Questions tagged [item-response-theory]
Item Response Theory (IRT) is a set of statistical methods for modeling responses to test items. Use this tag for on-topic questions about (a) the theory itself (b) models appropriate for a specific case (c) implementing IRT in a specific tool such as an R package.
277 questions
3
votes
3
answers
290
views
class imbalance in ML based IRT models
I’m testing an IRT model on large datasets, but running into class imbalance. most responses fall into the “correct” category. It is quite normal in IRT models I suppose as assessment data will be ...
- 47
2
votes
1
answer
96
views
Software for multigroup Rasch measurement with latent change, SEM regression and clustered data
I’m working on a longitudinal IRT project in which we want to:
Estimate student growth between two measurement points on a common latent‐ability scale (e.g. two grade levels).
Compare average growth ...
- 61
3
votes
1
answer
78
views
In multidimensional item response theory, why are graphs always in an orthogonal coordinate system even when the traits are correlated?
In multidimensional item response theory, the item characteristic curve from unidimensional models is replaced with an item characteristic surface.
To graphicaly plot this, we have axes x and y, ...
- 609
6
votes
3
answers
345
views
Low slope/discrimination questions: usefulness in IRT?
In Item Response Theory, a 2-PL model captures both slope and intercept, whereas the Rasch model, which only captures the intercept, shifting the curve left/right (captured below.)
The general ...
- 3,970
1
vote
0
answers
75
views
Which factor score estimation method for Rasch/2PL itemfit and reliability?
I want to calculate the item-fit values (infit and outfit) and the person separation reliability (PSR) for a Rasch and 2PL model using R. My goal is developing a reliable and good fitting scale that ...
- 95
0
votes
0
answers
41
views
How to measure unidimensionality using MNAR, binary data in R
So for my thesis I need my data to be unidimensional. I want to test the unidimensionality using CFA. However, my data has some issues that make a standard CFA difficult, as it is MNAR and binary. So ...
- 1
1
vote
0
answers
43
views
Using IRT for a non-proctored assessment [closed]
For the past couple of years, my research team has been administering an in-house multiple-choice knowledge assessment for students to take at the end of a college course. My boss is adamant that we ...
8
votes
1
answer
465
views
Why does Item Response Theory not depend on a representative sample, or why is it "sample-independent"?
Multiple resources describing Item Response Theory (IRT) and comparing it to Classical Test Theory (CTT) note that CTT depends on a representative sample whereas IRT does not.
Sample-independence of ...
- 1,464
2
votes
1
answer
77
views
Computing probabilities given overlapping intervals?
Suppose that you use an item response theory model to infer student ability and question difficulty on some training set. Now, we want to use this data for the purpose of dynamic testing where student ...
- 3,970
1
vote
1
answer
95
views
Dimension reduction to 1 ranked dimension
I have a dataset that contains a comprehensive questionnaire to 13,000 individuals containing about 40 questions regarding disability e.g., can I shower independently? Can I do groceries alone? Etc
...
- 1,136
4
votes
1
answer
268
views
Differential Item Functioning Criteria Selection
When testing differential item functioning (DIF) using a (ordered) logistic regression framework, a series of (ordered) logistic regressions are fit to the data, using item response ($Y$) as the ...
- 203
1
vote
1
answer
184
views
1
vote
1
answer
107
views
Item Response Theory (IRT) -- Problem (?) or unxepected result using the mirt package with a 2PL model for binary variables?
I'm working with 5 variables related to "child and family health" using Item Response Theory to check the model and identify its parameters. My model is composed of items such as "Do ...
- 244
3
votes
2
answers
164
views
Weighing Questions within a Survey
I work for a public child welfare agency. We have something called a level of care tool that categorizes children based on the amount of care they need. This helps us identify appropriate care options ...
- 33
2
votes
3
answers
319
views
IRT Nominal Response Model theta
I'm quite familiar with multiple IRT models; however, I can't seem to understand the Nominal Response Model. How is theta estimated? For example, if we have a science test with multiple-choice items (&...
- 21