Your browser doesn't support javascript.
loading
Validation of the Spanish adaptation of the School Attitude Assessment Survey-Revised using multidimensional Rasch analysis / Validación de la adaptación española del School Attitude Assessment Survey-Revised mediante el modelo de Rasch multidimensional
Veas, Alejandro; Castejón, Juan Luis; Gilar, Raquel; Miñano, Pablo.
Affiliation
  • Veas, Alejandro; University of Alicante. Alicante. Spain
  • Castejón, Juan Luis; University of Alicante. Alicante. Spain
  • Gilar, Raquel; University of Alicante. Alicante. Spain
  • Miñano, Pablo; University of Alicante. Alicante. Spain
An. psicol ; 33(1): 74-81, ene. 2017. ilus, tab, graf
Article in English | IBECS | ID: ibc-159590
Responsible library: ES1.1
Localization: BNCS
ABSTRACT
The School Attitude Assessment Survey-Revised (SAAS-R) was developed by McCoach and Siegle (2003b) and validated in Spain by Miñano, Castejón, and Gilar (2014) using Classical Test Theory. The objective of the current research is to validate SAAS-R using multidimensional Rasch analysis. Data were collected from 1398 students attending different high schools. Principal Component Analysis supported the multidimensional SAAS-R. The item difficulty and person ability were calibrated along the same latent trait scale. 10 items were removed from the scale due to misfit with the Rasch model. Differential Item Functioning revealed no significant differences across gender for the remaining 25 items. The 7- category rating scale structure did not function well, and the subscale goal valuation obtained low reliability values. The multidimensional Rasch model supported 25 item-scale SAAS-R measures from five latent factors. Therefore, the advantages of multidimensional Rasch analysis are demonstrated in this study
RESUMEN
Validación de la adaptación española del School Attitude Assessment Survey-Revised mediante el modelo de Rasch multidimensional. Resumen El School Attitude Assessment Survey-Revised (SAAS-R) fue desarrollado por McCoach y Siegle (2003b) y validado en España por Mi- ñano, Castejón, y Gilar (2014) a través del Modelo Clásico de Test. El objetivo del presente estudio es validar el SAAS-R a partir del análisis de Rasch Multidimensional. Los datos se obtuvieron de 1398 estudiantes que asistían a diferentes institutos de Educación Secundaria. El Análisis de Componentes Principales apoyó el modelo Rasch multidimensional. Se calibraron los parámetros de dificultad de los ítems y habilidad de los sujetos a partir de la misma escala latente. Se eliminaron 10 ítems por mostrar desajuste al modelo de Rasch. El Funcionamiento Diferencial del Ítem no mostró diferencias significativas de género con los 25 ítems restantes. La estructura escalar de 7 categorías no mostró un funcionamiento óptimo, y la subescala Valoración de Logro obtuvo niveles bajos de fiabilidad. El modelo Rasch multidimensional apoyó la escala SAAS-R con 25 ítems y 5 factores latentes. De esta forma, se demuestran las ventajas del modelo de Rasch multidimensional en el presente estudio
Subject(s)

Full text: Available Collection: National databases / Spain Database: IBECS Main subject: Psychological Tests / Psychometrics / Attitude Type of study: Prognostic study Limits: Humans Language: English Journal: An. psicol Year: 2017 Document type: Article Institution/Affiliation country: University of Alicante/Spain

Full text: Available Collection: National databases / Spain Database: IBECS Main subject: Psychological Tests / Psychometrics / Attitude Type of study: Prognostic study Limits: Humans Language: English Journal: An. psicol Year: 2017 Document type: Article Institution/Affiliation country: University of Alicante/Spain
...