example data set: weighted categorical panel data

[corybrunson]

The two data sets currently installed with ggalluvial, vaccinations and majors, illustrate key features—indeed, those features were developed around the data sets. Several features currently in development are not so well illustrated by these or other data sets (including datasets::Titanic and alluvial::Refugees). An ideal data set to include would consist of individual-level panel data with categorical responses and varying sample weights. This would allow example code, including alluvia–lodes conversion functions and alluvial plotting layers, to showcase the following features:

• variable cohort sizes (lode heights y) – see #18 and #50
• variable lode weights (e.g. map computed variable prop to label aesthetic) – see #50
• negative strata for discontinued participants – see #45

One possible source is the National Longitudinal Surveys, in particular the recoded labor force status variables.