Patrones de correlación entre medidas de rendimiento escolar en evaluaciones longitudinalesun estudio de simulación desde un enfoque multinivel

  1. Blanco Blanco, Angeles
  2. González Barberá, Coral
  3. Ordóñez Camacho, Xavier G.
Journal:
Revista de educación

ISSN: 0034-8082

Year of publication: 2009

Issue Title: El valor añadido en educación

Issue: 348

Pages: 195-216

Type: Article

More publications in: Revista de educación

Abstract

Value-added models are applied in longitudinal data structures that raise important statistical and psychometrical methodological challenges. In this context, the aim of the study was to explain the differential correlation patterns of observed, true and latent scores over time. In particular, the effect of the trait complexity factor and the second (student) and third (school) levels of the residual covariate patterns was analyzed, from the point of view of the lineal hierarchical models with repeated measurements. Observed, true and latent scores for 25000 subjects were generated using a simulation, in four consecutive assessments and six different conditions.The first result is a description of the correlation pattern in the latent scores when the unidimensionality of the trait is assumed.This pattern is different to the one usually observed in empirical studies.Thus, the magnitude of the correlation between two measures increases as the measures distance to the starting point increases.This pattern can be explained as the result of the effect of the two referred factors. From the results two main findings are proposed. First, the need of understanding of the correlation patterns in the latent variables level to explain the relations in the empirical level. Second, the correlation patterns of achievement measures in educational longitudinal assessment are mainly determined by the growing complexity of the trait

Bibliographic References

  • BRAUN, H. I. (2005). Using student progress to evaluate teachers: a primer on value-added models. Princenton, NJ: Educational Testing Service.
  • BHASKAR, R. (1978). A realist theory of science. Harvester Press: Sussex.
  • CASTRO, M., RUÍZ, C. Y LÓPEZ, E. (2009). Forma básica del crecimiento en los modelos de valor añadido: vías para la supresión del efecto de regresión y funciones de crecimiento no lineales. Revista de Educación, 348.
  • FERRAÕ, E. (2009). Sensibilidad de las especificaciones de los modelos de valor añadido: la medida del estatus socio-económico. Revista de Educación, 348.
  • HOUSE, E. (1991). Realism in Research. Educational Researcher, 20 (6), 2-9.
  • IHAKA, R. Y GENTLEMAN, R. (1996). R: a language for data analysis and graphics. Journal of Computational and Graphical Statistics, 5, 299-314.
  • LIZASOAÍN, L. Y JOARISTI, L. (2009). Análisis de la dimensionalidad en los modelos de valor añadido: estudio de las pruebas de matemáticas empleando técnicas factoriales y métodos no paramétricos basados en TRI. Revista de Educación, 348.
  • MARTÍNEZ ARIAS, R. (2009). Usos, aplicaciones y problemas de los modelos de valor añadido en educación. Revista de Educación, 348.
  • MARTÍNEZ ARIAS, R., GAVIRIA, J. L. Y CASTRO, M. (2009). Concepto y evolución de los modelos de valor añadido en educación. Revista de Educación, 348.
  • MCCAFFREY, D. L., LOCKWOOD, J. R., KORETZ, D. M. Y HAMILTON, L. S. (2003). Evaluating value-added models for teacher accountability. Santa Mónica, CA: RAND Corporation.
  • RASCH, G. (1960). Probabilistic models for some intelligence and attainment tests. Copenhagen: Danish Institute for Educational Research
  • SANDERS, W. L. (2006, October 16). Comparisons among various educational Assessment Value-Added Models. Paper presented at The Power of Two-National Value-Added Conference, Columbus, Ohio.
  • SINGER, J. D. Y WILLET, J. B. (2003). Applied Longitudinal Data Analysis. Modeling Change and Event Occurrrence. Oxford: Oxford University Press.
  • WILEY, E. V. (2006). A practitioner's guide to value added assessment. Tempe, AZ: Educational Policy Studies Laboratory, Arizona State University.