Self-Assessment and Motivational Beliefs: Applying CSE Scales into a Chinese Tertiary EFL Self-Regulated Learning Course
Description
Data are collected with four instruments: (a) a motivational beliefs questionnaire; (b) a set of self-assessment scales adapted from the CSE listening self-assessment scale; (c) two sets of CET-4 (College English Test-Band 4) for a pretest and a posttest; (d) a semi-structured interview. The study was conducted in an online College English listening course supported by the Unipus Intelligent Teaching and Learning Cloud Platform. The participants were 230 Chinese undergraduates from Forestry, Computer Science, and Commerce majors. This study adopted an eight-week quasi-experiment and compared the performance of two groups: an experiment group that received self-assessment intervention and a control group without intervention. Three researchers develop CSE-based self-assessment scales, send them in the Wechat group, and provide instant feedback when students have problems. All the participants were recruited voluntarily after their receiving a letter of consent, and they were informed of their rights to withdraw from the research at any time during or after the data collection. 181 valid participants completed all processes within eight weeks’ experiment (NFemale=116, NMale=65), with the experiment group being 85 and the control group being 96.
Files
Steps to reproduce
The study initially conducted a preliminary test on the listening self-assessment scale, by which we defined students’ proficiency levels on a scale of 4-6. During the pre-test and post-test, all participants were invited to complete a CET-4 listening test while using the refined self-assessment scale. Then a motivational belief questionnaire followed. In the eight-week intervention, the experiment group was required to finish the self-assessment scale as soon as they complete the listening task each week, while the control group received the usual listening courses. To obtain qualitative data, students were recruited to participate in the follow-up interview. Multiple regression analysis was used to explore the predictive effect of CSE self-assessment scales on motivational beliefs. A regression model was proposed: the dependent variable is the change of three motivational beliefs, including the change of task value beliefs, the change of control beliefs, and the change of test anxiety. Independent variables were the changes of the three factors of the CSE self-assessment scale, namely LG changes, LD changes and CSI changes. Besides, simultaneous regression, where all variables were added to the regression equation, was applied. Assumptions of multiple regression analysis were tested from the aspects of sample size, linear relationship, multiple normality, multicollinearity and homology difference. Homoscedasticity. Correlations of predictors (independent variables) were examined for checking multicollinearity. Lastly, the interview transcripts were analyzed to supplement the results of quantitative data.