Servicios Personalizados
Revista
Articulo
Indicadores
- Citado por SciELO
- Accesos
Links relacionados
- Similares en SciELO
Compartir
Actualidades en Psicología
versión On-line ISSN 2215-3535versión impresa ISSN 0258-6444
Resumen
PRIETO-ADANEZ, Gerardo. Analysis of a Writing Test with the MFRM Model. Act.Psi [online]. 2015, vol.29, n.119, pp.3-19. ISSN 2215-3535. http://dx.doi.org/10.15517/ap.v29i119.19822.
This paper describes how a Rasch model (Many-Facet Rasch Measurement) can be applied to performance assessment focusing on analysis of examinee, raters, tasks and variables. The article provides an introduction to MFRM, a description of analysis procedures, and an illustrative example to examine the effects of various sources of variability on students’ performance on a writing test by means of the FACETS program. Results highlight the usefulness of the MFRM to detect raters that have extreme values on the continuum of severity/leniency as well as providing objective measurement of examinee (scores free of rater severity).
Palabras clave : Rater-Mediated Assessment; Performance Assessment; Writing Assessment; Rasch Model; Many-Facet Rasch Measurement.