Learn R Programming

⚠️There's a newer version (1.7.0) of this package.Take me there.

TestDesign (version 1.6.1)

Optimal Test Design Approach to Fixed and Adaptive Test Construction

Description

Uses the optimal test design approach by Birnbaum (1968, ISBN:9781593119348) and van der Linden (2018) to construct fixed, adaptive, and parallel tests. Supports the following mixed-integer programming (MIP) solver packages: 'Rsymphony', 'gurobi', 'lpSolve', and 'Rglpk'. The 'gurobi' package is not available from CRAN; see .

Copy Link

Version

Install

install.packages('TestDesign')

Monthly Downloads

470

Version

1.6.1

License

GPL (>= 2)

Maintainer

Seung W. Choi

Last Published

March 31st, 2024

Functions in TestDesign (1.6.1)

Calculate the Fisher information using full Bayesian

Calculate expected scores

Calculate second derivative of log-likelihood

Calculate Fisher information

calc_likelihood

(C++) For multiple items, calculate likelihoods

checkConstraints

Check the consistency of constraints and item usage

Calculate first derivative of log-likelihood

calcLocation-methods

Calculate central location (overall difficulty)

Calculate Relative Errors

Calculate a posterior value of theta

calc_posterior_function

Calculate a posterior distribution of theta

constraint-class

Class 'constraint': a single constraint

Compute expected a posteriori estimates of theta

Run adaptive test assembly

(C++) Classify theta values into segments using cutpoints

Calculate Root Mean Squared Error

detectBestSolver

Detect best solver

Open TestDesign app

Calculate alpha angles from a-parameters

config_Static-class

Create a config_Static object

config_Shadow-class

Create a config_Shadow object

Calculate the mutual information using full Bayesian

(C++) Calculate expected scores

dataset_reading

Reading dataset

makeItemPoolCluster

Create an item pool cluster object

calc_posterior_single

Calculate a posterior value of theta for a single item

(C++) For multiple items, calculate Fisher information

(C++) Calculate first derivative of log-likelihood

item_pool_cluster-class

Class 'item_pool_cluster': an item pool

output_Static-class

Class 'output_Static': fixed-form assembly solution

item_pool-operators

Basic operators for item pool objects

Convert mean and standard deviation into log-normal distribution parameters

Print solution items

Split an item pool into partitions

getSolutionAttributes

Retrieve constraints-related attributes from solution

calcLogLikelihood

Calculate log-likelihood

calcProb-methods

Calculate item response probabilities

item_attrib-operators

Basic functions for item attribute objects

item_pool-class

Class 'item_pool': an item pool

dataset_science

Science dataset

constraints-class

Class 'constraints': a set of constraints

st_attrib-class

Load set/stimulus/passage attributes

Convert mean and standard deviation into logit-normal distribution parameters

(C++) Calculate item response probability

constraints-operators

Basic operators for constraints objects

Compute maximum likelihood estimates of theta using fence items

makeSimulationDataCache

Create a simulation data cache object

loadConstraints

Load constraints

(C++) Calculate Fisher information

item_attrib-class

Load item attributes

(C++) Calculate second derivative of log-likelihood

Run Test Assembly

Extension of show() for objects in TestDesign package

output_Shadow-class

Class 'output_Shadow': adaptive assembly solution for one simulee

test_cluster-class

Class 'test_cluster': data cache for simulations

Create a test object

Extension of summary() for objects in TestDesign package

Class 'test': data cache for simulations

(C++) Calculate a theta estimate using EB (Empirical Bayes) method

(C++) Calculate a theta estimate using FB (Full Bayes) method

Extension of plot() for objects in TestDesign package

Extension of print() for objects in TestDesign package

dataset_fatigue

Fatigue dataset

Simulate item response data

iparPosteriorSample

Generate item parameter samples using standard errors

makeTestCluster

Create a test cluster object

output_Shadow_all-class

Class 'output_Shadow_all': a set of adaptive assembly solutions

Compute maximum likelihood estimates of theta

Basic operators for test objects

output_Split-class

Class 'output_Split': partitioning solution

summary-classes

Summary classes

st_attrib-operators

Basic functions for stimulus attribute objects

simulation_data_cache-class

Class 'simulation_data_cache': data cache for Shadow()

(C++) Calculate a theta estimate using EAP (expected a posteriori) method

toggleConstraints

Toggle constraints

Run fixed-form test assembly

Open TestDesign app

buildConstraints

Build constraints (shortcut to other loading functions)

Calculate the Fisher information using empirical Bayes