Validating the Conceptions of Assessment-III Scale in Canadian Preservice Teachers

Abstract

The purpose of this study was to test the validity of the Teachers' Conceptions of Assessment Scale III-Abridged Version (CoA-IIIA; Brown, 2006), a measure created, validated, and applied outside of North America, in a sample of Canadian preservice teachers (n = 436). This work is important because although we have long known that teachers' beliefs influence the way they teach, research also suggests that teachers' beliefs related to assessment influence the way they assess students. Confirmatory factor analysis of the Canadian data led to a solution slightly different than the New Zealand factor structure, with nine latent first-order variables, each with two or three measured indicators. Although the preservice teachers endorsed the purposes of assessment similarly to several other countries, the factor measuring assessment as “inaccurate” was endorsed more strongly than in previous research. Our discussion highlights the differences and similarities between the current results and previous studies with the CoA-IIIA.

Notes

¹ According to independent sample t tests, the 436 participants with full data did not differ significantly from the 158 participants who did not have full data on any measure of conceptions of assessment (t values ranged from /.2/ to /1.8/). Visual inspection of the data suggested that there was no systematic reason for the incomplete data, and thus we chose to exclude the 158 from the main analyses in order to obtain the standardized root mean square residual.

² We tested for latent mean differences between male and female participants and elementary and secondary preservice teachers by comparing an unconstrained model to one in which the factor loadings were constrained to be equal between the two groups in AMOS 20.0. Neither of the models provided an admissible solution possibly because the sample size for each group became too small (Jöreskog & Sörbom, 1984).