In talbano/epmr: Tools for Educational and Psychological Measurement

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

The epmr package supports educational and psychological measurement in R, with methods and example data for psychometric analysis of results from standardized tests and assessments. This vignette summarizes the main features of the package. For a thorough demonstration, see the associated book at https://thetaminusb.com/intro-measurement-r/.

Overview

The epmr package is organized by type of analysis. Here's a list of the main functions, which all have the suffix study after an abbreviation of the type of analysis in the first few letters.

• astudy is for examining interrater agreement via percentage agreement, kappa, and weighted kappa statistics.
• difstudy reports differential item functioning results for a set of items.
• dstudy returns descriptive statistics for one or more variables in a data frame.
• fastudy is a simple wrapper for exploratory factor analysis with factanal.
• gstudy runs generalizability studies using multilevel modeling with the lme4 package.
• irtstudy currently runs the Rasch item reponse theory model using the lme4 package.
• ostudy is for option analysis, also called distractor analysis, with unscored, multiple choice item response data.
• rstudy estimates tau-equivalent and congeneric reliability, also called coefficients alpha and omega, as well as standard error of measurement.

Helper functions are also provided for

• plotting results from item response theory modeling using ggplot2 with irtplot
• rescaling variables to have a certain range, mean, and standard deviation with setmin, setmax, setmean, and setsd
• recoding variables with rescore, formerly named recode which conflicted with the dplyr function by the same name.

Data

The data in epmr come from two studies involving testing data.

• PISA09 is a data frame containing demographic, cognitive, and noncognitive variables for a subset of students participating in the 2009 administration of the Programme for International Student Assessment (PISA).
• BDI is a list containing the correlation matrix, covariance matrix, and variable labels for an administration of the Beck Depression Inventory II, as reported in a research study by Whisman, Perez, and Ramel (see the help file for reference).

talbano/epmr documentation built on May 1, 2024, 11:10 a.m.