Actionable companion · Accessibility & inclusion

Learning with agency: The utility of adaptive eLearning data in predicting dental students’ learning performance in a blended learning course

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paper Search paper agent CC BY 3.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Higher education. Health-professions learning must be distinguished from K–12 instruction.

Original authors’ abstract, reproduced under the linked source license:

Objectives To examine the impact of dental students’ usage patterns within an adaptive learning platform (ALP), using ALP-related indicators, on their final exam performance. Methods Track usage data from the ALP, combined with demographic and academic data including age, gender, pre- and post-test scores, and cumulative grade point average (GPA) were retrospectively collected from 115 second-year dental students enrolled in a blended learning review course. Learning performance was measured by post-test scores. Data were analyzed using correlation coefficients and linear regression tests. Results The ALP-related variables (without controlling for background demographics and academic data) accounted for 29.6% of student final exam performance (R2=0.296, F(10,104)=4.37, p=0.000). Positive significant ALP-related predictors of post-test scores were improvement after activities (β=0.507, t(104)=2.101, p=0.038), timely completed objectives (β=0.391, t(104)=2.418, p=0.017), and number of revisions (β=0.127, t(104)=3.240, p=0.002). Number of total activities, regardless of learning improvement, negatively predicted post-test scores (β= -0.088, t(104)=-4.447, p=0.000). The significant R2 change following the addition of gender, GPA, and pre-test score (R2=0.689, F(13, 101)=17.24, p=0.000), indicated that these predictors explained an additional 39% of the variance in student performance beyond that explained by ALP-related variables, which were no longer significant. Inclusion of cumulative GPA and pre-test scores showed to be the strongest and only predictors of post-test scores (β=18.708, t(101)=4.815, p=0.038) and (β=0.449, t(101)=6.513, p=0.038), respectively. Conclusions Track ALP-related data can be valuable indicators of learning behavior. Careful and contextual analysis of ALP data can guide future studies to examine practical and scalable interventions.

What kind of evidence is this?

Intelligent tutoring & adaptive learning · Comparative / pre-post study reported

Stage: Higher education. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

The University of Illinois at Chicago institutional review board reviewed and approved the study. The data were retrospectively obtained from a convenience sample of second-year dental students who were enrolled in a single course offered in a blended learning format during the academic years 2016-2018, where an ALP was used, and its online homework grades constituted 30–35% of the cumulative grade in the course. A total of 115 students represents the entire study population. Detailed tracking logs from the ALP present the study's largest and most novel data source (Table 1). We collected 15 tracking indicators from the ALP as independent variables (Table 1). Additional demographics and background academic information were collected including average age, gender, pre-test scores, cumulative GPA prior to course enrollment, and course grade after course completion (Table 2).

Limitations and uncertainty · source passage

The second significant predictor found in this study was timely submission of ALP assignments, reflecting students’ time management skills or regular study behaviors by allowing a greater amount of time for reflection and synthesis. This finding is consistent with numerous studies on self-regulated learning in online learning, in which reported failure in self-regulation reflected academic procrastination, and procrastination had a significant negative impact on success in exam performance17-19 and potential student dropout.4 Further, You (2016) found that two academic procrastination indicators in LMS data explained 59.7% of the variability in students' academic achievement.4 In line with previous studies, this study’s findings support the perspective that time management strategies have a considerable significant impact on academic achievement and, thus, should be considered as a variable worth addressing in future updates in the ALP and/or future intervention studies.20,21 Given the scope of this study, we did not analyze detailed logs to objectively measure consistent study behaviors, via indicators such as frequency of accessed sessions, dates of first and last activities, and interval between activities, which is a promising area for future research, using sophisticated data mining techniques.

Design the system before the pilot

Document the learner model, mastery rule, hint sequence and adaptation signal. Compare adaptive routing with the same content in a non-adaptive sequence. Check whether early errors lead to persistent low-expectation pathways.

Co-design accommodations with learners and specialists. Check access and error burden across groups without inferring subgroup effectiveness from small or unreported samples.

A 90-day district pilot

Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
Days 16–30 · Design. Choose one disciplinary task. Require a first attempt, a documented AI consultation, a critique of the output and an independent explanation. Align the comparison group and scoring rubric with the same objective. Name an accountable owner and set the baseline, success threshold and budget before implementation.
Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
Days 61–90 · Decide. Blind-score reasoning and accuracy on a new task without AI. Distinguish model examination scores, student perceptions and actual learner gains; only the last directly tests educational benefit. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Farhan H. Alwadei; Blasé P. Brown; Saleh H. Alwadei; Ilene B. Harris; Abdurahman H. Alwadei. The utility of adaptive eLearning data in predicting dental students’ learning performance in a blended learning course. International Journal of Medical Education (2023-10-06). https://doi.org/10.5116/ijme.64f6.e3db

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 3.0 license. This guide has not undergone independent peer review.