Skip to contents

Person fit

Source: R/easyRasch.R
RIpfit.Rd

Outputs a histogram of person fit ZSTD and a plot with person fit ZSTD and person location/score. Defaults to output a histogram and a hex heatmap.

Usage

RIpfit(
  dfin,
  model = "PCM",
  pointsize = 2.5,
  alpha = 0.5,
  bins = 30,
  group,
  output = c("hist", "heatmap"),
  infit_lim = c(-1.96, 1.96)
)

Arguments

dfin

Dataframe with item data only

model

Rasch model to use, "PCM" or "RM"

pointsize

Size of datapoints for grouped view

alpha

Transparency of points (0-1 where 1 = not transparent)

bins

Number of bins for hexplot

group

Optional grouping variable

output

Can also be "rowid" for a dataframe with rownumbers

infit_lim

Lower/upper limit for person infit ZSTD

Details

Optional grouped output with colorized points.

You can also get a vector with row numbers for persons with infit ZSTD over/under +/- 1.96 by using output = "rowid". Or the full dataframe with all respondents infit ZSTD and estimated theta values with output = "dataframe.

If you desire another cutoff than +/- 1.96, it can be set with infit_lim.

Note: theta estimation is done using ML, which is not optimal but should be sufficient for this analysis.