This study reports a multitrait-multimethod analysis of evidence relating to the validity of reporting on subdomains in science from national tests at age 14 in England and Wales in 1993. The data support the policy decision to report only a single test level. A similar analysis (multitrait-multiform) is also described, which may make this approach feasible in a wider range of operational testing contexts and which confirmed the conclusions reached.
Get full access to this article
View all access options for this article.
References
1.
Campbell, D. T. , & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
2.
Hoste, R. (1980). The construct validity of some CSE biology examinations: The evidence from multitrait-multimethod matrices. Journal of Biological Education, 14, 41-48.
3.
Kendall, M. G. (1948). Rank correlation methods, London: Griffin.
4.
Massey, A. J. , & Foulkes, J. (1994). Audit of the 1993 KS3 science national test pilot and the concept of quasi-reconciliation. Evaluation and Research in Education, 8, 119-132.
5.
Nuttall, D. L. , & Stobart, G. (1994). National curriculum assessment in the U.K.Educational Measurement: Issues and Practice, 13, 24-27, 39.
6.
School Assessment & Examinations Council . (1993). 1994 KS3 school assessment folder. London: HMSO.
7.
Wood, R. (1991). Assessment and testing: A survey of research. Cambridge, UK: Cambridge University Press.
