Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items
In ordinal data analysis, category collapse is the process of combining adjacent response options to create fewer response categories than were originally measured. When collapsing response categories, researchers need to be aware of inducing data-model misfit and of obtaining biased parameter estim...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PsychOpen GOLD/ Leibniz Institute for Psychology
2025-03-01
|
| Series: | Methodology |
| Subjects: | |
| Online Access: | https://doi.org/10.5964/meth.14303 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850198946134097920 |
|---|---|
| author | Yale Quan Chun Wang |
| author_facet | Yale Quan Chun Wang |
| author_sort | Yale Quan |
| collection | DOAJ |
| description | In ordinal data analysis, category collapse is the process of combining adjacent response options to create fewer response categories than were originally measured. When collapsing response categories, researchers need to be aware of inducing data-model misfit and of obtaining biased parameter estimates. Through mathematical derivation we show that category collapse induces data-model misfit when using Generalized Partial Credit IRT model (GPCM) generated data. This data-model misfit is not present when using Graded Response IRT model (GRM) generated data. Using simulation studies, we found that category collapse can indicate better data-model fit in GRM- and GPCM-generated data. In the case of GPCM data, this result is spurious and can lead practitioners to draw conclusions from models that do not fit the data well. Recovered GPCM IRT item parameters were also significantly biased. Recommendations for practitioners who wish to collapse categories are provided. |
| format | Article |
| id | doaj-art-e4ff8e987eb74464b6c6e35fcd024bea |
| institution | OA Journals |
| issn | 1614-2241 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | PsychOpen GOLD/ Leibniz Institute for Psychology |
| record_format | Article |
| series | Methodology |
| spelling | doaj-art-e4ff8e987eb74464b6c6e35fcd024bea2025-08-20T02:12:45ZengPsychOpen GOLD/ Leibniz Institute for PsychologyMethodology1614-22412025-03-01211467310.5964/meth.14303meth.14303Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous ItemsYale Quan0https://orcid.org/0000-0002-8582-7749Chun Wang1https://orcid.org/0000-0003-2695-9781College of Education, University of Washington, Seattle, WA, USACollege of Education, University of Washington, Seattle, WA, USAIn ordinal data analysis, category collapse is the process of combining adjacent response options to create fewer response categories than were originally measured. When collapsing response categories, researchers need to be aware of inducing data-model misfit and of obtaining biased parameter estimates. Through mathematical derivation we show that category collapse induces data-model misfit when using Generalized Partial Credit IRT model (GPCM) generated data. This data-model misfit is not present when using Graded Response IRT model (GRM) generated data. Using simulation studies, we found that category collapse can indicate better data-model fit in GRM- and GPCM-generated data. In the case of GPCM data, this result is spurious and can lead practitioners to draw conclusions from models that do not fit the data well. Recovered GPCM IRT item parameters were also significantly biased. Recommendations for practitioners who wish to collapse categories are provided.https://doi.org/10.5964/meth.14303category collapseitem response theorygeneralized partial credit modelparameter recoverydata-model fit |
| spellingShingle | Yale Quan Chun Wang Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items Methodology category collapse item response theory generalized partial credit model parameter recovery data-model fit |
| title | Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items |
| title_full | Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items |
| title_fullStr | Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items |
| title_full_unstemmed | Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items |
| title_short | Collapsing or Not? A Practical Guide to Handling Sparse Responses for Polytomous Items |
| title_sort | collapsing or not a practical guide to handling sparse responses for polytomous items |
| topic | category collapse item response theory generalized partial credit model parameter recovery data-model fit |
| url | https://doi.org/10.5964/meth.14303 |
| work_keys_str_mv | AT yalequan collapsingornotapracticalguidetohandlingsparseresponsesforpolytomousitems AT chunwang collapsingornotapracticalguidetohandlingsparseresponsesforpolytomousitems |