A flipped classroom is a blended instructional approach in which the in-class lecture is shifted to before-class learning using videos or other forms of online learning to free up in-class time for students to discuss issues, practise, or apply knowledge (Bergmann & Sams, 2014). A persistent problem with implementing flipped classroom is that, without proper guidance or assistance, students lacked the engagement and self-regulation skills to complete online learning activities before class, and hence failed to learn effectively in the following in-class lessons (Mason et al., 2013).

In an online learning environment, promoting students’ self-regulated learning (SRL) through planning, goal setting, organising, self-monitoring and self-evaluating during the learning process is imperative (Zimmerman, 1990). The theoretical framework of SRL is underpinned by three components of SDL identified as: (1) metacognitive strategies; (2) task management and control; and (3) cognitive strategies to learn the materials (Pintrich & DeGroot, 1990). Previous research has reported that motivational beliefs are positively associated with SRL (Credé & Phillips, 2011).

With the advancement of Artificial Intelligence (AI) technology, chatbot has gained prominence in education and regarded as a useful tool to provide personalised guidance, support or feedback to support students’ learning. There is a growing body of evidence on the use of chatbots to promote students’ SRL in an online environment (Du & Hew, 2021; Hew et al., 2022; Song & Kim, 2020). Other previous studies have reported that the use of chatbot-based learning have contributed to higher students’ learning achievements, self-efficacy, learning attitude (Lee et al., 2022), intrinsic motivation (Yin et al., 2021) and critical thinking (Chang et al., 2022).

In recent years, scholars have highlighted that there is a lack of studies to investigate the effectiveness of learning designs or learning strategies using chatbots (Chen et al., 2020). To date, there is a paucity of studies on chatbot-based learning which investigates the motivational factors that influence students’ SRL in the flipped classroom, particularly with using a critical thinking module and mixed-methods research methodology. Therefore, this present study is an attempt to address this gap in the literature by employing a chatbot combining with worksheet scaffold to investigate the extent to which motivational factors can influence students’ SRL before-class. Understanding the motivational factors that impact students’ SRL using chatbots in the flipped classroom is crucial for researchers and educators to reflect upon, and develop better learning approaches or interventions to support students’ learning in the future.

Specifically, we formulated the following research questions to guide in the data analysis:

RQ1: To what extent does the four motivational variables (i.e. intrinsic goal orientation, extrinsic goal orientation, task value and self-efficacy) correlate with SRL among students using chatbots?;

RQ2: To what extent do the four motivational variables (intrinsic goal orientation, extrinsic goal orientation, task value and self-efficacy) predict SRL for students using chatbots?;

RQ3: Based on the interviews, what are the students’ perspectives of their learning experiences in using chatbots?; and

RQ4: In what ways do the interview data reporting students’ perspectives of their learning experiences in using chatbots help to explain the quantitative results in the online questionnaire?

This study employed an explanatory sequential mixed-methods design, where data were collected in two phases via online questionnaires (N=72) and follow-up with individual semi-structured interviews (N=9). This research design allows the use of qualitative findings to provide an in-depth explanation of the quantitative findings (Creswell & Clark, 2018). The self-report online questionnaire was intended to measure five variables i.e. intrinsic goal orientation, extrinsic goal orientation, task value, self-efficacy and SRL. Based on the expectancy-value theory and achievement goal theory, these five variables were measured using selected subscales in the Motivated Strategies for Learning Questionnaire (MSLQ) developed by Pintrich et al. (1991). In particular, the metacognitive self-regulation subscale in the MSLQ was adopted as a measure of SRL. It assesses the extent to which learners utilise planning, monitoring, and regulating strategies for learning. All items in the study variables were measured on a 7-point Likert scale ranging from “1 = strongly disagree” and “7 = strongly agree”. Participants were first-year polytechnic students undertaking the Critical Thinking and Problem-Solving module. A flipped classroom approach was adopted for lesson 2 and 3 where students were required to complete a worksheet scaffold with the help of the chatbot to acquire an understanding of the learning content prior to the in-class lessons. The online questionnaire and semi-structured interviews were conducted at the end of lesson 3 and lesson 5 respectively. Descriptive statistics, correlations, reliability, and regression were used for data analysis using SPSS Statistical Package 24.0. The qualitative data from the interviews were transcribed and coded by two researchers using Nvivo Version 12 software and analysed thematically.

Preliminary analysis showed that the mean ratings of the five study variables ranged between 4.78 and 5.87 (0.94 ≤ SD ≤ 1.15). The Cronbach’s alphas of the 5 variables ranged between 0.78 and 0.95. There was no evidence of multicollinearity among the four predictor motivational variables (1.30 ≤ VIF ≤ 2.51). The Cohen’s kappa coefficient for inter-rater reliability was 0.72. To answer RQ1, all the four motivational variables revealed significant correlations with SRL. Out of the four motivational variables, intrinsic goal orientation had the highest significant correlation with SRL (r=.63, p<.01), followed by task value (r=.60, p<.01). Extrinsic goal orientation correlated the least with SRL among the motivational variables (r=.29, p<.01). With regards to RQ2, self-efficacy (β =.34, p<.01) and intrinsic goal orientation (β = .27, p<.05) were the only two independent variables that significantly predicted SRL. A total of 48% of the variance in SRL was explained by the four motivational variables, and self-efficacy alone contributed to 6.8% of the variance. In relation to RQ3, the thematic analysis of the qualitative data identified four emerging themes on usability, task strategies, motivation and perceived usefulness. Concerning RQ4, the qualitative findings suggested that SRL can be enhanced when students perceived the benefits of using the chatbot combining with the worksheet scaffold as an interactive learning tool to help them gain confidence in deepening their understanding of the learning concepts. In addition, the students adopted various task strategies, including help-seeking, self-practice, and note-taking to support their SRL. In conclusion, the study provided insights on the pedagogical affordances of the chatbots to enhance students’ SRL through a better understanding of the four motivational variables. Finally, implications of the findings, along with study limitations and directions for future research will be discussed in the paper.

Chang, C.-Y., Kuo, S.-Y., & Hwang, G.-H. (2022). Chatbot-facilitated Nursing Education. Educational Technology & Society, 25(1), 15-27. https://doi:/10.30191/ETS.202201_25(1).0002 Chen, X., Xie, H., Zou, D., & Hwang, G.-J. (2020). Application and theory gaps during the rise of artificial intelligence in education. Computers and Education: Artificial Intelligence, 1, 100002. https://doi:/10.1016/j.caeai.2020.100002 Credé, M., & Phillips, L. A. (2011). A meta-analytic review of the Motivated Strategies for Learning Questionnaire. Learning and Individual Differences, 21(4), 337-346. https://doi.org/10.1016/j.lindif.2011.03.002 Creswell, J. W., & Clark, V. L. P. (2017). Designing and conducting mixed methods research: Thousand Oaks, CA: Sage publications. Du, J., & Hew, K. F. T. (2021). Using recommender systems to promote self-regulated learning in online education settings: current knowledge gaps and suggestions for future research. Journal of Research on Technology in Education, 54(4), 1-22. https://doi:/10.1080/15391523.2021.1897905 Hew, K. F., Huang, W., Du, J., & Jia, C. (2022). Using chatbots to support student goal setting and social presence in fully online activities: learner engagement and perceptions. Journal of Computing in Higher Education, 1-29. https://doi:/10.1007/s12528-022-09338-x Lee, Y.-F., Hwang, G.-J., & Chen, P.-Y. (2022). Impacts of an AI-based chabot on college students’ after-class review, academic performance, self-efficacy, learning attitude, and motivation. Educational Technology Research and Development, 70(5), 1843-1865. https://doi:/10.1007/s11423-022-10142-8 Mason, G. S., Shuman, T. R., & Cook, K. E. (2013). Comparing the effectiveness of an inverted classroom to a traditional classroom in an upper-division engineering course. IEEE Transactions on Education, 56(4), 430-435. https://doi:/10.1109/TE.2013.2249066 Pintrich, P. R., Smith, D. A. F., Garcia, T., & McKeachie, W. J. (1991). A manual for the use of the Motivated Strategies for Learning Questionnaire (MSLQ): Ann Arbor, MI: National Center for Research to Improve Postsecondary Teaching and Learning. Pintrich, P. R., & DeGroot, E. V. (1990). Motivational and self-regulated learning components of classroom academic performance. Journal of Educational Psychology, 82(1), 33–40. https://doi.org/10.1037/0022-0663.82.1.33 Song, D., & Kim, D. (2021). Effects of self-regulation scaffolding on online participation and learning outcomes. Journal of Research on Technology in Education, 53(3), 249-263. https://doi:/10.1080/15391523.2020.1767525 Yin, J., Goh, T.-T., Yang, B., & Xiaobin, Y. (2021). Conversation technology with micro-learning: The impact of chatbot-based learning on students’ learning motivation and performance. Journal of Educational Computing Research, 59(1), 154-177. https://doi:/10.1177/0735633120952067 Zimmerman, B. J. (1989). A social cognitive view of self-regulated academic learning. Journal of Educational Psychology, 81(3), 329. https://doi:/10.1037/0022-0663.81.3.329