cat.pbc: Calculate Item Discrimination for Each Category of...

In eatPrep: Prepare Data for IRT Analyses

Description

catPbc calculates discrimination statistics for the categories of categorical variables, i.e. the correlations of each category with the total score on the test. This information can be useful in determining which categories of an item are influencing the overall fit and discrimination in item response scaling applications and/or to find mistakes in recoding.

Usage

catPbc(datRaw, datRec, idRaw, idRec, context.vars = NULL, values, subunits, file.name = NULL, verbose = F)

Arguments

datRaw	A merged unrecoded dataset
datRec	The same dataset as in datRaw, in recoded form
idRaw	Name or column number of the identifier (ID) variable in unrecoded dataset
idRec	Name or column number of the identifier (ID) variable in recoded dataset
context.vars	Names or column numbers of one or more context variables (e.g., sex, school). catPbc will ignore these variables.
values	Data frame with information about values, see inputList for details.
subunits	Data frame with information about subunits, see inputList for details.
file.name	Optional: Full path of csv file for results.
verbose	logical. If TRUE additional diagnostics are printed.

Details

The column names of datRaw and datRec must be consistent with the names provided by the columns subunit and subunitRecoded in data.frame subunits. Otherwise, catPbc will fail.

Value

A data frame with the discrimination values for each category of categorical variables. The data frame contains the following columns:

item	Name of unrecoded item
cat	Name of category
n	Number of responses for this item
freq	Absolute frequency of the category
freq.rel	Relative frequency of the category
catPbc	Discrimination value for the category (correlation with total score)
recodevalue	Recode value for the category
subunitType	Type of subunit, see inputList for details

Author(s)

Nicole Haag

Examples

data(inputDat)
data(inputList)

datRaw <- mergeData(newID = "ID", datList = inputDat, addMbd = TRUE)
datRec <- recodeData(datRaw, values = inputList$values, 
    subunits=inputList$subunits)
pbcs   <- catPbc(datRaw, datRec, idRaw = "ID", idRec = "ID", 
    context.vars = "hisei", values = inputList$values, 
    subunits = inputList$subunits)

eatPrep documentation built on May 2, 2019, 5:20 p.m.