elderly       2       14.6       14.1       15.1 |
+ 48. |         elderly       3       14.9       14.4       15.5 |
+ 49. |         elderly       4         15       14.5       15.6 |
+ 50. |         elderly       5       15.2       14.7       15.8 |
+     |----------------------------------------------------------|
+ 51. | segregation_wnw       1       29.1         27       31.3 |
+ 52. | segregation_wnw       2       29.6       27.6       31.8 |
+ 53. | segregation_wnw       3       30.1       27.9       32.3 |
+ 54. | segregation_wnw       4       30.8       28.7         33 |
+ 55. | segregation_wnw       5       31.9       29.7       34.1 |
+     +----------------------------------------------------------+
+
+
+
+ +
+
+ +
+
+
+
+
+

Methods Detail

Baseline Health Data

In order to identify explanatory health and socioeconomic indicators, we used the 2019 Robert Wood Johnson Foundation (RWJF) County Health Rankings (CHR) dataset.20,23 The publicly available dataset contains dozens of metrics compiled from national surveys and healthcare databases. It is a well-documented public health resource,20 including data from the American Community Survey and Center for Medicare and Medicaid Services.

+

We compared the intensity of social distancing to 11 metrics: three healthcare, two economic, three structural, and three demographic. These were selected from the CHR dataset because they are established indicators of other health and behavioral outcomes,20 with an emphasis on emergent concerns about equality arising during the pandemic.

+

Healthcare metrics

To gauge overall baseline healthcare access and utilization, we examined primary care providers per 100,000 population and percent uninsured under age 65 (e.g., Medicare eligibility). As a marker for a closely related preventive health behavior, we examined whether earlier influenza vaccination rates were associated with how much the county was likely to slow down during the current coronavirus outbreak. This was quantified as the percent of annual Medicare enrollees having an annual influenza vaccination.

+

Economic metrics

We explored two baseline economic metrics, one representing the overall wealth of the community and one proxy for poverty: 80th percentile of annual household income in dollars and the percent of school-age children eligible for subsidized or free lunches.

+

Structural metrics

Three lifestyle metrics were selected to provide a diverse snapshot of baseline structural factors that could influence defiance of prolonged stay-at-home orders. The percent of people experiencing food insecurity was established from survey responses and a cost-of-food index. Access to exercise opportunities was the percent of population with adequate access to locations for physical activity. The percent of households with overcrowding was based on housing condition surveys.

+

Demographic metrics

The three demographic metrics were: percent of youth (age under 18 years) because of concerns about non-compliance with stay-at-home orders, the percent of elderly (aged 65 years and above) because they are risk group for COVID-19 mortality, and a residential segregation index (white versus non-white).

+

Primary Mobility Data

The analytic dataset started with public, anonymized, aggregated county-level (or similar geopolitical units) data from smartphone GPS movement tracing, collected from January 1 through April 15, 2020 in the United States, pre-processed by Descartes Labs (Santa Fe, New Mexico, United States). Raw mobility data generated from location services were processed using a parallel bucket sort to create device-based (e.g., node) records that for a given day are longitudinal.24 Maximum distance mobility (Mmax) was defined as by the maximum Haversine (great circle) distance in kilometers from the first location report.7 Conceptually, this represents the straight-line distance between the first observation and the day’s farthest. Across all reports, the median accuracy of location measurement was 15 to 20 meters. Adjustments were made for poor GPS signal locks, too few observations (less than 10 reports per day), and signal accuracy (50 meter threshold).7 To reduce bias from devices merely transiting through a county (e.g., interstate highways), Mmax is limited to nodes with 8 hours of observation per day. The resulting analysis dataset contained 2,633 of 3,142 US counties. +
+The pre-processed mobility dataset in this analysis could not be used to identify individuals. +Using geotemporally coarse data provides further safeguards for protecting privacy. To this end, the values of Mmax were summarized by taking the median per county-day (m50). The median was indexed against the baseline period February 17 through March 7, 2020 (e.g., before widespread social distancing) to obtain our main study metric: the percentage change in mobility since baseline (m50_index) as a proxy for the intensity of social distancing.

+

Variable Construction

To account for weekly periodicity in movement (e.g., lower on weekends) we limited analysis to weekdays. To prevent undue influence from single-day variability, we averaged the last 3 weekday values of m50_index: April 13, 14, 15. The resulting distribution approximated a Gaussian function; to reduce outlier influence, we constructed a 5-level inverted stratification of m50_index, where the highest quintile (5) represented the greatest reduction in mobility since baseline, interpreted as the highest 20% of counties in terms of social distancing intensity. Category boundaries for m50_index by quintile were: lowest mobility change (1) +193% to -45.8%, (2) -46.0% to -55.4%, (3) -55.5% to -62.3%, (4) 62.4% to 74.8%, and highest (5) -75.0% to -100%. Only 7 counties (all less than 20,000 population) showed an increase in mobility from baseline and were included in the lowest change category.

+

Potential Confounders

We adjusted models for two potential confounders. First, state and municipal stay-at-home-orders issued from February through April 2020 were identified by county.25 These orders limited travel to basic necessities and employment in sectors deemed essential. Municipal stay-at-home orders were identified for the eight states that did not have stay-at-home orders. +
+Second, even though our main outcome was change in mobility from baseline, rurality might be considered a potential confounder due to distances traveled for essential activities. We used federal rural-urban continuum codes (RUCC) to adjust for rurality and transportation connections between city centers and satellite counties.26 Although RUCC can be conceptualized as a 9-point ordinal scale of urbanicity (or rurality), it was modeled using indicator coding to impose fewer assumptions. Other potential spatial and economic confounders (number of solo vehicle commuters, longer vehicle commuting times, and CHR socioeconomic status composite rank) were not included in the final model because they did not meaningful improve model fit.

+

External Validation

In the context of social distancing, coarse mobility data have the potential for misclassification. One way to cross-validate the findings is to compare these data to more granular location information, such as by type of visited venue. The data came from aggregated and anonymized GPS traces of devices for which the Location History setting within Google apps had been turned on (off by default). Since detailed information on data collection was not available, we did not consider the Google Location Services data appropriate for the primary analysis. +
+During the study period, Google published county-level datasets showing COVID-19-related mobility changes across six types of venues: grocery and pharmacy; parks, transit stations, retail and recreation, places of residence; and places of work. The metric was percent change in mobility changes since baseline, January 3 to February 6, 2020, controlling for day of week. A validation dataset was created for March 1 to April 11, the overlap period with mobility data used in the primary analyses, for which county-day could be established. Pearson product-moments were calculated for correlations between the six venue-specific changes in mobility and percent change in mobility from baseline in the primary location data. To have confidence in overall mobility change to serve a proxy for social distancing, we expected the strongest correlations with staying at home, transit, work and retail, and less correlation with use of other venues that were permissible or essential during stay-at-home orders.

+

Statistical Analysis

Datasets were analyzed with Stata MP (version 16, College Station, Texas, United States). Scaled Poisson regression with robust variance estimators was used in base models regressing each of the 11 metrics individually against quintiles of mobility change. Negative binomial (NB2) models were employed when warranted by further overdispersion. The adjusted models included indicator variables to control for rurality/urbanicity. For a given health or socioeconomic indicator, mean and 95% confidence intervals were calculated for each quintile using non-intercept models; pairwise contrasts of percent difference between quintiles were estimated using the full model with adjustment. We intentionally did not include all explanatory variables in a combined multivariable model because we wanted to highlight the individual associations, not explain away the variance. Adjusted model-predicted means and confidence intervals of bivariate associations were plotted visually. For cross-validation, pairwise pearson product-moment correlations were generated, comparing zero-recentered m50_index against percent change from baseline by venue, as reported by Google Location Services users. Code and datasets are available at https://github.com/opioiddatalab/covid.

+ +
+
diff --git a/docs/output_13_1.pdf b/docs/output_13_1.pdf deleted file mode 100644 index 723acaa26a583ddc96c61a32379fe09d62193725..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1525 zcmZWp2~ZPf7#&7XST#_^TCgUoLJ>qZ8weqVW~h+}4N(Sg>`d`kk{{ekvbfnGP*JB< z5hWgo0wn@P84mDA473l4R^5U{Zsn>8OBa0qu_8XQHG-;YeKjpc*CvH zu*$f#u|-FE4+olfW-aPJ+=q-zVBj8LVOavffW<&bsQ@eilW7?m5bCK&vZ3stf|`yy zZ8r}t*tm59(eU_b%!2tIoZw3t@)>8BN6gAF+CMeBRBRB*jZ?|2lg6{k3(s3pO}tYP ziej6O^P*!t>o1M#5y7 zK0i@s!o8dcE~wIFUn_oMI9aA-+`Ixs>w4!Oa=Ehfc2$>-i_Ht?dMJwikmH4yTz~V< zY}=MK@4LL~k&}LU_)b1>T@iJ|Rk8E9t)Qtb-&iqCY6;F-A8ekq=Avs``@`7WC8<-y zQ6;-d0_4>Ny49N-N$o!=*--~tuG2Rg(;HRb2}!h+#F0j zSsimYZnlX5oR!X|*~A^*X8#qS!G5UAgA4;-$MGKMON1e>~clhhE?wsbZ+=<$2Dd?wsk{@o>}YP?bYrQbboYkDp;gZe&-9Hs?mygc zS(3XxJVUqT)|;N|&WHP?Bxh$s;+i6hzP<8csf!^$jpEh@HS6bR>R*U)}&MY@?-c}(hy_wqH*T31uz)we6D;fT^M@)s4 z0;>)YE*^ncQH=;@*qB(!A=n~93#{V9LP0_jU`Qo_U{ShQ#%f+{B#vMq8Ym)C6OUpm zNX?fTEja>=a+72l#36h%0O5Z)sJ~F?D@5a_e!y_3TlS!u|1f{(Lt`*p0KIK2763;W zgZYZa<->8&I4+Khq0NqP8JZvy0ByDCNQ#1YasSaeTlP?;IQ<>uQgy-dVWlI2~}sdxCSKmIz>h z-s=I77Qt)bT_OxR`vg?@94Z3`2g(G^i|j!P zI9(1V#0Uk9a0NOfM-`37=ws;L*fP_rOzD7RJyYdtZty_pLC-7& z5vvWA3gPW~+D#VrZMWS`R?}`Ze`7zbA7|U0{R#}I1KQo%e(dHy`bEx*V||!3)zMQE z{fm)^!y*#-uX8@E9;TI5QwHFfcGMAZ`r` cFfcGMFfcJNFfx-r1w#rjIXMa?B}Gq03RPkpz5oCK delta 697 zcmV;q0!IC}3XBV|w*h}uO>+}5488BK@DUCy`m!x^?9jpi4$urIE}3*&wv#498m7a) zN49s9ZJR*>j|EJGV(km#UQisJKV1G>#s15Ye{{m_5|hhAGw_M1qXt~oIr zfF!fe-$Y*AG=bZc)ZIPl``@&TnW+97-DShr)|p{Q?2QzSbSi(F>3Ng9*+ip9mp}7} zNVXG@7~D-m5eWxvY393VpTdP8xW^UkIXEH(zr$7S9*)%{Rlowhwv>=Cilu@RVX#pW zQ1QozPDQ}n#b`_DV*vApQX*7IWEz1dahQlTYavl`Boesrf>G?ae>Nwx=IxoqOGIvV zS&|O8@`o((i#mQ-vXLq708P^BaX z!YUt67&(T_h`Q&bLy3^G`C*rGFB30kMJpD3NXrLk^A+%3N|q`li~yL7%?z5I3O@EFfcG6 fZVd`BFfcGMFflMVH - - - - - 38 - 40 - 42 - 44 - 46 - Medicare Beneficiary Flu Vaccination % + + + + + + + 7 + 8 + 9 + 10 + 11 + avg 1 @@ -28,7 +30,5 @@ 5 Social Distancing: Lowest (1) to Highest (5) - - - Elderly Flu Vaccination + uninsured_p diff --git a/docs/output_17_1.pdf b/docs/output_17_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..09be2f60d77658960188afbc4d797352a08f9638 GIT binary patch literal 1528 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtI9QntT!4tLqDr6=wvIZG#GwkQ6to076S@A;CfZ!P2Y-FbiJEz}q9Ex*UO+-${PwY{;|=SZHb4w^lGzGr>y?IRafseMm>f9w2Z zwi7O2b4$W~8@BzsY%X=fvrHx?^QAnpky`pb>BC8;PAQosGM64KZ2MwT z7UZF9ZS5O!(^L8UrtI_cg03#~zM&c+Aoq6f{XHxn+0FZz?v#svSiQjKN3zJuKUz*R zuf!Ogm$y7vROi59P-y>S&Q#Y%Ue8%m4O|%ZtgL6{J)-E>%wU+@#r8;qZx>^u;xnBq zu8|ypC;d%b`f62jG)^Y+mN6aB&bZ)k*@>lbp+t_y!v$;t@vjmWvUOYfo}O4-b>%X{ zk~zk|FE`d!w01c#&6}3;z>#6$_x8sJ`p+%?So7!VoVByQKQCLRv?w}u>gm(zvH?~T zx^I3^yp>&gd(X%GyRXte7hbKglbRjm!+N`_!cr<{`IW;TS#~Vix%$xI(+@oyllhkm zKR8uZ9A@GFF!;^Fr=Nn}f0}jh(2V_Q8u2zKf^2^+ynQeC-G93^F^%o^lE*Apq`W*b z@zb13^2=oO3-0Jg`Q1D?;mDGpv!Wl253w}!EZJIbm6{*%zb7tilSObB*7 zRA1NkZ;Dk_LE;57*JDMx8Id1%Z-4*Y|8D=;Uq#nATBz7mMHh`!}un z?YaEA-TBjBe@(k>d*MH`?uB+0P-y~5-LP^6lp8=9#N5aTCIre?usj1Pn?SjaOW!H6 zI2D}#^gU8@%2G=*lM}i0gG-Z2z~vW6r6HHTYaTEkW#*+T=!a+KIp!5-!UT{Cl!~I% zG%f=L3oZjVP%tw!H8xd9Vii($c;=<$D}ahsV0jjtS(OU(3b@!vEGhx} Q&Ctlwh)Y$~)!&T^044lE00000 literal 0 HcmV?d00001 diff --git a/docs/output_13_1.svg b/docs/output_17_1.svg similarity index 67% rename from docs/output_13_1.svg rename to docs/output_17_1.svg index 777e89e..3ecbaee 100644 --- a/docs/output_13_1.svg +++ b/docs/output_17_1.svg @@ -7,37 +7,37 @@ - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + - - - - - - + + + + + + 0 - .01 - .02 - .03 - .04 - .05 + .01 + .02 + .03 + .04 + .05 Density diff --git a/docs/output_18_1.pdf b/docs/output_18_1.pdf index 2438e934c681bfdc7a1497eb6ed90842ed4ab19c..c5aa805cfcb1493b6f809801ac74fc802ca6f5b1 100644 GIT binary patch delta 589 zcmV-T0IPvPo7Y80ZvNS>MH?r;)?#grk zQEh$skL>x@rO>E~e>@rc-`Ic!)QyH)x`0jhn4op+W_Q_e?S_0G{z4y=Q&^PcE}m?Z*Or8l;bWW_|= zz@UW5=BPk5Ucg!?TDY9;RZ2E&9LAa;LnJluj|#X{vnr#Q}WN%_>3N|1xAa7!73N@251o9RzIW{#QFfcGMFd%LX b3NSD*FfcGMIg>gCMG7!DF$yImMNdWwcp?&Z delta 689 zcmV;i0#5y<3W*D_w*h}uOK;mS48HqU@VFi3@gYiLkHuaEcG;TaE=7=LDVhXrGZg#x zqb%F;f^^F=8?Z>3A3utsL~K)vApy3TQ_wQ{K=8P1@P%JrKQ;&mq`4LIXCgf=noGwa z26OxJPUQKziPD;7cz)6M-)V&kQT;mIZ6nmynSL-)CnYDH%5Hyp-I%&fa(aUDx7K*Vep3evxxMl~P zU}S!1E@#c#4<3IlGsEn*+h8n3K%L-Z$@dVHIm2w-yb5i24YRxE{jB+NGMZ3R=;3+> zT+Ops;ql+$OdMkmPiMemBX5@g3$%q6IOpiO@;CqN~%-p-VQcG3_)XXn<{?~Ca1?eq`?xmoP&oSYRJce z>T0nktn&GUiTyk{Bjr6OA4^16^5ZHy<$P8WwQxu)M`+98iUV!0{0FTWMyD0tvxZ}= zeu~q`T9xxzIWnaKRz0VWN%_>3N|1xAa7!73O19@1M(IzFg7(HFfcGMFd%LX3NSD* XFfcGNFq75 - - - - - - - 4.4 - 4.5 - 4.6 - 4.7 - 4.8 - Income inequality (80:20) - + + + + + + 48 + 49 + 50 + 51 + avg + 1 - - 2 - - 3 - - 4 - - 5 - Social Distancing: Lowest (1) to Highest (5) - - - Income Inequality + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + fluvaccine diff --git a/docs/output_16_1.pdf b/docs/output_22_1.pdf similarity index 53% rename from docs/output_16_1.pdf rename to docs/output_22_1.pdf index c1bc322b388a0ab80950d81389b38fcef55eb8f5..6a5cb3e6c201add6f3249b870a47cd2acdb0f378 100644 GIT binary patch delta 590 zcmV-U0KOC-!3FB6q48UN3V0Tb8j1wbkSv9Anqs95&@(qG+P0wuOJ*I}H|-Odg>;#)wEB zLK1je}dT9!Smc~{|LuxbWe^{&B@l@SJ9^RB`r4tm5@?++YjvvT%W zp^tU|7K&nxXC;ibi+ZYjj0jSejZ2;wlIB8&#wykVRxVx^J2ni;r6D935>SlVWW6a2 z(m&np%)R?)=@Net%pU3q`*I7g8VDpw1p;G&*~Y!Gg6ww#vq$$~=DuD{fIe9(8bg@I z5Rs%Y@U$EV1LXt5v>YfZQ)7_x98TeXV=QcM3oUz6b0h~F(wI3`;`y)Dk+p&PizH>~ z?@@KydG;}k1QKEyrDpdb*C*GPtBE?wVmS)8dPkC~H}HQT*zD8Iyz~r?O-tQLc`4il&U-t9uV!d6xhcCO=<;xzv`{2gB@E9;TI5QwHFfcGMAZ`r` cFfcGMFfcJNFfx-u1w#rjIXMa?B}Gq03XVb!WdHyG delta 700 zcmV;t0z>_`3Xco0w*h}u$&M2-5WV{=eBb~Fnp(VaW5g1G{4Vt zyVD(_WQdwXvR&<}_sX@{)WjHEwA3Vt&X*Yz+lPgrGh07C8M0*Ar$#e=2djtHT)~dW zW$2gx!Fs+nL7SYDeL9Q%Z(8L-R7cR=8s(;b>8Gq7qzux-HB^7EP4>`!qN=@L7UOru?uF0xBG(rDL7f7xlmDkb91 z^d1R-Dmw&Tvv9!dGD(N0%+{4*Z6b+UB&ImQ64hDk23MrykTy;vg-9Ix2wLfnNnjYL zf0?V9dGitOiim%m-F6wKQU%;C*ouqUffL@%*5+mMMD)te?#+jp`F^oIlZFbAX&OLq zWxoJS^FYL<3cxhY1xem%k8te6q4{%&1?_c#RYLvP?4oawr&w_l?e6Ti_F>*`cFR?3 zzij5)b!&fu?ga)O87tt>KV=yDhlTn??{$>LI*-~|_v?SX%H@vlLy@C2k{tZs$Kr%j zU1|1&bv>Oh1at=&IgOlhDi8>wA5K|?>8vb@Za$*bQ?%%{Fwj=vKWGzE0IkZ%>Q1oM zp6*$DVLGd5;&{aBr&RZF4BR}8wnJ+hsxAyDyseRma z<;|tM?QVZ-KQ9+|`;Y#p<{NadmiNd<@nsG^p{Khi>-1Jhkjp`>WDa7prT~vP)`tsQ z*_B5Nmn#% - - - - - - - 85000 - 90000 - 95000 - 100000 - 105000 - 80th Percentile Income ($) + + + + + + + 120000 + 125000 + 130000 + 135000 + 140000 + avg 1 @@ -30,7 +30,5 @@ 5 Social Distancing: Lowest (1) to Highest (5) - - - Annual Income + income80 diff --git a/docs/output_8_1.pdf b/docs/output_26_1.pdf similarity index 52% rename from docs/output_8_1.pdf rename to docs/output_26_1.pdf index 9489d275761e608b0cc108f012c6b02fab70b621..b6040947223257c7dc9fc9eea1481857f7d01700 100644 GIT binary patch delta 610 zcmV-o0-gPo3$_Zdw*h~Y%W~T=3`O_+3NG74RR9D?u**zei*DL@*^S4O*v|Mdjos$= z3zD|%WaQL#r5O>BaPQ$$(4`bPdFYf9OKGh^|2TDU;?s{W4m@}i?LxNeD0;keH>LyN zZ1v?niYH4KLzlDqr<30Q#s(~)_B7nR3)rr!Qw$guFm&Iz5;nA1d91#gT z2PD6Al#s9`3oGvXdP0>WLa(aHIV$7~+RIvYj*fsJ^}3dwr4s=;=xX*%7t|tEz$$&P z=_i{4wrLNUFj{MUr0N&Y98!>Meks{ER8=%P5wDnnYViRq>;xdlTrbA3eBFBlrDmaYhYI%$JLi*#=nE z@d|ZL3Qet-t4SbfRWlr`kEO4<6{mN_G63bW@l@@jrK!3T19&b98cLVQmU!Ze(v_Y6>+n wIFmmF@E9;TI5HqGFfcGMAZ`r`FfcGMFfcJNFfo%y1w#rjIXDU>B}Gq03iXp7#sB~S delta 705 zcmV;y0zUn=3X}`5w*h}uOLG%348HHL&=C$S*2A{UvBRSWm;st2mrP09l)Oj-{rkxF zX0z=ikXd$T){NAB=Jk79E)jphy>NkBT!-)>_HXX8t5?}4J4$xh6Io6~PsQh_*=jdkz~yYZ*xkYB z?fUn8wtaxjY@4I@12*f;uf=pXU#||-)pMUb`O_y)R_cE#?`?8Ai4S5nRoXZ!+-~{X`NM9yx}L9Y z-@wQ9dfZe(+Ga%Ev{3T19&Z(sK7%?z6 nHXtxCFfcG6ZVd`BFfcGMFflMWHIwNCLkclDG72RnMNdWwGXY42 diff --git a/docs/output_26_1.svg b/docs/output_26_1.svg new file mode 100644 index 0000000..614795d --- /dev/null +++ b/docs/output_26_1.svg @@ -0,0 +1,34 @@ + + + + + Stata Graph - Graph + + + + + + + + + + 32 + 34 + 36 + 38 + 40 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + schoollunch + diff --git a/docs/output_10_1.pdf b/docs/output_29_1.pdf similarity index 52% rename from docs/output_10_1.pdf rename to docs/output_29_1.pdf index a3bdea34e2b3e602be1891459917359e882a7a9a..e4ea937b8dcc370b00ddf260e9b1e6940e32db11 100644 GIT binary patch delta 594 zcmV-Y0^G%*GrEi^*Wb6#~p`>-)^=F_)N1}s=)Xa1WaB_BMR9IiIQdUQglHjM(tnbrZsOr zYQ98d+jTiA$j5()#mEVB&?6CH+iuM(gx9vcGap*>^_y)%g$Az>=P+@--7Y5_-@xOF_h9gj5J``(?Mdx9_|C zVX<3wyUjcMrGH%Q5B57SpuW%^di%NC{2VqpZ_eGrq@~WDn&^KKi8w5h>%RZd?e?qX ziQ^Bso?sq93T19&b98cLVQmU!Ze(v_Y6>+mGLu3C@E9;THZ~wIFfcGM gAZ`r`FfcGMFfcGVI5m?>1w#rjI5G+)B}Gq03Wv5A$N&HU delta 708 zcmV;#0z3Vt3YQD8w*h}u$#UB;488L!cx(?gRVbA1{sPCZ%YfPul)BEn_CCU!{96j7?o>hr~`vQAvk$RlR>T$*E3IdbIgfi--i) zF_T>y3P@Omg(~;HoDq1RCoP_f6T7Gm;A?S-IwB%Vdhp|nTyIS2%f#OexUD)Nbv zm5j~2pcFm#A9H^*GVi~rzeHq54<#`mCl8iMao9Vn9ymK%nm6&S9X*=QBlF{Itfo#Z z*mOA-IjY+kjBo`ylIM!X%h5zR!UauNpcR}dnuq8i^fhbQ(z|Sufiy!g=Ue011t~`F zKnwcUYQEY&ZKjhwLB)5uSF)l@y&}{M=yVv#iu0m6b>4reobP9M>J+G5|^_ghlq!?e1taFpzCN`;d)vEj819wM(tVU z7;CNdmbD81!RlR12dwUzWbO?L=s!z7MvDPzdEBnY^ZSMUG~e#-SL69=_Q8H%KTo$i zdj}3sKWKkioTj24noEN})BuePXq7NavgrRmq z-3yYtpx(Qxk;-+cH=ydFw0#i0<6%o|qruFF(P)=yTNvgNB=QZjXh)#rT&?WBsO`te z(_**PDxXX@)(-;z7rIlnD+*<9WOH + + + + Stata Graph - Graph + + + + + + + + + 10 + 11 + 12 + 13 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + foodinsec + diff --git a/docs/output_5_1.pdf b/docs/output_32_1.pdf similarity index 52% rename from docs/output_5_1.pdf rename to docs/output_32_1.pdf index 5c1a0367ccd36499f9c9b79f6a488b45c8c6d593..6d0f079e2ebaff764898d21c882b263d8b0d0761 100644 GIT binary patch delta 609 zcmV-n0-pV=3$+Tcw*h~YO>f&U42JLe6*_K*c_bxD)Q-iz4!dmiahD))vlLB(CBeF% zKT3-2q;NJjivf$2_41)8G0?;qe6-LAMbCNE0qt&O;KHYGpA1;A$kMo^zY}ZMnH$vs z_@wLeb7C(yCb%Xgu}>Fc{~PNtgX&vxcgA5;XO2U1tWZ?wsjYuYZ%txd6BUm(KN=B{ zzJ6y;(XIt=*t2-kHvcS|Wf& zdavatErHjvyF?gt^a-fMbBLBPIt>@2enJq*(a@KjsmiHGz~n9JnL+kKqGTyLPK67K z7}!6}O>5qM)O>%5$hLRofFWN3EQTf=Ip{%(ux&TyRfN~Jy*D3P^L4g3QG_bymy6K{ zu4uUeO#m)c^%ZPluZ7I&bjCiF{w-RL^d^UNKuH}htnC=Yg%Wh}Yc)o#qM;%3YiNk< z)T8Uyqs-L2kF72t+`44JfNzbT!%?%y`deAdV zK}7#opmx;lcDY_W*!S!GVezWN%_>3N|1xAa7!73Nxin delta 728 zcmV;}0w?{o3abmSw*h}uOOw+u48G@A_y`ArBwM!3ahKPDLt%T|OQ&>8mOhqjVTXT@ zY^Q13X`x$^nOJi4>62{Pf;t56f`M8nI8LJuXxe!NCw_hTP=Nu1EVYgL45W5eT`3O0 zMYT`wK%OouZ|fMvJe~CYZ(56qsC<>~uCiE{nQn-dl@yh9D64GWkmrAi?UGTKGqYcNY3`Ctek^Fhk<)o(ayp}gs{V9?Hn9Au+rm-b{0-zfE_Mt zdpIj9i2!Eky>6Q>0Xp3-8wtIRE&^qL7SVX&sP010Uol8pVog3JhQ!G=0h7~GLYERF zA1N9N($Ru4^wfVps;hDJ`n`rrL}q-K>jF~kU=}pqz+U%Xgw1$ay>R@}jPI*=9u4u6IUXz)iwh$q60yxQEccc}t#NC6^W`LfUGm zeblYaK1bIh^Dt`-0(>{4^~2A_WYeryJEG#JKbYy$AIyKGlv>!@q@(KCJGCkD*5v$n zx7o))HHp)(yY%r$P`P?L!XjTz7;Nx~5oO0o#}>lnop_{r6uZ>iaxbNy@1ss=8F^bVsYxGYl!IC4try_8rIsw)##>5f7q)f z2(d=eE}$&Tt{m~USx(yD=6cf3%&1-eYQE0e$6X}0e*mroxP1y`Ze(+Ga%Ev{3T19& zZ(?c+HXtw{Z(?c+HZ(Pp-~;d&F)%qfATTg6Ffbr)4GJ(YFfcGMF)=YXljsCP3NkS@ K3MC~)Peuw!6k>t^ diff --git a/docs/output_32_1.svg b/docs/output_32_1.svg new file mode 100644 index 0000000..519f1a6 --- /dev/null +++ b/docs/output_32_1.svg @@ -0,0 +1,34 @@ + + + + + Stata Graph - Graph + + + + + + + + + + 70 + 75 + 80 + 85 + 90 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + exercise + diff --git a/docs/output_36_1.pdf b/docs/output_36_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..2a5ed03c1e2595d09d80b8d6d8d67a03f8c72766 GIT binary patch literal 1339 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtRa5AQ7# z!RFG;DU4qVm!D}}zppmPO`2EF?8qm{$m45w3C=Z~rfO%aTwNS6PwLr<^Ww$(`&2y^ zd=k4eS1IF^kGX+mW0S-2Z7uh-%7lbj_^oUw2{QC32^~d{LH>F;w9zKyWE_D?s~>#NyQm1dp3UcWl_8kPB**Uc|?zNWNl?u@vt!jpsf_V;`4 z&78YkDEN(C%D+dy_7+XJ<))pzWoekf#uYWu)(IP@Gc$V~=Ma2!;K)j+Da&4z&z$=- zBvWyl{zj8V-?z`JbDsZUpLVn2;3?NtA=a)_?f9og&3PK~h2tUXRo$HjO%<8UPuRJ= z^gL|&Fn9OO_3~eS|Gk{~cgIcpck8}R|F);r>gD@mk`u&V1h2M#UvB@~bGq7csfnqp zCqM1ZFg>PG&&PSwF28?va`ivOEsAIJYojL^gEAx}fxxmRC_#bJn5i*X62t^0O<1yn zWN%Qy<DoPqlpl01OeB-jbrk%$(FBa6uQ8pARgVAlc6|FD+jIR5AgJgy77oRG?SD WIWMuO1nf6MLjz!0sp{(Q#svUgn%boR literal 0 HcmV?d00001 diff --git a/docs/output_36_1.svg b/docs/output_36_1.svg new file mode 100644 index 0000000..a6e7c81 --- /dev/null +++ b/docs/output_36_1.svg @@ -0,0 +1,36 @@ + + + + + Stata Graph - Graph + + + + + + + + + + + 1.9 + 2 + 2.1 + 2.2 + 2.3 + 2.4 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + overcrowding + diff --git a/docs/output_39_1.pdf b/docs/output_39_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..63df728926b8a486b1c7d0e8c15272e3082db199 GIT binary patch literal 1325 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPt#@go7JD{6zkoy4Wv1|%hGf!|Gt`-t9fCmi1Vil znoAW@w=n*FeClAQf4rT=hX;XM9_Q?5e*N)Q>qPy6PaW~d?c<-)e5)AhR7dbYX8%XN2nJYZJ-*}IrSXm)XacYBQg3IT;y z$7_2wmn_m*ysql`3yz~{)v=$?T~eK9^J-Q3l4MhF(G6bj-!PrvQ@nAoSgpsbjVI)> z;UfY4)(Pz<6I>^|A7waty7Nf#vY5U}F3Q!M8|U&%Bqh`ri*tYfwfCvFOI=~`dI>Md z1XCVy);HHekLsJfV=YwI|NH3N-sm0Y(@z&`uF&HDujj_Op!%zfg`!% z{A=M4|Bqg4!c{exQ;VCrwx)*knZCK&e^p{@XWKE2^$QKVF6(WG4~|mx(22e9rtaj= zlV4|V?Rx3a{8VDcvQ}ZwoBtWU97sFQuI8fQa%G}tFni+EwU6~y8_)0ZajZZ7N#q)z z3g3-qf7s17yB&^F<-KCysJm9G>eaoO{;GnnXC%K`!zH(EVW;(}i|-CZXPP8sX(uQ^IBq23D|Fj1{RiFs;aL3 GZd?F8HrZML literal 0 HcmV?d00001 diff --git a/docs/output_39_1.svg b/docs/output_39_1.svg new file mode 100644 index 0000000..53cc351 --- /dev/null +++ b/docs/output_39_1.svg @@ -0,0 +1,34 @@ + + + + + Stata Graph - Graph + + + + + + + + + + 23.5 + 24 + 24.5 + 25 + 25.5 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + youth + diff --git a/docs/output_42_1.pdf b/docs/output_42_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..fca33d674aaf9dec42e7fb710bd2f5c39ecb5931 GIT binary patch literal 1340 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPt_wsqju?)Bg6>{pU|pEseYHgo4klmd+FW~X&Es3KGx;1g9i9Gu=H6SoPlTR(C9n7FpZT5E)?H(S|7_NHkqe?UyZJ^Qlq%rr)qaWr<^%+S-)?M3&>P^ z6S~#dwN<8Kg6Tb#xr+`gSiM)%YLN(A>|@_cs(h6fuHI9YT^+fqVe6M`a?&@ib=vA} z+uO0ti*40s!=#(nE|m$LEWN&zE84V@Nl-J+PsFP=$V0^BlcYG$#m)6QSGQLGzW7R2 zY5$Iu?98V=B;+nwE^#2u)%3^FdwZQ$zStUnXBR*B_tH%6RxRuDG-7$|zk589(t-5~rD#7U$(^Nn6gRblLAvl7vj5xtk*UG)h3_OFre#*NR7oPu0}r3sgAxZQYo zVd3dpGyPrFE-sIdJAA9{h}><9|BYwYT-;#hX&dUc;^`J|_uRQZL%15BaGIW(`CuQf zZ%km{Hy#7Wb;5D-VYi>2+kgMw{AuN@-`;&T_kU4E#Qphy$~hX!9P6!qoWFhc{JS&_ z^W%O=+rp0TRGAm@)_#$~_1*HXAHO?4;i3Km?tU{+j)bHTSl$GsC{Q9ZH8z0>fzl={ z-9hp5S_CfXg7Wi$MH3|ddFG|%D}ah7V3`n{S(OU( Y3OMT}7L|bgW@u<&!lkO}>hHz{06RtJ0ssI2 literal 0 HcmV?d00001 diff --git a/docs/output_42_1.svg b/docs/output_42_1.svg new file mode 100644 index 0000000..70368d1 --- /dev/null +++ b/docs/output_42_1.svg @@ -0,0 +1,36 @@ + + + + + Stata Graph - Graph + + + + + + + + + + + 14.2 + 14.4 + 14.6 + 14.8 + 15 + 15.2 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + elderly + diff --git a/docs/output_46_1.pdf b/docs/output_46_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..8f7a5ef3705be6fb6e15b1a515d13fb2c5e282a6 GIT binary patch literal 1324 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtUvpR| zB0Z{Bpa)I71g^XQ46 zZ7b9d$-K!)y*J^`jhl;q8W_jxvnjXD;JS5CG-DxyXzzOW?5}-VvyQzm{F0wD^JVU` z5{Fw3Uu3>yxs+|3aISv4`ZX@8Hsk$m56}3$t&^X8{A&9Q^^Lb`Pdq*Ob*a~)wj0@f z_oi+sP+DoE?|v{VUVlbRSES(2C7!{(AC~?|p879RJ#KxMp5eYJrNWzjGPT5gdl94W z-5?VZ_BJiOST_0wx4Xi?DPC${QfbJg@0tfpdzpFZ3i{!hd5(F-nJ@vQOjuErn#N_IV8LYo z2MT7UrpBfUX}oYTLtt_T0fjt-m>DqpqKO$A0gD85F$MSjR z#TiVjq$n{nC$$J%vIXVm1B)U^e)G&r%U1vuMZhv3II}7h=oN5QODrk@`_0h6!h%ax J)z#mP3jo9{+wuSa literal 0 HcmV?d00001 diff --git a/docs/output_46_1.svg b/docs/output_46_1.svg new file mode 100644 index 0000000..e5226bc --- /dev/null +++ b/docs/output_46_1.svg @@ -0,0 +1,32 @@ + + + + + Stata Graph - Graph + + + + + + + + + 29 + 30 + 31 + 32 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + segregation_wnw + diff --git a/docs/output_50_1.pdf b/docs/output_50_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..79b1789b846ea78149410bc1b21437f0218b7f49 GIT binary patch literal 1418 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtNMA@dZWl=AtrKnAlsr{7_Ic;I*M&66g zlQq~*F6MnY0l0939hLQIljsQTw;rk%-9;{S$pEf8E&_a2mT(8?Y(olTufi^=fy2K zhEI$RC`rUl3EQ(_X8*Kzkv`}DzBsp6cBg-Ov!G*=>uc5*W=XE<6;`u4lG~4QU*O!~ zd|AM@<=@%pZ(o#lrzn`V=5Cid?Y6=%|G>HHRj<~Ye~zDgw(NX>>=)6HZ`<@2EV@$r zCRgGlyQXW{(|b?;IQ&k{UUJOq$|K$wrz6UaQ%}9-SY2b8_~b^Kn@_c$)wGTPvHNR{ zY)&1y962%hkaLus(`^4qU$)FGveac%GmKhhn{9C{%0K6L)YSh8xvDQ2SH}dM-Fw(! zO|O&E>v;=~-v3&&`}a1f^WmGr|2_Uyx9RRrX#t1d3+}J|+y3*-w)so|M&VynO&UE zx6^-_Ike{1fbusay}@!iD1CyGteK$^ObC?HVW}8WT7dEam%dYCaVj`R=zFB*l%f)0|KJ@eA?6+i_Yu!ITDtV#uX1)Px+i%P(LGc>d?;!;(0^>^a} E0Aq& - - - - - - - - 8 - 9 - 10 - 11 - 12 - 13 - % Uninsured + + + + + + + 78 + 79 + 80 + 81 + 82 + % of Drivers 1 @@ -31,8 +29,8 @@ 4 5 - Social Distancing: Lowest (1) to Highest (5) - - - Uninsured Adults Under 65 + Distancing: Lowest (1) to Highest (5) + + + Commuting Alone by Vehicle diff --git a/docs/output_52_1.pdf b/docs/output_52_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..59675492742d818bf4e0397e2ec1fb02bcf9916d GIT binary patch literal 1374 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtzi|D{)Q>3<|llam}*ndfm5Nmb&M!uX0iO&^YnWtw0}%RYl8>Ij_-QFrlTe zA@r?;jH8G|L8vv?^dgyuUEf%adoG>Az+JMmRmE2=gV9oWn^hZ6$REQ+l^L6Vb!$FV z+Z&M^c}apjy(*P{o_|W- z#(m7TkKOq_`1{ukohTDlnSZiA2ad4ns5eY*;mSMy!0?m*xic4MR;Db-sTFRNX1OS5 zK0p51>Gzv^=f}jxyxAN6;lhG%l2TWu>|);AyESKm*$UsO)sr7LzcD^HQOe6P`EAfK zX2+@1UUICiu}YjWp;z$oyvcl`i%Qbwr-u5ayB~`xoWk|6MNnxOl$EY?;)%EiO6idtHR2qdi~kzQfu+eT@r)(>$YXPre@75MXdk z+Hrc2ns#Ks?z?n0#s08dcLqlT= LE>%@me>W}whhgvu literal 0 HcmV?d00001 diff --git a/docs/output_52_1.svg b/docs/output_52_1.svg new file mode 100644 index 0000000..62befcb --- /dev/null +++ b/docs/output_52_1.svg @@ -0,0 +1,34 @@ + + + + + Stata Graph - Graph + + + + + + + + + 3 + 4 + 5 + 6 + Meidan RUCC + + + 1 + + 2 + + 3 + + 4 + + 5 + Distancing: Lowest (1) to Highest (5) + + + Urban-Rural + diff --git a/docs/output_53_1.pdf b/docs/output_53_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..f983d45d16eace40b012f8dac6bae17466246cff GIT binary patch literal 1425 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtEA!+o?VX>-Vy5i?qTHZ|I1R( zxO~knS>3jAt@hysI&NKVR`X}SN?n{}b@9c;r`L6qge=Z`tjpF=N)&zPQ~BwuBpBhKl4pv|F z10ZKKTbsd-NPaQcOIWhUr@+~zDX67?`wlo)5=`yMrMlG{l zZGR+6sxoEmq52JMp1)c`{WL#aT(CVZQ&h;ZCt&xpn!R^x_wV19Rrg-+SNqel=j$98 z-Ztl5->v_8w*2EoPmV@jS?hn~=evm>liWXF&zb)E<%h79>_ObV7yEd&+&HtoW8=M` z7c({P2b4{l)izC8;7!9qi9-eMH-Db|{5tCd({07yD%^*IY+pS$eUj{xyQN{$#*QD8 zI=b&kRo#33+u=XM|2+$zfU-Fx*}*bAD4Bv%u9=}FObC?VVTl=1Xn?W`JC^0i9wFq2%2Ic1i%R5M>_RLGmR{)iFz=9??vnmzn6>xq|EGhx}&Ct-o Ml1o+9)!&T^0NxD~>i_@% literal 0 HcmV?d00001 diff --git a/docs/output_5_1.svg b/docs/output_53_1.svg similarity index 62% rename from docs/output_5_1.svg rename to docs/output_53_1.svg index 1c77e8f..d22730e 100644 --- a/docs/output_5_1.svg +++ b/docs/output_53_1.svg @@ -7,19 +7,17 @@ - - - - - - - 45 - 50 - 55 - 60 - 65 - 70 - PCP per 100k Population + + + + + + 29 + 30 + 31 + 32 + 33 + % of Solo Commuters Driving 30+ mins 1 @@ -31,8 +29,8 @@ 4 5 - Social Distancing: Lowest (1) to Highest (5) - - - Primary Care Providers + Distancing: Lowest (1) to Highest (5) + + + Long Solo Commute diff --git a/docs/output_8_1.svg b/docs/output_8_1.svg deleted file mode 100644 index 6932167..0000000 --- a/docs/output_8_1.svg +++ /dev/null @@ -1,36 +0,0 @@ - - - - - Stata Graph - Graph - - - - - - - - - - 120 - 140 - 160 - 180 - 200 - Mental Health Proviers per 100k population - - - 1 - - 2 - - 3 - - 4 - - 5 - Social Distancing: Lowest (1) to Highest (5) - - - Mental Health Providers - diff --git a/docs/output_9_1.pdf b/docs/output_9_1.pdf new file mode 100644 index 0000000000000000000000000000000000000000..5ae6c987d79b43f2fcbb75f9e762d78b8b99b2aa GIT binary patch literal 1350 zcmY!laB1BLvgEG`=xF8z?of>Z^4=fsl4ocwey{eZ;u)M5oA zpn@PSJ3Fq_ycCc^5Fb?oSRI$XcVOB&`8O`6;Ez zsYMDJ9*ISz3T{QIsR}@sDfnb26(tr`Dj4aR=ozG>mT4mC2N{iNHJ83`YD#9JQ+@@| zb^{EO#R{fCH9=hZL8-<0rA5g=FM|9IbWd_{Y6;K?eULNs zLsBbB6!blFfnneTrJbR)C&+hh`9RNsG`blon1c-j1`!YeL#+sC3r4VkEQPtiqdU5gT4CeYj-5aQ1!^^Ht0Rp4XrM@7#N1)&XCyr1|>C{+UlR zY<{(3hxOqbSJaPdZVcDZ`f}yk((P)MvBIMLX8%oL)@`#w*tM{&G?}AB7 z5~D*4kFA)nx5Y1PZ=lsdiR##o$6j@-;gLtO_r#+$JMBwesSr6o7|m~mNHD_7MpTmVg1b6 zGb_IbURG7I-mU$e>C@eVNl$8OFZ@!sU2^lEE9nK2syBE4E%e^vJ)jNq0BR z3lD4O%vig&ebu}JYDw*`vg#YYKA5@cS5K|h-{>teFLU`$I(W*e$L(iyw3xEs^ZnI! zS)!$}Guqd2Zw}_Zzb;Ybb$21um0rE z`Yh9my#>r`Be|4rO$gZivuu0BdGl|7f4wxQk60rgcKzGwca=X&HrzkR{Ho#G zhRyS@ub%(bdA>Z~*ThR}`NDj^SbHeN%>GgM@kg?~z0PledIrOby*EL56OvY7xfGPP zK*`P2!~`Y;N~N$A2+8T7G|Z*%lvtb!PR;rrsX1k-C7H>IT>8PKNhRRY0Ho58OW!pQ zm?|^#(iQZ>GxHqtiZfvXNZGTZC^e1CK*55`01gz)OihhV71DU&Vurws00Ihm2r)BY z#zqq}Gy)bD=wcQ?FQADTn`4@1X=#D3&d|URNpDF}VrEWi5xC3?%FhQDQjnbJnU|KY j04k(_r9*ILRVvUc;7piUR08&!p`oD(m#V6(zZ(|-yCd&E literal 0 HcmV?d00001 diff --git a/docs/output_9_1.svg b/docs/output_9_1.svg new file mode 100644 index 0000000..c442ba5 --- /dev/null +++ b/docs/output_9_1.svg @@ -0,0 +1,36 @@ + + + + + Stata Graph - Graph + + + + + + + + + + + 50 + 55 + 60 + 65 + 70 + 75 + avg + + + 1 + + 2 + + 3 + + 4 + + 5 + Social Distancing: Lowest (1) to Highest (5) + pcp_rate + diff --git a/google_x_valid.dta b/google_x_valid.dta index 6f7dbc8dfc3136f8917a6c8834db3af7e0a09c02..a82eefe5e83f3b00b2f5275ce660d113283b35dd 100644 GIT binary patch delta 3367 zcmeIy%WD%s90%~(*tluZBpZ|Xs%c|wsfyp~bg0erIPU2HIlN^q7TDzs&Bu_P3Lczj|mI zhvBR_8i_|D*M`P)(w{y_$QKJSuCF4$iTp0|hsd8Ie~J7p@{h<#k$I7%n$N3e4av? zhxL4^7YW2tWPp%T$K|ITIIHzBvMoHvCI2&UruC=Y>}+1S#JyVo%}I}`ezqOG2Op@4 zpMk<(?MWbZTQ4-pP0>!?`V?jNWhN%m_U+?=ci8W*@;(k{hzkm1j*!eu$yXe>W;*HV1@-(ZlzP=-I zC(DuECYLexkj;+cN#r~vcdv%E#dD|Ep&*bX~jC+vc5*bP0f2lm3g(VTaG_tn3k_!TJt delta 4764 zcmeH|%S%*Y7{=eZ9A})dQJa~{)Ee*i(NV|L%1qNN%}uz;D6LGwC{bI~=7e?=!V8D~ zgrH=C3e-ikh-eWlTto>iN^p}!5e7vN+H=nL9#LFajm~W5gCEa)xA%R&=X~eP{L9D2 zc`$NK@|9KkD@F&W<(~Gw4l>efyF{}l~!)OYesd!O>bkbErDN?OZe&1TJE z&1KDF&1WrO^|BVS7P0zB$#ZzxMR3}{EaziGk>x>LD6)Kw4@H(jXDG6CyF6_h{q&Vq zDapQQc}`}sQZzY~eyK%%`0=ulkowz{P-4A0^TeQzw;h?#=IOnd@1~53Ug{)TpN+D5 z>c2?oE&9)tYiyJ^;FN2mT#?a6OU5cEEFQ&Vy&<>S;R)ITe;f=J(Lc`P7o$YKCPwH5 zN$3QEsm)>%`qa$I^HritT|DpSI77T{Qq0RlVXXl9!&-rsKa3S{{;*Xb6JLZdRiK_f zoE3<-Vp18fd*x9Y|2a103%JCqK{LZmUlvfHt_B{j9unhYucogost4e8>CfJ2{o9Q# zW_4}yXPZgYN3%b)K4p%L{Kk4c4)x#H`kYUu=1*dS9w0yQ){2d#CW97`KNU-b>465D zNt>s)qgvoq1@Hk3Jg@JwHK+xnvQ-=SWCqry4ID@>n7>6t!B{A(UI*!6Fc!+1XTfBO*i})ez5cgv) zKH|r6KXRQBKbHH^<%;;R%#VS9`)0sx07hT}W*~tmU;$PT4Q#*;V!$Tg0I?tr#DmSi z30y#pxdm(m+du-?4iZ5UNCrE=PT&D4AQhy6bdUiuK^DjcIUpD0fqYN^yr2*i0Usy^ zC7=}eK^fQu%0UIF1iQf=uoqN;eV`ib2Q{D;8~}Bo9vlP>;1D\n", "` ^\n", - " levels1 | 48.03302 2.404615 77.34 0.000 43.5439 52.98495\n", - " levels2 | 50.28804 2.545734 77.39 0.000 45.53803 55.53351\n", - " levels3 | 56.80765 2.835276 80.94 0.000 51.51376 62.64558\n", - " levels4 | 61.25371 3.008065 83.79 0.000 55.63286 67.44247\n", - " levels5 | 71.49253 3.509299 86.98 0.000 64.93493 78.71236\n", + " levels1 | 49.92198 2.325957 83.93 0.000 45.56514 54.69541\n", + " levels2 | 54.89074 2.441119 90.06 0.000 50.30882 59.88996\n", + " levels3 | 57.89873 2.709556 86.73 0.000 52.82437 63.46053\n", + " levels4 | 61.49219 2.753973 91.97 0.000 56.32462 67.13387\n", + " levels5 | 73.62823 3.207945 98.67 0.000 67.60175 80.19195\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | .6718608 .027037 -9.88 0.000 .6209052 .7269981`" + "` levels1 | .6780277 .0239497 -11.00 0.000 .6326751 .7266314`" ] }, { @@ -775,14 +749,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "48.8\n", - "LL: 61.1\n", + "47.5\n", + "LL: 58.1\n", "UL: 37.6\n" ] } ], "source": [ - "invert .6718608 .6209052 .7269981" + "invert .6780277 .6326751 .7266314" ] }, { @@ -790,7 +764,7 @@ "metadata": {}, "source": [ "\n", - "The counties showing the smallest declines in mobility had 48 primary care providers per 100,000, whereas the most social distancing counties had 71 per 100,000 after adjusting for rurality and stay-at-home orders, a 49% (95% CI: 38%, 61%) difference. \n", + "The counties showing the smallest declines in mobility had 50 primary care providers per 100,000, whereas the most social distancing counties had 74 per 100,000 after adjusting for rurality and stay-at-home orders, a 47% (95% CI: 38%, 58%) difference. \n", "\n", "---" ] @@ -812,48 +786,55 @@ "name": "stdout", "output_type": "stream", "text": [ - "----- RURALITY-ADJUSTED POISSON MODEL -----\n", - "note: uninsured_p has noninteger values\n", + "----- RURALITY-ADJUSTED NEGBIN MODEL -----\n", + "note: you are responsible for interpretation of non-count dep. variable\n", "\n", - "Iteration 0: log likelihood = -7726.4116 \n", - "Iteration 1: log likelihood = -7718.2877 \n", - "Iteration 2: log likelihood = -7718.2864 \n", - "Iteration 3: log likelihood = -7718.2864 \n", + "Fitting Poisson model:\n", "\n", - "Generalized linear models Number of obs = 2,633\n", - "Optimization : ML Residual df = 2,619\n", - " Scale parameter = 1\n", - "Deviance = 4539.377304 (1/df) Deviance = 1.733248\n", - "Pearson = 4761.738841 (1/df) Pearson = 1.818152\n", + "Iteration 0: log pseudolikelihood = -25368.195 \n", + "Iteration 1: log pseudolikelihood = -7847.5733 \n", + "Iteration 2: log pseudolikelihood = -7718.5574 \n", + "Iteration 3: log pseudolikelihood = -7718.2864 \n", + "Iteration 4: log pseudolikelihood = -7718.2864 \n", "\n", - "Variance function: V(u) = u [Poisson]\n", - "Link function : g(u) = ln(u) [Log]\n", + "Fitting full model:\n", "\n", - " AIC = 5.873366\n", - "Log likelihood = -7718.286439 BIC = -16087.55\n", + "Iteration 0: log pseudolikelihood = -8982.7909 \n", + "Iteration 1: log pseudolikelihood = -7464.8863 \n", + "Iteration 2: log pseudolikelihood = -7462.9819 \n", + "Iteration 3: log pseudolikelihood = -7462.981 \n", + "Iteration 4: log pseudolikelihood = -7462.981 \n", + "\n", + "Negative binomial regression Number of obs = 2,633\n", + "Dispersion = mean Wald chi2(14) = 94084.62\n", + "Log pseudolikelihood = -7462.981 Prob > chi2 = 0.0000\n", "\n", "------------------------------------------------------------------------------\n", - " | OIM\n", + " | Robust\n", " uninsured_p | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 10.71643 .3930656 64.66 0.000 9.973072 11.51519\n", - " levels2 | 9.441651 .3438872 61.64 0.000 8.79114 10.1403\n", - " levels3 | 8.465887 .3104513 58.25 0.000 7.878765 9.09676\n", - " levels4 | 7.986939 .2926295 56.71 0.000 7.433505 8.581577\n", - " levels5 | 7.028877 .2568552 53.36 0.000 6.543055 7.55077\n", - " homeorder | 1.159956 .0292505 5.88 0.000 1.10402 1.218726\n", + " levels1 | 10.74092 .3917303 65.09 0.000 9.999938 11.5368\n", + " levels2 | 9.464582 .3546025 59.99 0.000 8.794479 10.18574\n", + " levels3 | 8.480079 .3192442 56.78 0.000 7.876899 9.129449\n", + " levels4 | 8.000203 .3006194 55.34 0.000 7.432174 8.611645\n", + " levels5 | 7.044127 .2647529 51.94 0.000 6.543873 7.582624\n", + " homeorder | 1.156171 .0313348 5.35 0.000 1.096358 1.219246\n", " |\n", " rucc |\n", - " 2 | 1.036294 .0321618 1.15 0.251 .9751373 1.101287\n", - " 3 | 1.03009 .0327635 0.93 0.351 .967835 1.096349\n", - " 4 | 1.052689 .0378063 1.43 0.153 .9811374 1.129458\n", - " 5 | 1.059101 .0510379 1.19 0.233 .963647 1.164009\n", - " 6 | 1.166658 .0323102 5.57 0.000 1.105019 1.231735\n", - " 7 | 1.123239 .0344586 3.79 0.000 1.057692 1.192849\n", - " 8 | 1.230292 .0503446 5.06 0.000 1.135472 1.333031\n", - " 9 | 1.158536 .0502063 3.40 0.001 1.064197 1.261239\n", + " 2 | 1.036269 .0311154 1.19 0.235 .9770437 1.099084\n", + " 3 | 1.02913 .0307182 0.96 0.336 .9706509 1.091133\n", + " 4 | 1.050335 .0360446 1.43 0.152 .9820124 1.123411\n", + " 5 | 1.05925 .0478817 1.27 0.203 .9694404 1.157379\n", + " 6 | 1.170248 .0327939 5.61 0.000 1.107706 1.236321\n", + " 7 | 1.123108 .0352113 3.70 0.000 1.056173 1.194286\n", + " 8 | 1.237528 .0477472 5.52 0.000 1.147396 1.33474\n", + " 9 | 1.1618 .0488127 3.57 0.000 1.069962 1.261521\n", + "-------------+----------------------------------------------------------------\n", + " /lnalpha | -2.682096 .0589782 -2.797691 -2.566501\n", + "-------------+----------------------------------------------------------------\n", + " alpha | .0684196 .0040353 .0609506 .0768038\n", "------------------------------------------------------------------------------\n", - "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n", + "Note: Estimates are transformed only in the first equation.\n", "Compare to tabular data:\n", "\n", "----------------------------------------------------------\n", @@ -870,56 +851,68 @@ " 5 | 535 8.6297361 .1775762\n", "----------------------------------------------------------\n", "----- PERCENT DIFFERENCE MODEL -----\n", + "note: you are responsible for interpretation of non-count dep. variable\n", "note: levels5 omitted because of collinearity\n", - "note: uninsured_p has noninteger values\n", "\n", - "Iteration 0: log likelihood = -7726.4116 \n", - "Iteration 1: log likelihood = -7718.2877 \n", - "Iteration 2: log likelihood = -7718.2864 \n", - "Iteration 3: log likelihood = -7718.2864 \n", + "Fitting Poisson model:\n", "\n", - "Generalized linear models Number of obs = 2,633\n", - "Optimization : ML Residual df = 2,619\n", - " Scale parameter = 1\n", - "Deviance = 4539.377304 (1/df) Deviance = 1.733248\n", - "Pearson = 4761.738841 (1/df) Pearson = 1.818152\n", + "Iteration 0: log pseudolikelihood = -7718.2865 \n", + "Iteration 1: log pseudolikelihood = -7718.2864 \n", "\n", - "Variance function: V(u) = u [Poisson]\n", - "Link function : g(u) = ln(u) [Log]\n", + "Fitting constant-only model:\n", "\n", - " AIC = 5.873366\n", - "Log likelihood = -7718.286439 BIC = -16087.55\n", + "Iteration 0: log pseudolikelihood = -9016.9274 \n", + "Iteration 1: log pseudolikelihood = -7680.503 \n", + "Iteration 2: log pseudolikelihood = -7675.9149 \n", + "Iteration 3: log pseudolikelihood = -7675.8619 \n", + "Iteration 4: log pseudolikelihood = -7675.8619 \n", + "\n", + "Fitting full model:\n", + "\n", + "Iteration 0: log pseudolikelihood = -7480.4687 \n", + "Iteration 1: log pseudolikelihood = -7463.1407 \n", + "Iteration 2: log pseudolikelihood = -7462.981 \n", + "Iteration 3: log pseudolikelihood = -7462.981 \n", + "\n", + "Negative binomial regression Number of obs = 2,633\n", + " Wald chi2(13) = 482.26\n", + "Dispersion = mean Prob > chi2 = 0.0000\n", + "Log pseudolikelihood = -7462.981 Pseudo R2 = 0.0277\n", "\n", "------------------------------------------------------------------------------\n", - " | OIM\n", + " | Robust\n", " uninsured_p | IRR Std. Err. z P>|z| [95% Conf. Interval]\n", "-------------+----------------------------------------------------------------\n", - " levels1 | 1.524629 .0404712 15.89 0.000 1.447335 1.60605\n", - " levels2 | 1.343266 .0359434 11.03 0.000 1.274634 1.415594\n", - " levels3 | 1.204444 .0330621 6.78 0.000 1.141356 1.271019\n", - " levels4 | 1.136304 .0310814 4.67 0.000 1.076989 1.198885\n", + " levels1 | 1.524804 .0395603 16.26 0.000 1.449206 1.604347\n", + " levels2 | 1.343613 .0368115 10.78 0.000 1.273367 1.417735\n", + " levels3 | 1.203851 .0339756 6.57 0.000 1.139068 1.272318\n", + " levels4 | 1.135727 .0316513 4.57 0.000 1.075355 1.199488\n", " levels5 | 1 (omitted)\n", - " homeorder | 1.159956 .0292505 5.88 0.000 1.10402 1.218726\n", + " homeorder | 1.156171 .0313348 5.35 0.000 1.096358 1.219246\n", " |\n", " rucc |\n", - " 2 | 1.036294 .0321618 1.15 0.251 .9751373 1.101287\n", - " 3 | 1.03009 .0327635 0.93 0.351 .967835 1.096349\n", - " 4 | 1.052689 .0378063 1.43 0.153 .9811374 1.129458\n", - " 5 | 1.059101 .0510379 1.19 0.233 .963647 1.164009\n", - " 6 | 1.166658 .0323102 5.57 0.000 1.105019 1.231735\n", - " 7 | 1.123239 .0344586 3.79 0.000 1.057692 1.192849\n", - " 8 | 1.230292 .0503446 5.06 0.000 1.135472 1.333031\n", - " 9 | 1.158536 .0502063 3.40 0.001 1.064197 1.261239\n", + " 2 | 1.036269 .0311154 1.19 0.235 .9770437 1.099084\n", + " 3 | 1.02913 .0307182 0.96 0.336 .9706509 1.091133\n", + " 4 | 1.050335 .0360446 1.43 0.152 .9820124 1.123411\n", + " 5 | 1.05925 .0478817 1.27 0.203 .9694404 1.157379\n", + " 6 | 1.170248 .0327939 5.61 0.000 1.107706 1.236321\n", + " 7 | 1.123108 .0352113 3.70 0.000 1.056173 1.194286\n", + " 8 | 1.237528 .0477472 5.52 0.000 1.147396 1.33474\n", + " 9 | 1.1618 .0488127 3.57 0.000 1.069962 1.261521\n", " |\n", - " _cons | 7.028877 .2568552 53.36 0.000 6.543055 7.55077\n", + " _cons | 7.044127 .2647529 51.94 0.000 6.543873 7.582624\n", + "-------------+----------------------------------------------------------------\n", + " /lnalpha | -2.682096 .0589782 -2.797691 -2.566501\n", + "-------------+----------------------------------------------------------------\n", + " alpha | .0684196 .0040353 .0609506 .0768038\n", "------------------------------------------------------------------------------\n", - "Note: _cons estimates baseline incidence rate.\n", - "(Standard errors scaled using square root of Pearson X2-based dispersion.)\n" + "Note: Estimates are transformed only in the first equation.\n", + "Note: _cons estimates baseline incidence rate.\n" ] }, { "data": { - "application/pdf": "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", + "application/pdf": "JVBERi0xLjMKJbe+raoKMSAwIG9iago8PAovVHlwZSAvQ2F0YWxvZwovUGFnZXMgMiAwIFIKPj4KZW5kb2JqCjIgMCBvYmoKPDwKL1R5cGUgL1BhZ2VzCi9LaWRzIFsgNCAwIFIgXQovQ291bnQgMQo+PgplbmRvYmoKMyAwIG9iago8PAovUHJvZHVjZXIgKEhhcnUgRnJlZSBQREYgTGlicmFyeSAyLjQuMGRldikKPj4KZW5kb2JqCjQgMCBvYmoKPDwKL1R5cGUgL1BhZ2UKL01lZGlhQm94IFsgMCAwIDE1OC4zOTk5OSAxNTguMzk5OTkgXQovQ29udGVudHMgNSAwIFIKL1Jlc291cmNlcyA8PAovUHJvY1NldCBbIC9QREYgL1RleHQgL0ltYWdlQiAvSW1hZ2VDIC9JbWFnZUkgXQovRm9udCA8PAovRjEgNyAwIFIKPj4KPj4KL1BhcmVudCAyIDAgUgo+PgplbmRvYmoKNSAwIG9iago8PAovTGVuZ3RoIDYgMCBSCi9GaWx0ZXIgWyAvRmxhdGVEZWNvZGUgXQo+PgpzdHJlYW0NCnicpVRNk9MwDL37V/gIB4IkW7bMcYevAxfYHJlhOqVbstMPaLvs30e2m7S7k0Jbkhlbcvzek2Qr0MQYfLTQCEXPedYJ7WZu4MSnLx8MWLDI0rikz5G1mZkbxQH4pPvze2RPlya4JiWxjye5/y3rMAvVUeXujMOmruyn5FhFUUWPbd16axLrnqjyaPOrjIgE1jskuzQUkbA6C908xHoChQDRD7DqnYVjcXzAFe8cHCHrao+r3lm46DgdcMV7gnOilghggd7rAZAWwWvhXSIMOvsEVPgcKgMHB8oXo35DyePCoC+pUeJcBAImX0rishcSpVxXLLqgEoP2L3PTmtfvNV4isu1dOWLI1wCLkQPmoEcpwrZdmhfxpW3vzbvWfD4bm/Jtyli5HIvsQQo4XQ4mxAgFjHAFOiRHFY3H6GPo39gy06tM6YCSOl5YuzGIC4V08nt+YK2goR/yMPRDduq59Xtyh3gO+fJXu99yfnB9qilSQ+I1Ru9qriNBlZu2F6z29YLRxZAVUQ6aNFYIX5LfZ+mfVOJyVXRR8LmsG5HNncO97N65XpYAhVVW79Gg6sdUa4fuVavzH6qBJT1T5Wtu8EDoQc8sBRGrDZUK4e162k0W9m233U1W0241f2M/rR9n2539CgyoA9rd2n7s5j/6Rc6LY32YNNXRUxP9i5GIwyL5sOpW24fN7Pu3nweWP/bbd5oKZW5kc3RyZWFtCmVuZG9iago2IDAgb2JqCjUyOQplbmRvYmoKNyAwIG9iago8PAovVHlwZSAvRm9udAovQmFzZUZvbnQgL0hlbHZldGljYQovU3VidHlwZSAvVHlwZTEKL0VuY29kaW5nIC9XaW5BbnNpRW5jb2RpbmcKPj4KZW5kb2JqCnhyZWYKMCA4CjAwMDAwMDAwMDAgNjU1MzUgZg0KMDAwMDAwMDAxNSAwMDAwMCBuDQowMDAwMDAwMDY0IDAwMDAwIG4NCjAwMDAwMDAxMjMgMDAwMDAgbg0KMDAwMDAwMDE4NyAwMDAwMCBuDQowMDAwMDAwMzc1IDAwMDAwIG4NCjAwMDAwMDA5ODMgMDAwMDAgbg0KMDAwMDAwMTAwMiAwMDAwMCBuDQp0cmFpbGVyCjw8Ci9Sb290IDEgMCBSCi9JbmZvIDMgMCBSCi9TaXplIDgKPj4Kc3RhcnR4cmVmCjEwOTkKJSVFT0YK", "image/svg+xml": [ "\n", "\n", @@ -934,7 +927,7 @@ "\t\n", "\t\n", "\t\n", - "\t\n", + "\t\n", "\t7\n", "\t8\n", "\t9\n", @@ -970,7 +963,7 @@ "\t<line x1="55.44" y1="800.17" x2="1342.53" y2="800.17" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="514.95" x2="1342.53" y2="514.95" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", "\t<line x1="55.44" y1="229.73" x2="1342.53" y2="229.73" style="stroke:#FFFFFF;stroke-width:4.75"/>\n", - "\t<path d=" M64.90 315.32 L381.95 686.07 L698.96 942.78 L1015.97 1085.39 L1333.03 1370.61" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", + "\t<path d=" M64.90 315.32 L381.95 657.56 L698.96 942.78 L1015.97 1085.39 L1333.03 1370.61" stroke-linejoin="round" style="fill:none;stroke:#3E647D;stroke-width:19.01"/>\n", "\t<text x="1364.51" y="1391.74" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">7</text>\n", "\t<text x="1364.51" y="1106.52" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">8</text>\n", "\t<text x="1364.51" y="821.35" style="font-family:'Helvetica';font-size:60.49px;fill:#000000">9</text>\n", @@ -1012,7 +1005,7 @@ ], "source": [ "// Comparing percent uninsured to social distancing\n", - "modelpoisson uninsured_p " + "modelrun uninsured_p " ] }, { @@ -1024,17 +1017,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 11.0404 .4479299 59.19 0.000 10.19647 11.95417\n", - " levels2 | 10.5449 .4268925 58.19 0.000 9.74054 11.41568\n", - " levels3 | 9.052178 .3649673 54.64 0.000 8.364388 9.796524\n", - " levels4 | 8.318344 .3368231 52.32 0.000 7.683699 9.005408\n", - " levels5 | 7.144656 .291952 48.12 0.000 6.594755 7.74041\n", + " levels1 | 10.74092 .3917303 65.09 0.000 9.999938 11.5368\n", + " levels2 | 9.464582 .3546025 59.99 0.000 8.794479 10.18574\n", + " levels3 | 8.480079 .3192442 56.78 0.000 7.876899 9.129449\n", + " levels4 | 8.000203 .3006194 55.34 0.000 7.432174 8.611645\n", + " levels5 | 7.044127 .2647529 51.94 0.000 6.543873 7.582624\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | 1.545267 .0467263 14.39 0.000 1.456346 1.639617`\n", + "` levels1 | 1.524804 .0395603 16.26 0.000 1.449206 1.604347`\n", "\n", - "Counties with lower social distancing also had a higher proportion of people without health insurance. The lowest social distancing counties had 11.0% uninsured adults, whereas the most social distancing counties had only 7.1% uninsured after adjusting for rurality and social distancing orders, a 54% (95% CI: 46%, 64%) difference. \n", + "Counties with lower social distancing also had a higher proportion of people without health insurance. The lowest social distancing counties had 10.7% uninsured adults, whereas the most social distancing counties had only 7.0% uninsured after adjusting for rurality and social distancing orders, a 52% (95% CI: 45%, 60%) difference. \n", "\n", "---" ] @@ -1442,17 +1435,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 48.23875 .8912549 209.80 0.000 46.52317 50.01759\n", - " levels2 | 47.49881 .8681976 211.22 0.000 45.82729 49.23129\n", - " levels3 | 48.55726 .8674856 217.34 0.000 46.88644 50.28762\n", - " levels4 | 49.43355 .8763789 220.02 0.000 47.74538 51.18141\n", - " levels5 | 50.63066 .8872871 223.94 0.000 48.92114 52.39993\n", + " levels1 | 47.99562 .7799153 238.23 0.000 46.4911 49.54882\n", + " levels2 | 48.2899 .7663704 244.31 0.000 46.81096 49.81556\n", + " levels3 | 49.35107 .7754712 248.13 0.000 47.85434 50.89462\n", + " levels4 | 49.66889 .7746225 250.41 0.000 48.17363 51.21056\n", + " levels5 | 50.96012 .7778188 257.55 0.000 49.4582 52.50765\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | .9527577 .013099 -3.52 0.000 .927427 .9787802`\n", + "` levels1 | .9418269 .0112491 -5.02 0.000 .9200352 .9641349`\n", "\n", - "The lowest social distancing counties had 48.2% flu vaccine coverage among Medicare beneficiaries, whereas the most social distancing counties had 50.6% after adjusting for rurality and social distancing orders, a 5.0% (95% CI: 2.2%, 7.8%) difference. " + "The lowest social distancing counties had 48.0% flu vaccine coverage among Medicare beneficiaries, whereas the most social distancing counties had 51.0% after adjusting for rurality and social distancing orders, a 6.2% (95% CI: 3.7%, 8.7%) difference. " ] }, { @@ -1464,14 +1457,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "5\n", - "LL: 7.8\n", - "UL: 2.2\n" + "6.2\n", + "LL: 8.7\n", + "UL: 3.7\n" ] } ], "source": [ - "invert .9527577 .927427 .9787802" + "invert .9418269 .9200352 .9641349" ] }, { @@ -1728,18 +1721,18 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 120492.2 2165.852 650.87 0.000 116321.1 124812.9\n", - " levels2 | 121736 2143.795 664.93 0.000 117605.9 126011.1\n", - " levels3 | 126707.4 2289.893 650.15 0.000 122297.8 131275.9\n", - " levels4 | 132681.5 2399.382 652.28 0.000 128061.2 137468.6\n", - " levels5 | 140242.2 2690.91 617.65 0.000 135066.1 145616.7\n", + " levels1 | 118675.2 1873.364 740.18 0.000 115059.7 122404.3\n", + " levels2 | 122077.6 1869.469 764.83 0.000 118468 125797.3\n", + " levels3 | 125227.2 2022.182 726.89 0.000 121325.9 129254\n", + " levels4 | 129432 1971.797 772.66 0.000 125624.5 133354.9\n", + " levels5 | 139390 2261.967 729.93 0.000 135026.4 143894.6\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | .8591722 .011219 -11.62 0.000 .8374625 .8814448`\n", + "` levels1 | .8513898 .0096039 -14.26 0.000 .832773 .8704228`\n", "\n", "\n", - "The lowest social distancing counties the 80th percentile of annual household income was around `$120,000`, whereas in the most social distancing counties it was `$140,000`, after adjusting for rurality and social distancing orders, a 16% (95% CI: 13%, 19%) difference. \n" + "The lowest social distancing counties the 80th percentile of annual household income was around `$120,000`, whereas in the most social distancing counties it was `$140,000`, after adjusting for rurality and social distancing orders, a 17% (95% CI: 15%, 20%) difference. \n" ] }, { @@ -1751,14 +1744,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "16.4\n", - "LL: 19.4\n", - "UL: 13.5\n" + "17.5\n", + "LL: 20.1\n", + "UL: 14.9\n" ] } ], "source": [ - "invert .8591722 .8374625 .8814448" + "invert .8513898 .832773 .8704228" ] }, { @@ -2015,18 +2008,18 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 41.4066 1.367529 112.74 0.000 38.8112 44.17555\n", - " levels2 | 40.94799 1.337411 113.66 0.000 38.40885 43.65499\n", - " levels3 | 37.43485 1.260305 107.60 0.000 35.04443 39.98832\n", - " levels4 | 34.79884 1.160151 106.47 0.000 32.59769 37.14863\n", - " levels5 | 33.23793 1.123619 103.64 0.000 31.10705 35.51478\n", + " levels1 | 40.67673 1.205293 125.06 0.000 38.38169 43.10901\n", + " levels2 | 37.48201 1.116649 121.64 0.000 35.35609 39.73576\n", + " levels3 | 35.28 1.065025 118.04 0.000 33.25315 37.4304\n", + " levels4 | 33.67241 1.010585 117.17 0.000 31.74882 35.71253\n", + " levels5 | 32.23657 .9852857 113.63 0.000 30.36215 34.22671\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | 1.245763 .0271148 10.10 0.000 1.193737 1.300057`\n", + "` levels1 | 1.26182 .0244192 12.02 0.000 1.214855 1.310599`\n", "\n", "\n", - "In the lowest social distancing counties, 41% of schoolage children were eligible for free or reduced price lunches. By comparison, in the most social distancing counties 33% were eligible, after adjusting for rurality and social distancing orders, a 24% (95% CI: 19%, 30%) difference. \n" + "In the lowest social distancing counties, 41% of schoolage children were eligible for free or reduced price lunches. By comparison, in the most social distancing counties 32% were eligible, after adjusting for rurality and social distancing orders, a 26% (95% CI: 21%, 31%) difference. \n" ] }, { @@ -2254,18 +2247,18 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 12.98755 .3612017 92.19 0.000 12.29855 13.71514\n", - " levels2 | 12.55802 .348123 91.28 0.000 11.89392 13.2592\n", - " levels3 | 11.60655 .3183067 89.39 0.000 10.99915 12.24749\n", - " levels4 | 11.0192 .3022595 87.48 0.000 10.44242 11.62783\n", - " levels5 | 10.22625 .2805256 84.75 0.000 9.690954 10.79112\n", + " levels1 | 12.93501 .3216935 102.93 0.000 12.31962 13.58114\n", + " levels2 | 11.84488 .2918237 100.33 0.000 11.28651 12.43088\n", + " levels3 | 11.64897 .2858669 100.05 0.000 11.10194 12.22295\n", + " levels4 | 10.89867 .2676669 97.26 0.000 10.38648 11.43612\n", + " levels5 | 10.17634 .2475631 95.37 0.000 9.702515 10.67331\n", "`\n", "

\n", "Percent difference
\n", - "`levels1 | 1.27002 .0257634 11.78 0.000 1.220515 1.321533`\n", + "` levels1 | 1.271086 .0226617 13.45 0.000 1.227437 1.316288`\n", "\n", "\n", - "The lowest social distancing counties had greater food insecurity, among 13.0% of residents. The most social distancing counties had 10.2%, after adjusting for rurality and social distancing orders, a 27% (95% CI: 22%, 32%) difference. \n", + "The lowest social distancing counties had greater food insecurity, among 12.9% of residents. The most social distancing counties had 10.2%, after adjusting for rurality and social distancing orders, a 27% (95% CI: 23%, 32%) difference. \n", "\n", "---" ] @@ -2519,15 +2512,15 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 68.69511 2.008972 144.63 0.000 64.86832 72.74766\n", - " levels2 | 74.2871 2.075381 154.20 0.000 70.32878 78.46819\n", - " levels3 | 77.96532 2.065593 164.43 0.000 74.02015 82.12077\n", - " levels4 | 82.76289 2.142898 170.55 0.000 78.66768 87.07129\n", - " levels5 | 90.49707 2.206371 184.79 0.000 86.27436 94.92647\n", + " levels1 | 70.82756 1.902913 158.57 0.000 67.19442 74.65715\n", + " levels2 | 76.05735 1.841866 178.86 0.000 72.5317 79.75439\n", + " levels3 | 78.16616 1.906506 178.71 0.000 74.51738 81.9936\n", + " levels4 | 82.19944 1.871535 193.65 0.000 78.61194 85.95066\n", + " levels5 | 91.41497 2.000766 206.31 0.000 87.57646 95.42173\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | .7590865 .0177023 -11.82 0.000 .7251717 .7945875`\n", + "` levels1 | .7747917 .0157072 -12.59 0.000 .7446098 .8061969`\n", "\n", "\n", "In the lowest social distancing counties, 69% of residents had access to physical spaces for exercise, whereas in the most social distancing counties 90% had access, after adjusting for rurality and social distancing orders, a 32% (95% CI: 26%, 40%) difference. " @@ -2542,14 +2535,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "31.7\n", - "LL: 37.9\n", - "UL: 25.9\n" + "29.1\n", + "LL: 34.3\n", + "UL: 24\n" ] } ], "source": [ - "invert .7590865 .7251717 .7945875" + "invert .7747917 .7446098 .8061969" ] }, { @@ -2785,17 +2778,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 2.340737 .179192 11.11 0.000 2.014605 2.719663\n", - " levels2 | 2.256858 .1718827 10.69 0.000 1.943912 2.620184\n", - " levels3 | 2.106022 .1585213 9.89 0.000 1.817158 2.440806\n", - " levels4 | 2.077067 .155689 9.75 0.000 1.793278 2.405766\n", - " levels5 | 2.010469 .1498266 9.37 0.000 1.737253 2.326653\n", + " levels1 | 2.367699 .1589751 12.84 0.000 2.075745 2.700717\n", + " levels2 | 2.162979 .1437291 11.61 0.000 1.898848 2.46385\n", + " levels3 | 1.876126 .1261733 9.36 0.000 1.644436 2.14046\n", + " levels4 | 1.934634 .1285511 9.93 0.000 1.698396 2.203733\n", + " levels5 | 2.068613 .1338108 11.24 0.000 1.822293 2.348229\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | 1.164274 .0640694 2.76 0.006 1.045235 1.29687`\n", + "` levels1 | 1.144583 .0535491 2.89 0.004 1.044297 1.2545`\n", "\n", - "The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders. " + "The lowest social distancing counties had 14% (95% CI: 4.4%, 25%) less overcrowding, after adjusting for rurality and social distancing orders. " ] }, { @@ -3028,21 +3021,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 25.34917 .3433241 238.69 0.000 24.68512 26.03108\n", - " levels2 | 25.1898 .3382312 240.29 0.000 24.53552 25.86152\n", - " levels3 | 24.94332 .328453 244.27 0.000 24.3078 25.59546\n", - " levels4 | 24.55284 .3220041 244.06 0.000 23.92977 25.19214\n", - " levels5 | 23.45215 .3067361 241.22 0.000 22.85859 24.06111\n", + " levels1 | 25.54333 .3027171 273.42 0.000 24.95686 26.14359\n", + " levels2 | 24.97196 .2904891 276.61 0.000 24.40905 25.54785\n", + " levels3 | 24.65616 .2854101 276.88 0.000 24.10306 25.22195\n", + " levels4 | 24.76482 .2843834 279.48 0.000 24.21366 25.32852\n", + " levels5 | 23.59149 .2675883 278.67 0.000 23.07281 24.12183\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | 1.080889 .0110241 7.63 0.000 1.059497 1.102713`\n", - "\n", - "Counties that did not restrict movement \n", + "` levels1 | 1.082735 .0095077 9.05 0.000 1.06426 1.101531`\n", "\n", - "As the share of the youth popuplation increased \n", - "\n", - "The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders. " + "Counties with the least restriction of movement had 8.2% more children (95% CI: 6.4%, 10%) than areas that most greatly had their movement reduced." ] }, { @@ -3278,17 +3267,17 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 14.13568 .3077327 121.67 0.000 13.54522 14.75188\n", - " levels2 | 14.34026 .3107842 122.88 0.000 13.74389 14.96251\n", - " levels3 | 14.63441 .3106594 126.41 0.000 14.03802 15.25613\n", - " levels4 | 14.94357 .3143969 128.54 0.000 14.3399 15.57266\n", - " levels5 | 15.1625 .3164327 130.28 0.000 14.55482 15.79556\n", + " levels1 | 14.16821 .2681426 140.07 0.000 13.65229 14.70363\n", + " levels2 | 14.60455 .2709946 144.50 0.000 14.08295 15.14547\n", + " levels3 | 14.94807 .2750933 146.96 0.000 14.4185 15.49708\n", + " levels4 | 15.00888 .2744157 148.15 0.000 14.48055 15.55647\n", + " levels5 | 15.22254 .2732011 151.71 0.000 14.69638 15.76753\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | .9322789 .0144558 -4.52 0.000 .9043722 .9610467`\n", + "` levels1 | .930739 .0124421 -5.37 0.000 .9066697 .9554473`\n", "\n", - "The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders. " + "Counties that did the best at restricting movement had 7.4% (95% CI: 4.7%, 10%) more elderly people, compared to the lowest tier of movement restriction." ] }, { @@ -3300,14 +3289,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "7.3\n", - "LL: 10.6\n", - "UL: 4.1\n" + "7.4\n", + "LL: 10.3\n", + "UL: 4.7\n" ] } ], "source": [ - "invert .9322789 .9043722 .9610467" + "invert .930739 .9066697 .9554473" ] }, { @@ -3556,15 +3545,15 @@ "\n", "Relative effect measures
\n", "` ^\n", - " levels1 | 28.47816 1.219824 78.19 0.000 26.18495 30.97219\n", - " levels2 | 28.01057 1.171094 79.71 0.000 25.80679 30.40253\n", - " levels3 | 29.37394 1.228875 80.80 0.000 27.06149 31.88399\n", - " levels4 | 29.15434 1.174793 83.70 0.000 26.94037 31.55026\n", - " levels5 | 31.81989 1.253444 87.84 0.000 29.45562 34.37392\n", + " levels1 | 29.07819 1.106746 88.54 0.000 26.98794 31.33033\n", + " levels2 | 29.62044 1.078855 93.03 0.000 27.57963 31.81226\n", + " levels3 | 30.05066 1.117545 91.50 0.000 27.93823 32.32281\n", + " levels4 | 30.79493 1.091354 96.71 0.000 28.72851 33.00999\n", + " levels5 | 31.866 1.11741 98.72 0.000 29.74948 34.1331\n", "`\n", "

\n", "Percent difference
\n", - "` levels1 | .8949799 .0268652 -3.70 0.000 .843844 .9492145`\n", + "` levels1 | .9125145 .0241028 -3.47 0.001 .8664759 .9609994`\n", "\n", "The lowest social distancing counties had 16% (95% CI: 4.5%, 30%) less overcrowding, after adjusting for rurality and social distancing orders. " ] @@ -3578,14 +3567,14 @@ "name": "stdout", "output_type": "stream", "text": [ - "11.7\n", - "LL: 18.5\n", - "UL: 5.4\n" + "9.6\n", + "LL: 15.4\n", + "UL: 4.1\n" ] } ], "source": [ - "invert .8949799 .843844 .9492145" + "invert .9125145 .8664759 .9609994" ] }, { @@ -4278,6 +4267,152 @@ "frame change default" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n", + "\n", + "# Figure Data" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "\n", + "\n", + " +----------------------------------------------------------+\n", + " | strat level avg LL UL |\n", + " |----------------------------------------------------------|\n", + " 1. | pcp_rate 1 49.9 45.6 54.7 |\n", + " 2. | pcp_rate 2 54.9 50.3 59.9 |\n", + " 3. | pcp_rate 3 57.9 52.8 63.5 |\n", + " 4. | pcp_rate 4 61.5 56.3 67.1 |\n", + " 5. | pcp_rate 5 73.6 67.6 80.2 |\n", + " |----------------------------------------------------------|\n", + " 6. | uninsured_p 1 10.7 10 11.5 |\n", + " 7. | uninsured_p 2 9.5 8.8 10.2 |\n", + " 8. | uninsured_p 3 8.5 7.9 9.1 |\n", + " 9. | uninsured_p 4 8 7.4 8.6 |\n", + " 10. | uninsured_p 5 7 6.5 7.6 |\n", + " |----------------------------------------------------------|\n", + " 11. | fluvaccine 1 48 46.5 49.5 |\n", + " 12. | fluvaccine 2 48.3 46.8 49.8 |\n", + " 13. | fluvaccine 3 49.4 47.9 50.9 |\n", + " 14. | fluvaccine 4 49.7 48.2 51.2 |\n", + " 15. | fluvaccine 5 51 49.5 52.5 |\n", + " |----------------------------------------------------------|\n", + " 16. | income80 1 118675.2 115059.7 122404.3 |\n", + " 17. | income80 2 122077.6 118468 125797.3 |\n", + " 18. | income80 3 125227.2 121325.9 129254 |\n", + " 19. | income80 4 129432 125624.5 133354.9 |\n", + " 20. | income80 5 139390 135026.4 143894.6 |\n", + " |----------------------------------------------------------|\n", + " 21. | schoollunch 1 40.7 38.4 43.1 |\n", + " 22. | schoollunch 2 37.5 35.4 39.7 |\n", + " 23. | schoollunch 3 35.3 33.3 37.4 |\n", + " 24. | schoollunch 4 33.7 31.7 35.7 |\n", + " 25. | schoollunch 5 32.2 30.4 34.2 |\n", + " |----------------------------------------------------------|\n", + " 26. | foodinsec 1 12.9 12.3 13.6 |\n", + " 27. | foodinsec 2 11.8 11.3 12.4 |\n", + " 28. | foodinsec 3 11.6 11.1 12.2 |\n", + " 29. | foodinsec 4 10.9 10.4 11.4 |\n", + " 30. | foodinsec 5 10.2 9.7 10.7 |\n", + " |----------------------------------------------------------|\n", + " 31. | exercise 1 70.8 67.2 74.7 |\n", + " 32. | exercise 2 76.1 72.5 79.8 |\n", + " 33. | exercise 3 78.2 74.5 82 |\n", + " 34. | exercise 4 82.2 78.6 86 |\n", + " 35. | exercise 5 91.4 87.6 95.4 |\n", + " |----------------------------------------------------------|\n", + " 36. | overcrowding 1 2.4 2.1 2.7 |\n", + " 37. | overcrowding 2 2.2 1.9 2.5 |\n", + " 38. | overcrowding 3 1.9 1.6 2.1 |\n", + " 39. | overcrowding 4 1.9 1.7 2.2 |\n", + " 40. | overcrowding 5 2.1 1.8 2.3 |\n", + " |----------------------------------------------------------|\n", + " 41. | youth 1 25.5 25 26.1 |\n", + " 42. | youth 2 25 24.4 25.5 |\n", + " 43. | youth 3 24.7 24.1 25.2 |\n", + " 44. | youth 4 24.8 24.2 25.3 |\n", + " 45. | youth 5 23.6 23.1 24.1 |\n", + " |----------------------------------------------------------|\n", + " 46. | elderly 1 14.2 13.7 14.7 |\n", + " 47. | elderly 2 14.6 14.1 15.1 |\n", + " 48. | elderly 3 14.9 14.4 15.5 |\n", + " 49. | elderly 4 15 14.5 15.6 |\n", + " 50. | elderly 5 15.2 14.7 15.8 |\n", + " |----------------------------------------------------------|\n", + " 51. | segregation_wnw 1 29.1 27 31.3 |\n", + " 52. | segregation_wnw 2 29.6 27.6 31.8 |\n", + " 53. | segregation_wnw 3 30.1 27.9 32.3 |\n", + " 54. | segregation_wnw 4 30.8 28.7 33 |\n", + " 55. | segregation_wnw 5 31.9 29.7 34.1 |\n", + " +----------------------------------------------------------+\n" + ] + } + ], + "source": [ + "frame change results\n", + "list" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n", + "\n", + "# Methods Detail\n", + "\n", + "## Baseline Health Data\n", + "In order to identify explanatory health and socioeconomic indicators, we used the 2019 Robert Wood Johnson Foundation (RWJF) County Health Rankings (CHR) dataset.20,23 The publicly available dataset contains dozens of metrics compiled from national surveys and healthcare databases. It is a well-documented public health resource,20 including data from the American Community Survey and Center for Medicare and Medicaid Services. \n", + "\n", + "We compared the intensity of social distancing to 11 metrics: three healthcare, two economic, three structural, and three demographic. These were selected from the CHR dataset because they are established indicators of other health and behavioral outcomes,20 with an emphasis on emergent concerns about equality arising during the pandemic.\n", + "\n", + "## Healthcare metrics\n", + "To gauge overall baseline healthcare access and utilization, we examined primary care providers per 100,000 population and percent uninsured under age 65 (e.g., Medicare eligibility). As a marker for a closely related preventive health behavior, we examined whether earlier influenza vaccination rates were associated with how much the county was likely to slow down during the current coronavirus outbreak. This was quantified as the percent of annual Medicare enrollees having an annual influenza vaccination.\n", + "\n", + "## Economic metrics\n", + "We explored two baseline economic metrics, one representing the overall wealth of the community and one proxy for poverty: 80th percentile of annual household income in dollars and the percent of school-age children eligible for subsidized or free lunches.\n", + "\n", + "## Structural metrics\n", + "Three lifestyle metrics were selected to provide a diverse snapshot of baseline structural factors that could influence defiance of prolonged stay-at-home orders. The percent of people experiencing food insecurity was established from survey responses and a cost-of-food index. Access to exercise opportunities was the percent of population with adequate access to locations for physical activity. The percent of households with overcrowding was based on housing condition surveys.\n", + "\n", + "## Demographic metrics\n", + "The three demographic metrics were: percent of youth (age under 18 years) because of concerns about non-compliance with stay-at-home orders, the percent of elderly (aged 65 years and above) because they are risk group for COVID-19 mortality, and a residential segregation index (white versus non-white).\n", + "\n", + "## Primary Mobility Data\n", + "The analytic dataset started with public, anonymized, aggregated county-level (or similar geopolitical units) data from smartphone GPS movement tracing, collected from January 1 through April 15, 2020 in the United States, pre-processed by Descartes Labs (Santa Fe, New Mexico, United States). Raw mobility data generated from location services were processed using a parallel bucket sort to create device-based (e.g., node) records that for a given day are longitudinal.24 Maximum distance mobility (Mmax) was defined as by the maximum Haversine (great circle) distance in kilometers from the first location report.7 Conceptually, this represents the straight-line distance between the first observation and the day’s farthest. Across all reports, the median accuracy of location measurement was 15 to 20 meters. Adjustments were made for poor GPS signal locks, too few observations (less than 10 reports per day), and signal accuracy (50 meter threshold).7 To reduce bias from devices merely transiting through a county (e.g., interstate highways), Mmax is limited to nodes with 8 hours of observation per day. The resulting analysis dataset contained 2,633 of 3,142 US counties.\n", + "
\n", + "The pre-processed mobility dataset in this analysis could not be used to identify individuals.\n", + "Using geotemporally coarse data provides further safeguards for protecting privacy. To this end, the values of Mmax were summarized by taking the median per county-day (m50). The median was indexed against the baseline period February 17 through March 7, 2020 (e.g., before widespread social distancing) to obtain our main study metric: the percentage change in mobility since baseline (m50_index) as a proxy for the intensity of social distancing. \n", + "\n", + "## Variable Construction\n", + "To account for weekly periodicity in movement (e.g., lower on weekends) we limited analysis to weekdays. To prevent undue influence from single-day variability, we averaged the last 3 weekday values of m50_index: April 13, 14, 15. The resulting distribution approximated a Gaussian function; to reduce outlier influence, we constructed a 5-level inverted stratification of m50_index, where the highest quintile (5) represented the greatest reduction in mobility since baseline, interpreted as the highest 20% of counties in terms of social distancing intensity. Category boundaries for m50_index by quintile were: lowest mobility change (1) +193% to -45.8%, (2) -46.0% to -55.4%, (3) -55.5% to -62.3%, (4) 62.4% to 74.8%, and highest (5) -75.0% to -100%. Only 7 counties (all less than 20,000 population) showed an increase in mobility from baseline and were included in the lowest change category.\n", + "\n", + "## Potential Confounders\n", + "We adjusted models for two potential confounders. First, state and municipal stay-at-home-orders issued from February through April 2020 were identified by county.25 These orders limited travel to basic necessities and employment in sectors deemed essential. Municipal stay-at-home orders were identified for the eight states that did not have stay-at-home orders.\n", + "
\n", + "Second, even though our main outcome was change in mobility from baseline, rurality might be considered a potential confounder due to distances traveled for essential activities. We used federal rural-urban continuum codes (RUCC) to adjust for rurality and transportation connections between city centers and satellite counties.26 Although RUCC can be conceptualized as a 9-point ordinal scale of urbanicity (or rurality), it was modeled using indicator coding to impose fewer assumptions. Other potential spatial and economic confounders (number of solo vehicle commuters, longer vehicle commuting times, and CHR socioeconomic status composite rank) were not included in the final model because they did not meaningful improve model fit.\n", + "\n", + "## External Validation \n", + "In the context of social distancing, coarse mobility data have the potential for misclassification. One way to cross-validate the findings is to compare these data to more granular location information, such as by type of visited venue. The data came from aggregated and anonymized GPS traces of devices for which the Location History setting within Google apps had been turned on (off by default). Since detailed information on data collection was not available, we did not consider the Google Location Services data appropriate for the primary analysis.\n", + "
\n", + "During the study period, Google published county-level datasets showing COVID-19-related mobility changes across six types of venues: grocery and pharmacy; parks, transit stations, retail and recreation, places of residence; and places of work. The metric was percent change in mobility changes since baseline, January 3 to February 6, 2020, controlling for day of week. A validation dataset was created for March 1 to April 11, the overlap period with mobility data used in the primary analyses, for which county-day could be established. Pearson product-moments were calculated for correlations between the six venue-specific changes in mobility and percent change in mobility from baseline in the primary location data. To have confidence in overall mobility change to serve a proxy for social distancing, we expected the strongest correlations with staying at home, transit, work and retail, and less correlation with use of other venues that were permissible or essential during stay-at-home orders.\n", + "\n", + "## Statistical Analysis\n", + "Datasets were analyzed with Stata MP (version 16, College Station, Texas, United States). Scaled Poisson regression with robust variance estimators was used in base models regressing each of the 11 metrics individually against quintiles of mobility change. Negative binomial (NB2) models were employed when warranted by further overdispersion. The adjusted models included indicator variables to control for rurality/urbanicity. For a given health or socioeconomic indicator, mean and 95% confidence intervals were calculated for each quintile using non-intercept models; pairwise contrasts of percent difference between quintiles were estimated using the full model with adjustment. We intentionally did not include all explanatory variables in a combined multivariable model because we wanted to highlight the individual associations, not explain away the variance. Adjusted model-predicted means and confidence intervals of bivariate associations were plotted visually. For cross-validation, pairwise pearson product-moment correlations were generated, comparing zero-recentered m50_index against percent change from baseline by venue, as reported by Google Location Services users. Code and datasets are available at https://github.com/opioiddatalab/covid.\n" + ] + }, { "cell_type": "markdown", "metadata": {},