merge multiple LISS modules into a single panel dataset

takes the output of merge_liss_modules() (or a list of per-module data frames) and joins them into one wide dataset keyed by respondent and wave year. module-specific columns are prefixed with the module code (e.g. ch_s004, cv_s004) to avoid name collisions.

Usage

merge_liss_panel(
  results,
  join_by = c("nomem_encr", "wave_year"),
  shared_cols = c("nohouse_encr"),
  join_type = c("full", "inner", "left"),
  write_to = NULL
)

Arguments

results

either the named list returned by merge_liss_modules() (where each element has a $data tibble), or a named list of data frames directly. names should be module codes (e.g. "ch", "cv").

join_by

character vector of columns to join on. defaults to c("nomem_encr", "wave_year").

shared_cols

character vector of additional columns to keep unprefixed (carried from the first module that has them). defaults to c("nohouse_encr").

join_type

character. type of join: "full" (default), "inner", or "left" (keeps all rows from the first module).

write_to

optional file path. if provided, the merged panel is written as SAV (SPSS format, preserving labels). should end in .sav.

Value

a tibble with all modules joined side by side.

Examples

if (FALSE) { # \dontrun{
results <- merge_liss_modules(recipe_paths, data_dir = "~/Downloads/liss")
panel <- merge_liss_panel(results)
panel <- merge_liss_panel(results, write_to = "./output/liss_panel.sav")
} # }