Abstract:
Primary and secondary students, like teachers, need to develop their assessment literacy through experience and explicit teaching encompassing students’ conceptualised perceptions, attitudes, skills and knowledge in developing strategies to engage in assessment and reflective practice, monitor learning, and develop affective skills. Therefore, determining students’ current state of and building their assessment literacy is critical for improving overall performance and engagement, including empowering students to develop autonomy, agency and the necessary assessment related skills to change the possibilities of their own world.
However, at present there is no assessment literacy tool that can be used to measure students’ assessment literacy. We aim to fill this gap by drawing on theoretical and factor analytic approaches to establish the tool’s psychometric properties to support its reliability and validity.
In Stage 1, we build on our previous work about the student assessment literacy framework and the results of teacher interviews (n=10) to identify individual items that comprise students’ assessment literacy with a five-point Likert scale (poor to excellent). We coded fifteen articles accessed from three databases (ProQuest, Eric, and Scopus) using combinations of key terms. The codes were determined to the two extant definitions of student assessment literacy. The codes were verified from the analysis of teacher interview data (n=10), explicitly identifying specific students’ dispositions, knowledge, and skills in assessment. Results show that there are 38 items clustered around six dimensions: general knowledge of assessment, development of strategies to engage in assessment, active engagement in assessment, monitoring learning progress, engagement in reflective practice, and disposition in assessment. The tool has undergone an expert validation process (n=3) to establish language clarity and inclusion of a range of, and accurate items. The theoretical dimensions are currently being tested in Stage 2. Pilot testing is underway using a generalised partial credit model to establish the fit measures, including reliability and construct validity. Student participants are recruited for the final phase. Both exploratory and confirmatory factor analyses will be conducted to establish the dimensionality of the construct. The weighted least squares means and variance adjusted estimation will be used in both analyses. Several fit indices will be used to evaluate the model fit including chi-square statistics, Root Mean Square Error of Approximation, Standardized Root Mean Square Residual, Comparative Fit Index, Tucker Lewis Index, and the Weighted Root Mean Square Residual, following the conventional cut-off values.
The results of our study can facilitate address any inconsistencies between teachers’ and school leaders’ expectations and understanding of students’ assessment literacy, thereby providing support for students for more effective learning.
However, at present there is no assessment literacy tool that can be used to measure students’ assessment literacy. We aim to fill this gap by drawing on theoretical and factor analytic approaches to establish the tool’s psychometric properties to support its reliability and validity.
In Stage 1, we build on our previous work about the student assessment literacy framework and the results of teacher interviews (n=10) to identify individual items that comprise students’ assessment literacy with a five-point Likert scale (poor to excellent). We coded fifteen articles accessed from three databases (ProQuest, Eric, and Scopus) using combinations of key terms. The codes were determined to the two extant definitions of student assessment literacy. The codes were verified from the analysis of teacher interview data (n=10), explicitly identifying specific students’ dispositions, knowledge, and skills in assessment. Results show that there are 38 items clustered around six dimensions: general knowledge of assessment, development of strategies to engage in assessment, active engagement in assessment, monitoring learning progress, engagement in reflective practice, and disposition in assessment. The tool has undergone an expert validation process (n=3) to establish language clarity and inclusion of a range of, and accurate items. The theoretical dimensions are currently being tested in Stage 2. Pilot testing is underway using a generalised partial credit model to establish the fit measures, including reliability and construct validity. Student participants are recruited for the final phase. Both exploratory and confirmatory factor analyses will be conducted to establish the dimensionality of the construct. The weighted least squares means and variance adjusted estimation will be used in both analyses. Several fit indices will be used to evaluate the model fit including chi-square statistics, Root Mean Square Error of Approximation, Standardized Root Mean Square Residual, Comparative Fit Index, Tucker Lewis Index, and the Weighted Root Mean Square Residual, following the conventional cut-off values.
The results of our study can facilitate address any inconsistencies between teachers’ and school leaders’ expectations and understanding of students’ assessment literacy, thereby providing support for students for more effective learning.