responseCurve: Produce data suitable for plotting item response curves

In pcFactorStan: Stan Models for the Paired Comparison Factor Model

Produce data suitable for plotting item response curves

Description

Selects samples random draws from the posterior and evaluates the item response curve on the grid given by seq(from,to,by). All items use the same responseNames. If you have some items with a different number of thresholds or different response names then you can call responseCurve for each item separately and rbind the results together.

Usage

responseCurve(
  dl,
  fit,
  responseNames,
  item = dl$nameInfo$item,
  samples = 100,
  from = qnorm(0.1),
  to = -from,
  by = 0.02
)

Arguments

dl	a data list prepared by prepData
fit	a stanfit object
responseNames	a vector of labels for the possible responses
item	a vector of item names
samples	number of posterior samples
from	the starting latent difference value
to	the ending latent difference value
by	the grid increment

Value

A data.frame with the following columns:

response

Which response

worthDiff

Difference in worth

item

Which item

sample

Which sample

prob

Associated probability

responseSample

A grouping index for independent item response samples

Response model

See cmp_probs for details.

See Also

Other data extractor: parDistributionCustom(), parInterval()

Examples

 vignette('manual', 'pcFactorStan') 

pcFactorStan documentation built on Sept. 14, 2023, 1:09 a.m.