Articles producció científica> Psicologia

Assessing retest effects at the individual level: A general IRT-based approach

  • Dades identificatives

    Identificador: imarina:9297983
    Autors:
    Lorenzo-Seva, Urbano
    Resum:
    Test-retest studies for assessing stability and change are widely used in different domains and allow improved or additional individual estimates of interest to be obtained. However, if these estimates are to be validly interpreted the responses given at Time-2 must be free of retest effects, and the fulfilment of this assumption must be empirically checked. This article proposes a comprehensive item response theory-based approach for assessing retest effects at the individual level and test the assumption of local independence under repetition. The approach can be used with a wide array of unidimensional and multidimensional models, and is based on correlation-type and mean-square-type indices. Procedures for (a) establishing critical values for detection purposes and (b) interpreting the magnitude of the retest effects for the detected respondents are also proposed. Furthermore, the article discusses the consequences of not addressing retest effects in stability and change studies. The procedures were assessed with simulation and used in three empirical studies. In all cases they worked well and provided meaningful information.
  • Altres:

    Autor segons l'article: Lorenzo-Seva, Urbano
    Departament: Psicologia
    Autor/s de la URV: Ferrando Piera, Pere Joan
    Paraules clau: Tests Selection Personality-assessment Performance Model Item response theory Dependence Bootstrap confidence-intervals
    Resum: Test-retest studies for assessing stability and change are widely used in different domains and allow improved or additional individual estimates of interest to be obtained. However, if these estimates are to be validly interpreted the responses given at Time-2 must be free of retest effects, and the fulfilment of this assumption must be empirically checked. This article proposes a comprehensive item response theory-based approach for assessing retest effects at the individual level and test the assumption of local independence under repetition. The approach can be used with a wide array of unidimensional and multidimensional models, and is based on correlation-type and mean-square-type indices. Procedures for (a) establishing critical values for detection purposes and (b) interpreting the magnitude of the retest effects for the detected respondents are also proposed. Furthermore, the article discusses the consequences of not addressing retest effects in stability and change studies. The procedures were assessed with simulation and used in three empirical studies. In all cases they worked well and provided meaningful information.
    Àrees temàtiques: Statistics and probability Psychology, experimental Psychology (miscellaneous) Psicología Experimental and cognitive psychology Developmental and educational psychology Ciencias sociales Ciencias humanas Arts and humanities (miscellaneous)
    Accès a la llicència d'ús: https://creativecommons.org/licenses/by/3.0/es/
    Adreça de correu electrònic de l'autor: perejoan.ferrando@urv.cat
    Identificador de l'autor: 0000-0002-3133-5466
    Data d'alta del registre: 2024-10-12
    Versió de l'article dipositat: info:eu-repo/semantics/publishedVersion
    Enllaç font original: https://psicologicajournal.com/assessings-retests-effects-at-the-individual-level-a-general-ir-based-approach/
    URL Document de llicència: https://repositori.urv.cat/ca/proteccio-de-dades/
    Referència a l'article segons font original: Psicologica. 36 (1): 143-163
    Referència de l'ítem segons les normes APA: Lorenzo-Seva, Urbano (2015). Assessing retest effects at the individual level: A general IRT-based approach. Psicologica, 36(1), 143-163
    Entitat: Universitat Rovira i Virgili
    Any de publicació de la revista: 2015
    Tipus de publicació: Journal Publications
  • Paraules clau:

    Arts and Humanities (Miscellaneous),Developmental and Educational Psychology,Experimental and Cognitive Psychology,Psychology (Miscellaneous),Psychology, Experimental,Statistics and Probability
    Tests
    Selection
    Personality-assessment
    Performance
    Model
    Item response theory
    Dependence
    Bootstrap confidence-intervals
    Statistics and probability
    Psychology, experimental
    Psychology (miscellaneous)
    Psicología
    Experimental and cognitive psychology
    Developmental and educational psychology
    Ciencias sociales
    Ciencias humanas
    Arts and humanities (miscellaneous)
  • Documents:

  • Cerca a google

    Search to google scholar