Comparison of SEM, IRT and RMT-based methods for response shift detection at item level: a simulation study - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Statistical Methods in Medical Research Année : 2020

Comparison of SEM, IRT and RMT-based methods for response shift detection at item level: a simulation study

Résumé

When assessing change in patient-reported outcomes, the meaning in patients' self-evaluations of the target construct is likely to change over time. Therefore, methods evaluating longitudinal measurement non-invariance or response shift (RS) at item-level were proposed, based on structural equation modelling (SEM) or on item response theory (IRT). Methods coming from Rasch Measurement Theory (RMT) could also be valuable. The lack of evaluation of these approaches prevents determining the best strategy to adopt. A simulation study was performed to compare and evaluate the performance of SEM, IRT and RMT approaches for item-level RS detection.Performances of these three methods in different situations were evaluated with the rate of false detection of RS (when RS was not simulated) and the rate of correct RS detection (when RS was simulated).The RMT-based method performs better than the SEM and IRT-based methods when recalibration was simulated. Consequently, the RMT-based approach should be preferred for studies investigating only recalibration RS at item-level. For SEM and IRT, the low rates of reprioritization detection raise issues on the potential different meaning and interpretation of reprioritization at item-level.
Fichier principal
Vignette du fichier
author accepted manuscript.pdf (1000.4 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02318621 , version 1 (17-10-2019)

Identifiants

Citer

Myriam Blanchin, Alice Guilleux, Jean-Benoit Hardouin, Véronique Sébille. Comparison of SEM, IRT and RMT-based methods for response shift detection at item level: a simulation study. Statistical Methods in Medical Research, 2020, 29 (4), pp.1015-1029. ⟨10.1177/0962280219884574⟩. ⟨hal-02318621⟩
211 Consultations
715 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More