Sbinsmth.init: Initialize surprisal smoothing of choice data.

In TestGardener: Information Analysis for Test and Rating Scale Data

Initialize surprisal smoothing of choice data.

Description

This version of Sbinsmth.init() uses direct least squares smoothing of the surprisal values at bin centers to generate dependent variables for a model for the vectorized K by M-1 parameter matrix Bmat. The estimates of the surprisal curves are approximated using functions in the fda package.

Usage

  Sbinsmth.init(percntrnk, nbin, Sbasis, grbgvec, noption, chcemat)

Arguments

percntrnk	Percent rank values of sum score values, usually after jittering
nbin	The number of bins used to bin the choice data.
Sbasis	A bspline functional basis object for surprisal smoothing.
grbgvec	A logical vector of length n indicating whether or not the choice data for an item is added for missing of illigetimate choices.
noption	An integer vector indicating the number of options for each item, not including a possible added garbage option.
chcemat	An N by n matrix with each row containing the indices of the options chosen by a person.

Value

A list vector of length n, each element being a list vector containing objects necessary for surprisal smoothing.

Author(s)

Juan Li and James Ramsay

References

Ramsay, J. O., Li J. and Wiberg, M. (2020) Full information optimal scoring. Journal of Educational and Behavioral Statistics, 45, 297-315.

Ramsay, J. O., Li J. and Wiberg, M. (2020) Better rating scale scores with information-based psychometrics. Psych, 2, 347-360.

See Also

ICC_plot, Sbinsmth

TestGardener documentation built on Sept. 30, 2024, 9:35 a.m.