摘要：Formative assessment policies argue that student self-assessment of work products and processes is useful for raising academic performance. This view draws on self-regulation of learning theories about setting targets and evaluating progress against criteria as a basis for meta-cognitively informed improvement of learning outcomes. However, the reliability of assessment is necessary for the validity of assessment interpretations. Research into psychological processes underlying the human ability to self-evaluate work raises serious doubts about the quality of students’ judgments. It has been shown that novices (i.e., students) tend to lack sufficient knowledge by which to evaluate their own work appropriately and, when possessing that knowledge, they are likely not to make use of it. Additionally, psychological safety factors indicate that self-assessment can be compromised by interpersonal relations present in classroom environments. Furthermore, since progress in many educational domains is relatively ill-defined, it is difficult for learners, let alone instructors, to validly evaluate progress or status. A recent review of studies in elementary and secondary schooling (K-12) found that the correlation between self-ratings and teacher ratings, between self-estimates of performance and actual test scores, and between student and teacher rubric-based judgments tended to be positive, ranging from weak to moderate (i.e., values ranging from r ≈ .20 to .80), with few studies reporting correlations r > .60. However, these values were not consistent across student experience and academic proficiency, with much less accuracy seen with younger and less proficient students. This paper argues that, instead of treating student self-assessment as an assessment method, a more pedagogical approach is needed that treats self-assessment as an important student competence or skill. As such self-evaluation needs a developmentally appropriate curriculum that is developed incrementally over time through consistent, scaffolded training and practice. Hence, to improve student skills in self-assessment, research must examine how the underlying skills required for self-assessment map onto student cognitive capabilities and the task characteristics of the work students are likely to encounter as they progress through schooling. This progression (like those developed for other key competencies like literacy, numeracy, and critical thinking) would help teachers consistently implement developmentally appropriate versions of self-assessment, giving students opportunities to develop these life-long skills.
