Person misfit on the Inventory of Depressive Symptomatology: Low quality self-report or true atypical symptom profile?

Publication Type:

Journal Article

Source:

International Journal of Methods in Psychiatric Research, Volume 26, Issue 4 (2017)

DOI:

10.1002/mpr.1548

Keywords:

atypical depression symptoms; careless and random responding; item response theory; person‐fit analysis

Abstract:

Person misfit on a self‐report measure refers to a response pattern that is unlikely given a
theoretical measurement model. Person misfit may reflect low quality self‐report data, for
example due to random responding or misunderstanding of items. However, recent research in
the context of psychopathology suggests that person misfit may reflect atypical symptom profiles
that have implications for diagnosis or treatment. We followed‐up on Wanders et al. (Journal of
Affective Disorders, 180, 36–43, 2015) who investigated person misfit on the Inventory of
Depressive Symptomatology (IDS) in the Netherlands Study of Depression and Anxiety
(n = 2,981). Our goal was to investigate the extent to which misfit on the IDS reflects low‐quality
self‐report patterns and the extent to which it reflects true atypical symptom profiles. Regression
analysis showed that person misfit related more strongly to self‐report quality indicators than to
variables quantifying theoretically‐derived atypical symptom profiles. A data‐driven atypical
symptom profile explained most variance in person misfit, suggesting that person misfit on the
IDS mainly reflects a sample‐ and questionnaire‐specific atypical symptom profile. We concluded
that person‐fit statistics are useful for detecting IDS scores that may not be valid. Further
research is necessary to support the interpretation of person misfit as reflecting a meaningful
atypical symptom combination.