This winter’s edition of the Research Clearinghouse newsletter examines key factors shaping online learning in K–12 and higher education contexts. The studies in this issue explore differences in teachers’ motivation for online teaching, the role of self-directed learning in student satisfaction, and the ways assessment practices are developing through the use of digital tools. Two articles focus on artificial intelligence in online learning, offering both a large-scale analysis of student AI use and achievement in virtual courses, as well as a synthesis of research on AI-supported student engagement. Together, these contributions provide insight into how instructional design, teacher support, and emerging technologies continue to influence the quality and effectiveness of online learning.
Anna Morrison, Julie Smart, Daphne Wiles, Luke Bennett
This quantitative study aimed to determine the level of motivation among K-12 teachers for online teaching and investigate possible differences in motivation according to group demographics. The Online Teaching Motivation Scale (OTMS) (Wiles et al., 2023) was used to collect data from K-12 teachers (N = 62) in the Southeastern United States between October 2023 and July 2024. Participants reported a mean motivation for online teaching of 2.66 (SD = 0.52) on a 4-point Likert-like scale. Significant differences in overall motivation and self-efficacy for online teaching were found when examining content areas (STEM versus non-STEM) and grade bands (elementary, middle, and secondary). Additionally, significant differences in perceptions of administrative support for online teaching were found among grade bands. Finally, teachers’ perceptions of online teaching and learning showed significant differences between grade bands, content areas, and gender. Implications for school and district leaders are discussed based on these differences among demographic groups. Results also imply the need to develop and implement differentiated systems of support to meet the diverse needs of teachers using online tools in their classrooms.
This study follows more than 26,000 Michigan students over two years to see how they actually use AI in their online courses—and what happens to their grades when they do. AI adoption nearly doubled, with sophisticated “tool + tutor” use growing fastest, especially among high-achieving students. Early achievement gaps between AI users and non-users have almost disappeared; yet, teacher responsiveness and course design still matter far more than any AI tool. The findings offer a grounded look at how AI is reshaping K–12 learning right now, without replacing the humans at the heart of it.
Laura Sara Agrati, Loredana Perla, Viviana Vinci, Arianna Beri
This study explores how online assessment tools can support the development of pre-service secondary teachers’ assessment competencies, with a focus on the impact of prior teaching experience. It was conducted with a sample of 3,780 pre-service teachers at Pegaso University in Italy, 64.9% of whom had experience of working in a school for an average of eight years each. A questionnaire was administered in early 2024 to examine the participants’ attitudes toward knowledge and use of digital assessment tools. The results revealed a strong preference for online tools due to their perceived efficiency, clarity, and transparency. Although there was little correlation between experience and overall frequency of use (r = 0.02), more experienced teachers reported a higher appreciation of tool selection (r = 0.11), format suitability (r = 0.15), and assessment utility (r = 0.42). These findings highlight that teaching experience enhances critical engagement with digital assessment as well as familiarity. Based on the teacher assessment literacy in practice model, this study recommends that teacher education programs should adapt their training to different levels of experience to promote the development of effective and ethical digital assessment skills.
Student engagement is crucial for success in online learning, but challenging to sustain due to the high demands for autonomy and self-regulation. Advances in artificial intelligence (AI) offer solutions through personalized learning, real-time feedback, and adaptive content. This study systematically reviews 24 studies from the Web of Science database to explore AI’s role in enhancing engagement. Six key applications emerged: 1) chatbots for course design, 2) emotion, facial, voice recognition, and eye tracking, 3) machine learning for data analysis, 4) teacher-student interaction support, 5) personalized feedback and recommendations, and 6) AI-powered bots in smart learning environments. Common data sources include video recordings, activity logs, standardized datasets, and surveys. Engagement is typically measured through multi-method approaches combining surveys, AI recognition, and coded activity data. Findings show that integrating diverse AI tools and data sources provides more accurate, real-time insights into cognitive, emotional, and behavioral engagement. Limitations include the focus on peer-reviewed studies from a single database, the lack of distinction between asynchronous and synchronous online learning, and the exclusion of broader AI applications beyond student engagement. Despite these limitations, this review offers strategies for designing effective AI systems, improving engagement measurement, and creating interactive, personalized learning experiences. These insights guide researchers, educators, and practitioners seeking to leverage AI to foster student engagement in online education.
The increasing number of online courses places an expectation on learners to possess self-directed learning skills to succeed in online learning. Moreover, student satisfaction is a crucial factor in the online learning process. The current study aims to examine whether self-directed learning (i.e., motivation, self-monitoring, and self-management) impacts satisfaction in online educational courses. Data were collected through a survey involving 448 online learners and were analyzed using structural equation modeling. The questionnaire was developed by the authors based on Garrison’s self-directed learning framework, which includes motivation, self-monitoring, and self-management. The study found that motivation has a positive impact on self-monitoring, online learning satisfaction, and an indirect significant influence on self-management through self-monitoring. Furthermore, self-monitoring was found to have a positive effect on self-management while simultaneously yielding a negative influence on satisfaction with online learning. Self-management was not found to significantly influence learners’ satisfaction with online learning. The results highlight the importance of motivation in supporting the self-monitoring learning process and enhancing learning satisfaction within online education. Self-directed learning’s component can predict education learners’ online learning satisfaction.
Subscribe to the Digital Backpack for more regular updates!
Want more frequent updates? Our research team regularly publishes blog articles on online and blended learning. By subscribing to the Digital Backpack blog, you'll receive a weekly digest of Michigan Virtual's latest articles sent to your inbox every Thursday.
Explore over 1,000 publications in our Research Clearinghouse
Although we feature only four or five resources per newsletter, we encourage you to visit our Research Clearinghouse to search for citations that may be of use to you. The Clearinghouse currently houses over 1,000 citations in the field of online and blended learning, and over 100 new citations are added each year.
If you have any questions about navigating this resource, please refer to our "Getting Started Guide."
What else do you want to see?Please feel free to offer any feedback on the Clearinghouse or any other MVLRI initiatives by emailing us at mvlri@michiganvirtual.org. Thank you!
Michigan Virtual, 920 Municipal Way, Lansing, Michigan 48917
