-
Notifications
You must be signed in to change notification settings - Fork 0
/
surveys_plots_all_3.Rmd
277 lines (206 loc) · 10.6 KB
/
surveys_plots_all_3.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
---
title: "BqExAvI Survey Analysis (2017-2022)"
author: "Modesto"
date: "`r Sys.Date()`"
format:
html:
page-layout: full
toc: true
toc-location: left
toc-depth: 2
number-sections: true
code-overflow: wrap
code-fold: true
code-summary: "Show the code"
link-external-icon: true
link-external-newwindow: true
#output:
# pdf_document:
# latex_engine: xelatex
---
# Contents and Disclaimer
```{r}
```
This file contains the data from the "Experimental Advanced Biochemistry I" course (Biochemistry Degree, [Universidad Autónoma de Madrid](https://www.uam.es)). It is a 6 ECTS practical course for 3rd year undergrad students. In this course, starting from of a simple signal transduction pathway and an experimental system, students design their own experimental program, carry out the experiments, draw conclusions and present the results. Over the last years, a number of lecturers and professors from the [Biochemistry Department](https://www.bq.uam.es/Index.php) have participated in this course, as follows (in alphabetical order): Julián Aragonés, Juan J. Arredondo, Víctor Calvo, José G. Castaño, Alicia González-Martín, Benilde Jiménez, Marina Lasa, Óscar Martínez-Costa, Luis del Peso, Modesto Redrejo-Rodríguez, Ana I. Rojo, Alejandro Samhan-Arias, and Isabel Sánchez-Pérez.
The questionnaires were completed by the students in 2017-2022 directly on Moodle on the last day of each course. This is a preliminary summary of the data analysis. The GitHub [repo](https://github.com/mredrejo/bqexav) contains the original files of all analyzes. This report only contains the data analysis and plots, without any results discussion.
We are presenting this project in the workshop [*Evolving molecular biosciences education*](https://www.eventsforce.net/biochemsoc/frontend/reg/thome.csp?pageID=82803&eventID=164) (UK Biochemical Society and FEBS joint event), with a [Poster](poster_may23.pdf). A full manuscript will be also available soon.
All these data are made available under the Creative Common License ([CC BY-NC-ND 3.0 ES](https://creativecommons.org/licenses/by-nc-nd/4.0/)).
::: {.callout-warning}
## Preliminary
This is only a preliminary analysis. Contact [modesto.redrejo\@uam.es](mailto:modesto.redrejo@uam.es) or [juan.arredondo\@uam.es](mailto:juan.arrredondo@uam.es) for any feedback or queries.
:::
# Moodle survey
```{r results='hide', message=FALSE, warning=FALSE}
#Load/install requires packages
paquetes <- c("ggplot2","data.table","kableExtra","corrplot","likert","ggpubr","heatmaply","reshape2","plotly","dplyr")
unavailable <- setdiff(paquetes, rownames(installed.packages()))
invisible(install.packages(unavailable))
invisible(lapply(paquetes, library, character.only = TRUE))
```
We import data from student responses in 5 years, between 2018-2022. The quiz consists of up to 77 questions, including 5 free text questions (50, 51, 52, 75, 76 & 77), 7 questions with three options, and the rest as a 5-degrees *Likert* scale.
```{r warning=FALSE}
#load questions
questions <- read.csv("questions.csv", head=TRUE, sep=";")
questions <- cbind(row.names(questions),questions[,c(3,1,2)])
#add type variable
questions$type <- "Pos."
questions$type[questions$Section=="Open"] <- ""
questions[c(4,10,56,63,63,64,65,66,67,72),5] <- "Neg."
questions$type[questions$Section=="Open"] <- "NA"
colnames(questions) <- c("No.","Since","Section","Question","Type")
#write.csv(questions,"questions_final.csv", row.names=FALSE)
#diplay the table
kbl(questions[,1:4], align = "cccl", caption = "Table 1. Students opinion quizz. The bulk of the questionaire was designed for the year 2017 and new questions were added as indicated.") %>%
kable_styling(bootstrap_options = "striped", full_width = F) %>%
column_spec(1, italic = T)
```
# Load data and pairwise t.test
Survey responses were downloaded from Moodle as txt/csv files. Moodle updates caused some format differences that could be worked around after opening the files with Numbers and exporting them as tables with ";" as the column separator. We now show all vs. all pairwise t-tests performed to detect significant differences in responses per year. The tables below contain the pairwise p-values for each comparison, with significant values highlighted in blue (p\<0.05) and red (p\<0.01)
```{r warning=FALSE, results='asis'}
#read the data in a list of dataframes
#didn't use the headers to avoid mistakes
quiz <- lapply(2017:2022, function(x) read.csv(paste0("survey",x,".csv"),header=FALSE,skip=1,sep=";"))
#add Year as the third variable (empty so far)
curso <- c("2017","2018","2019","2020","2021","2022")
for (i in 1:length(quiz)){
quiz[[i]][,3] <- curso[i]
}
#adjust questions changes
#remove questions in column 32 & 40 from 2017, because we removed it in the following years
quiz[[1]] <- quiz[[1]][,-c(32,40)]
quiz[[6]] <- quiz[[6]][,-10]
names(quiz[[6]]) <- names(quiz[[5]])
names(quiz[[1]]) <- names(quiz[[2]])
#merge all dataframes and name the columns
data <- Reduce(function(x, y) merge(x, y, all=TRUE), quiz)
#take the colnames from the last quiz that contains all the questions
colnames(data)[3] <- "Curso"
#statistics analysis
#subset questions 1/2: remove leftmost junk columns
subdata_all <- data[,c(3,11:87)]
names(subdata_all) <- c("Curso",paste0("Q",1:77))
#write.csv(subdata_all, "merged_data.csv")
#subset questions 2/2: remove open questions
open <- c(row.names(questions[questions$Section=="Open",]))
subdata <- subdata_all[,-(as.integer(open)+1)]
subdata <- sapply(subdata,as.numeric)
subdata <- as.data.frame(subdata)
#perform tests and display table in a loop
tests <- list()
nombres <- c()
for (i in 2:ncol(subdata)){
subdata[,i][!(subdata[,i] %in% c(1,2,3,4,5))] <- NA
#subset for years with answers to avoid void groups
kkk <- subset(subdata,!is.na(subdata[,i]))
tests[[i-1]] <- pairwise.t.test(x=as.numeric(kkk[,i]),g=as.numeric(kkk[,1]),paired = F)
nombres[i-1] <- questions[,4][(which(questions$No. %in% gsub("\\D", "",colnames(subdata[i]))))]
names(tests) <- nombres
print(as.data.frame(format(tests[[i-1]]$p.value, scientific=F,nsmall=6)) %>% replace(., . < 0, "") %>%
mutate_all(~cell_spec(.x, color = ifelse(.x < 0.01, "firebrick", ifelse(.x < 0.05, "steelblue",
"black")))) %>%
kable(escape = F, align = "cccl", caption =paste("<b>",names(tests[i-1]),"</b>"), digits=4) %>%
kable_styling(bootstrap_options = "striped", full_width = F, position = "center") %>%
column_spec(1, bold = T))
}
```
# Likert scale plots
*Likert* scale responses are grouped by the sections in the quiz.
## General Methodology
```{r Fig2, echo=TRUE, fig.height=7,warning=FALSE}
#lickert
#change question names
tablita <- data.frame(matrix(NA, # Create empty data frame
nrow = length(colnames(subdata)),
ncol = 2))
for (i in 2:length(colnames(subdata))){
tablita[i-1,] <- cbind(colnames(subdata[i]),questions$Question[as.numeric(gsub("\\D", "",colnames(subdata[i])))])
colnames(subdata)[i] <- tablita[i-1,2]
}
for (i in 2:72){
subdata[,i] <- factor(subdata[,i])
}
subdata$Curso <- factor(subdata$Curso,levels=c(2022,2021,2020,2019,2018,2017))
#questions with 3 options
#General Methodology
subdata[,c(8:12)] <- lapply(subdata[,c(8:12)], function(x) factor(x,
labels = c("Traditional","Same","Open Question"))
)
xlikgroup3a = likert(subdata[,c(8:12)], grouping = subdata$Curso)
plot(xlikgroup3a, type = "bar", centered = T)
```
## Activities Length
```{r Fig3, echo=TRUE, fig.height=10,warning=FALSE}
subdata[,c(29:35)] <- lapply(subdata[,c(29:35)], function(x) factor(x,
labels = c("Less","Fine","More"))
)
xlikgroup3b = likert(subdata[,c(29:35)], grouping = subdata$Curso)
plot(xlikgroup3b, type = "bar", centered = T)
#title(main = "Activities Length", xlab = "X axis", ylab = "Y axis", cex.main = 4, font.main = 3)
#legend("bottom", c("Less","Fine","More"))
```
## Equipment
```{r Fig4, echo=TRUE,warning=FALSE}
subdata[,c(2:4)] <- lapply(subdata[,c(2:4)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
xlikgroup5a = likert(subdata[,c(2:4)], grouping = subdata$Curso)
plot(xlikgroup5a, type = "bar", centered = T)
```
## Length and Schedule
```{r Fig5, echo=TRUE, fig.height=5,warning=FALSE}
subdata[,c(5:7)] <- lapply(subdata[,c(5:7)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
xlikgroup5b = likert(subdata[,c(5:7)], grouping = subdata$Curso)
plot(xlikgroup5b, type = "bar", centered = T)
```
## Method Objectives
```{r Fig6, echo=TRUE, fig.height=13,warning=FALSE}
subdata[,c(13:21)] <- lapply(subdata[,c(13:21)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
xlikgroup5c = likert(subdata[,c(13:21)], grouping = subdata$Curso)
plot(xlikgroup5c, type = "bar", centered = T)
```
## Activities Interest
```{r Fig7, echo=TRUE, fig.height=11, warning=FALSE}
subdata[,c(22:28)] <- lapply(subdata[,c(22:28)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
xlikgroup5d = likert(subdata[,c(22:28)], grouping = subdata$Curso)
plot(xlikgroup5d, type = "bar", centered = T)
```
## Assessment
```{r Fig8, echo=TRUE, fig.height=9,warning=FALSE}
subdata[,c(36:41)] <- lapply(subdata[,c(36:41)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
xlikgroup5e = likert(subdata[,c(36:41)], grouping = subdata$Curso)
plot(xlikgroup5e, type = "bar", centered = T)
```
## Learning Objectives
```{r Fig9, echo=TRUE, fig.height=12,warning=FALSE}
subdata[,c(42:50)] <- lapply(subdata[,c(42:50)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
xlikgroup5f = likert(subdata[,c(42:50)], grouping = subdata$Curso)
plot(xlikgroup5f, type = "bar", centered = T)
```
## ELN
```{r Fig10, echo=TRUE, fig.height=28,warning=FALSE}
#subset to remove empty years
subdata[,c(51:72)] <- lapply(subdata[,c(51:72)], function(x) factor(x,
labels = c("Not at all","Disagree","OK","Agree","Completely Agree"))
)
eln <- subset(subdata[,c(1,51:72)][subdata$Curso==2019|subdata$Curso==2020|subdata$Curso==2021|subdata$Curso==2022,])
xlikgroup5g = likert(eln[2:23], grouping = eln$Curso)
plot(xlikgroup5g, type = "bar", centered = T, title="ELN")
```
# [Open questions](surveys_free_text.html)
Free-text questions have been analyzed independently by automatic text *lemmatization* and plot methods. The detailed report can be found [here](surveys_free_text.html).
# Acknowledgments
This work has been supported by UAM Teaching Innovation Grants (M-015.17-INN, M-020.18-IMP, M_002.19_INN and M_009.20_IMP).
### Session Info
```{r}
sessionInfo()
```