[Rasch] Reversed Thresholds in Partial Credit Models: A Reason for Collapsing Categories?
srmillis at yahoo.com
Sat Nov 8 01:54:37 EST 2014
Assessment. 2014 Apr 30. [Epub ahead of print]
Reversed Thresholds in Partial Credit Models: A Reason for Collapsing Categories?
Wetzel E(1), Carstensen CH.
Author information: (1)Otto-Friedrich-University Bamberg, Germany.
When questionnaire data with an ordered polytomous response format are analyzedin the framework of item response theory using the partial credit model or thegeneralized partial credit model, reversed thresholds may occur. This led to the discussion of whether reversed thresholds violate model assumptions and indicate disordering of the response categories. Adams, Wu, and Wilson showed thatreversed thresholds are merely a consequence of low frequencies in the categoriesconcerned and that they do not affect the order of the rating scale. This articleapplies an empirical approach to elucidate the topic of reversed thresholds usingdata from the Revised NEO Personality Inventory as well as a simulation study. Itis shown that categories differentiate between participants with different trait levels despite reversed thresholds and that category disordering can be analyzed independently of the ordering of the thresholds. Furthermore, we show thatreversed thresholds often only occur in subgroups of participants. Thus,researchers should think more carefully about collapsing categories due toreversed thresholds.
PMID: 24789857 [PubMed - as supplied by publisher]
Scott R Millis, PhD, ABPP, CStat, PStat®
Wayne State University School of Medicine
Email: aa3379 at wayne.edu
Email: srmillis at yahoo.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch