Does test-taking motivation predict test results in a high-stakes testing context?

Abstract

ABSTRACT Test-taking motivation (TTM) has been associated with test performance in low-stakes testing contexts. However, there have been few studies about TTM in high-stakes testing contexts, and these have contradictory results. Our aim was to explore the relationship between test-taking effort and test performance in a real-life high-stakes testing context (n = 1,515). We collected time-based and self-reported data about test-taking effort and used a structural equation model (SEM) to predict test performance. We found that the motivational indicators added about 15% of predictive power to the SEM model, where gender and previous performance had been controlled for. Altogether, the SEM model predicted 69% of the variance in test results. We compared the findings to previous studies and concluded that the possible effect of TTM should be considered in various testing contexts, whether low-stakes or high-stakes.