A lightweight julia package providing basic implementations of item response models
MIT License
ItemResponseFunctions.jl implements basic functions for Item Response Theory models. It is built based on the interface designed in AbstractItemResponseModels.jl.
You can install ItemResponseFunctions.jl from the General package registry:
] add ItemResponseFunctions
ItemResponseFunctions.jl exports the following functions for Item Response Theory models:
irf
: The item response functioniif
: The item information functionexpected_score
: The expected score / test response functioninformation
: The test information functionCalling the function requires a model type M
, a person ability theta
and item parameters beta
.
For a simple 1-Parameter Logistic model,
using ItemResponseFunctions
beta = (; b = 0.5)
irf(OnePL, 0.0, beta, 1)
iif(OnePL, 0.0, beta, 1)
evaluates the item response function and item information function for response y
at ability value 0.0
for an item with difficulty 0.5
.
Given an array of item parameters (a test) and an ability value, the test response function and test information can be calculated by
betas = [
(; b = -0.3),
(; b = 0.25),
(; b = 1.0),
]
expected_score(OnePL, 0.0, betas)
information(OnePL, 0.0, betas)