Validity of personality measurement in adults with anxiety disorders: Psychometric properties of the Spanish NEO-FFI-R using Rasch analyses

  1. Inchausti, F. 14
  2. Mole, J. 2
  3. Fonseca-Pedrero, E. 3
  1. 1 Complejo Hospitalario Universitario de Badajoz, Spain
  2. 2 University of Oxford
    info

    University of Oxford

    Oxford, Reino Unido

    ROR https://ror.org/052gg0110

  3. 3 Universidad de La Rioja
    info

    Universidad de La Rioja

    Logroño, España

    ROR https://ror.org/0553yr311

  4. 4 Universidad de Salamanca
    info

    Universidad de Salamanca

    Salamanca, España

    ROR https://ror.org/02f40zc51

Revista:
Frontiers in Psychology

ISSN: 1664-1078

Año de publicación: 2015

Volumen: 6

Número: MAR

Tipo: Artículo

DOI: 10.3389/FPSYG.2015.00465 SCOPUS: 2-s2.0-84926676709 WoS: WOS:000353691600002 GOOGLE SCHOLAR

Otras publicaciones en: Frontiers in Psychology

Repositorio institucional: lock_openAcceso abierto Editor

Resumen

The aim of this study was to analyze the psychometric properties of the Spanish NEOFFI- R using Rasch analyses, in order to test its rating scale functioning, the reliability of scores, internal structure, and Differential Item Functioning (DIF) by gender in a psychiatric sample. The NEO-FFI-R responses of 433 Spanish adults (154 males) with an anxiety disorder as primary diagnosis were analyzed using the Rasch model for rating scales. Two intermediate categories of response ('neutral' and 'agree') malfunctioned in the Neuroticism and Conscientiousness scales. In addition, model reliabilities were lower than expected in Agreeableness and Neuroticism, and the item fit values indicated each scale had items that did not achieve moderate to high discrimination on its dimension, particularly in the Agreeableness scale. Concerning unidimensionality, the five NEO-FFI-R scales showed large first components of unexplained variance. Finally, DIF by gender was detected in many items. The results suggest that the scores of the Spanish NEO-FFI-R are unreliable in psychiatric samples and cannot be generalized between males and females, especially in the Openness, Conscientiousness, and Agreeableness scales. Future directions for testing and refinement should be developed before the NEO-FFI-R can be used reliably in clinical samples. © 2015 Inchausti.