Use Cases

Please add use cases, pseudo code and examples that you feel are necessary.

1. Rename/ recode alt/ aatie variables (i.e. to make them correspond with ego variables). This can be done the easiest and fastest with a flat/ global data frame. Lists of data frames/ tibbles would make lapply/map like operations necessary, that are less easy for beginners to understand and more cumbersome for everyone.

2. Conduct calculations on alts/ aaties per ego. Here lists of tibbles and flat/ global dataframes are at par. There should only be slight differences in speed and the handling is almost the same for both formats. Flat/ global dataframes could be processed with split()ing -> *apply()ing, aggregate()ing, or by going with dplyr's group_by() and do(). Lists of tibbles would simply be processed with *apply()/map() commands, which would have slight speed benefits. All of these would need the users to understand the concept of the aggregate/lapply logic and the definition of (anonymous) functions.

3. Combining alter-attributes and alter-alter ties for structural measures. I think the best format here is a flat/ global aaties data frame where the corresponding source and target alter attributes are added as additional columns. Another option for those operations is an igraph object, which has some operators for such procedures.

4. Comparing/ combining ego and alter attributes. Here I would suggest using a flat/ global alts data frame with the ego attributes added as columns to and repeated for each alter of an ego. Alternatively mapply()/map2() could be employed here, but I feel that this would make the handling much harder. The current implementation of the comp_ply() command supports this.