Learn R Programming

TestGardener

R code for analyzing data from tests and rating scales.

January 9, 2024 A major cleanup of the code was required to solve the issue that the plot of the last item was greatly distorted.

March 15, 2024 A number of bugs fixed.

September 16, 2024 Bspline basis functions will return values of 0 if the order of the spline is equal to the number of basis functions. This is especially helpful for the elementary two-basis case that is the spline basis version of the classic nominal model used by Darrell Bock for modelling multi-option tests and rating scales.

Copy Link

Version

Install

install.packages('TestGardener')

Monthly Downloads

314

Version

3.3.6

License

GPL (>= 2)

Maintainer

James Ramsay

Last Published

October 14th, 2025

Functions in TestGardener (3.3.6)

Plot probability and surprisal curves for test or scale items.

Estimate the option probability and surprisal curves.

Quant_13B_problem_parmList

Parameter list for 24 items from the quantitative SweSAT subtest.

Initialize surprisal smoothing of choice data.

Simulate Choice Data from a Previous Analysis

Sensitivity_plot

Plots all the sensitivity curves for selected items or questions.

List vector containing numbers of options and boundaries.

Quant_13B_problem_key

Option information for the short form of the SweSAT Quantitative test.

Quant_13B_problem_infoList

Arclength or information parameter list for 24 items from the quantitative SweSAT subtest.

Quant_13B_problem_dataList

List of objects essential for an analysis of the abbreviated SweSAT Quantitative multiple choice test.

Functional principal components analysis of information curve

Plot the test information or scale curve in either two or three dimensions.

Plot the probability density function for a set of test scores

Entropy measures of inter-item dependency

Plot the score index index as a function of arc length.

chcemat_simulate

Simulate a test or scale data matrix.

Simulation Based Estimates of Error Variation of Score Index Estimates

scorePerformance

Calculate mean squared error and bias for a set of score index values from simulated data.

Image of the Test Tnformation Curve in 2 or 3 Dimensions

Analyses of Tests and Rating Scales using Information or Surprisal

Compute and plot a score density histogram and and curve.

Statistics for Multiple choice Tests, Rating Scales and Other Choice Data)

Ensure that estimated score index is global

Compute a Probability Density Function

Make a list object containing information required for analysis of choice data.

Compute results using arc length or information as the abscissa.

Values of a Functional Data Object Defining Surprisal Curves.

Plot expected test score as a function of score index

Compute the expected test score by substituting probability of choices for indicator variable 0-1 values. Binary items assumed coded as two choice items.

Objects resulting for assessing fit of surprisal matrix to surprisal data

Compute score density

Compute optimal scores

Smooth binned probability and surprisal values to make an ICC object.

Fit data with surprisal smoothing.

Compute the negative log likelihoods associated with a vector of score index values.

Plot item entropy curves for selected items or questions.

Plot item power curves for selected items or questions.

Plot a selection of fit criterion F functions and their first two derivatives.

Plotting probability and surprisal curves for an item

Compute the first and second derivatives of the negative log likelihoods

Quant_13B_problem_chcemat

Test data for 24 math calculation questions from the SweSAT data.

Analyze test or rating scale data defined in dataList.

Construct grid of 101 values of the fitting function