A closer look at a marginalized test method: Self-assessment as a measure of speaking proficiency

Abstract Second language (L2) teachers may shy away from self-assessments because of warnings that students are not accurate self-assessors. This information stems meta-analyses in which self-assessment scores on average did correlate highly with proficiency test results. However, researchers mostly used Pearson correlations, when polyserial could be used. Furthermore, today can computer adaptive. With them, nonlinear statistics needed to investigate their relationship other measurements. We wondered, if we explored the between and using more robust measurements (polyserial correlation, continuation-ratio modeling), would find different results? had 807 L2-Spanish learners take a computer-adaptive, L2-speaking ACTFL Oral Proficiency Interview – (OPIc). The correlated at .61 (polyserial). Using modeling, found each unit increase OPIc scale was associated 131% odds passing thresholds. In words, student likely move higher subsections they rating. computer-adaptive appropriate for low-stakes L2-proficiency measurements, especially cost-effective, make intuitive sense learners, promote learner agency.