Self-regulated learning intervention on grammar achievement, self-regulated learning strategy use, and calibration accuracy among Vietnamese high school learners

Self-regulated learning (SRL) was documented as effective in enhancing academic performance and strategy use for school subjects but has yet to be investigated for grammar instruction in English as a foreign language (EFL) contexts, including Vietnam. Because most Vietnamese public high school students lack the self-regulatory ability and have problems with English grammar learning, SRL strategy training is essential, hypothesized to enhance grammar achievement and strategy use. Grounded on Bandura's (1986) social cognitive theory (SCT) and Nelson and Narens' (1990) metamemory theoretical framework (MMTF), an SRL intervention on grammar learning was designed. The study adopted a mixed-methods explanatory sequential approach in which the quantitative data collection and analysis preceded the qualitative phase and received primary importance. The quantitative phase examined the effects of an SRL intervention on Vietnamese public high school students' grammar achievement, SRL strategy use, and calibration accuracy (CA), measured at three-time points. Meanwhile, the qualitative phase explored the sources students use to make calibration judgments and their experience of participating in the experiment. Thirty Vietnamese tenth-grade students participated in a quasi-experiment. While the treatment (n = 15) received a five-week SRL intervention that introduced an SRL model and trained SRL strategies, the control (n = 15) received the traditional instruction. Repeated measures analyses showed that the intervention had a moderate within-subject effect on grammar achievement (p = .04) and a high within-subject effect on SRL strategy use (p = .00). Moderate interaction effects between time and group were found for grammar achievement (p = .03) and SRL scores (p = .03). However, no between-subject effect was found for grammar achievement (p = .58) and SRL scores (p = .55), although the treatment scored higher than the control on these outcomes. The Friedman tests showed a small within-subject effect on predictive CA (p = .02) but no within-subject effects on postdictive CA for treatment (p = .90). The Mann-Whitney U tests showed that the treatment calibrated more accurately than the control for predictive CA on post-test (p = .00) and postdictive CA on post-test (p = .01) and delayed post-test (p = .048). After the post-test, participants had their reflection journals collected, and three from each group were invited for the interview. The findings reveal that both groups ascribed calibration judgments to controllable (e.g., effort and previous learning experience) and uncontrollable factors (i.e., memory-based cues). However, strategy use (e.g., guessing based on native language) is exclusive to the treatment. Besides, both groups recorded positive changes in personal (i.e., personal attributes, language-related knowledge, and skills) and behavioral (i.e., self-regulation) factors with varying degrees. Nevertheless, the treatment reported employing various SRL strategies, especially metacognitive strategies, more than the control, and their knowledge of cognition and thinking skills also improved. The findings provide valuable information for Vietnamese curriculum designers and English instructors about the effects of an SRL intervention and its successful elements in improving grammar performance and strategy use. The study also has significant theoretical implications concerning the necessity of integrating SCT and MMTF to explain self-regulation in grammar learning. SCT expounds on essential components of SRL, which embraces the environmental, personal, and behavior factors but does not detail how knowledge is acquired, retrieved, and retained, which MMTF complements and explains under crucial SRL processes (monitoring and control).

Text FPP 2022 1 IR.pdf Download (1MB)

Actions (login required)

View Item