Actionable companion · Assessment & reasoning

Learning with agency: Introducing Data-Driven Materials Informatics into Undergraduate Courses through a Polymer Science Workshop

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 4.0

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. Verify the participant population and task before transferring the findings.

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

With the rapid growth of artificial intelligence and machine learning across scientific disciplines from materials discovery to data-driven problem solving, there is increasing opportunity to integrate these tools into a broad range of applications. Successful adoption of these approaches in research can be enhanced by foundational exposure during undergraduate education. The objective of this study is to introduce fundamental machine learning concepts to undergraduate students through a hands-on, application-focused workshop during a polymer science and engineering course. Students were guided through key steps of the machine learning workflow, including data cleaning, model training, performance evaluation, and result interpretation, using a polymer solubility data set generated via visual inspection. The effectiveness of the workshop was assessed through pre- and postworkshop student surveys, which indicated a measurable improvement in students’ understanding and confidence in applying machine learning techniques. The integration of this workshop into a materials course introduces the students to the new concepts while extending the application of the course material.

What kind of evidence is this?

Supporting AI learning science · Evidence synthesis

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

This workshop is designed to equip participants with foundational knowledge and practical skills in essential machine learning techniques, specifically in the context of polymer solubility (Figure ). Upon completion, participants will be able to Analyze the nature of a data set using visualization techniques. Explain and implement cross-validation techniques to evaluate how well a model will generalize to an independent data set, specifically The hold-out method, by splitting data sets into training and testing subsets. The k-fold cross-validation method, including selecting an appropriate number of folds and interpreting validation results for model evaluation. Understand the difference between supervised/unsupervised machine learning models and the purpose of a decision tree classifier. Compute and interpret a confusion matrix, including Calculating and explaining the metrics of accuracy, precision, and recall. Applying these metrics to assess model performance.

Limitations and uncertainty · source passage

Using the concepts of cross-validation, the students divided the data into two sets: a training set and a validation set. Through this activity, students learned how to implement the hold-out method using the train_test_split function, initially allocating 80% of the data for training and 20% for validation. By experimenting with additional splits, such as 70/30 and 50/50, they gained insight into how the choice of training and test set proportions can impact model performance. This helped reinforce the concept of data partitioning and allowed students to observe the trade-offs between training data availability and evaluation reliability, ultimately deepening their understanding of model generalization and overfitting in practical machine learning workflows. Once the initial split was completed, students implemented 5-fold cross-validation on the training set using the KFold method. Students were asked to further analyze different folds. To ensure reproducibility and avoid bias, the data were shuffled before splitting using shuffle = True and a fixed random_state. This iterative process helped students observe how model performance can vary across different subsets of data and reinforced the value of cross-validation for building robust and generalizable models before evaluating the final results on the held-out validation set.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.

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

Mona Amrihesari; Blair Brettmann. Introducing Data-Driven Materials Informatics into Undergraduate Courses through a Polymer Science Workshop. Journal of Chemical Education (2025-08-15). https://doi.org/10.1021/acs.jchemed.5c00562

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 4.0 license. This guide has not undergone independent peer review.