prepCleanData: Transforms data into a form tailored for efficient evaluation by Stan
Description
Vertex names, if not already factors, are converted to
factors. The number of thresholds per item is determined by the
largest absolute response value. Missing responses are filtered
out. Responses on the same pair of vertices on the same item are
grouped together. Within a vertex pair and item, responses
are ordered from negative to positive.
Usage
prepCleanData(df)
Value
a data list suitable for passing as the data
argument to pcStan or stan
Arguments
df
a data frame with pairs of vertices given in columns pa1 and pa2, and item response data in other columns
Details
Note: Reordering of responses is likely unless something like
normalizeData has been used with .sortRows=TRUE.
See Also
Other data preppers:
prepData(),
prepFactorModel(),
prepSingleFactorModel()