Teaching guide
Audience and delivery
Assume basic but uneven statistics knowledge and no prior R. Explain the statistical idea needed for the current task. For classroom delivery, the pilot plans below assume 120-minute sessions. Module pacing can be adapted. Keep explanations short enough for learners to predict, run, inspect, explain, and modify code themselves.
Course 1 introduces data preparation. Course 2 covers regression and assumes basic statistics and R. Course 3 assumes prior regression training and covers computational social science. Course 4 develops programming skills from a basic R foundation. Courses 1 and 3 are the authoring priorities; Courses 2 and 4 will be developed later. Most modules include guided local R practice. Images as Social Data and Rcpp are demonstrations. Do not make specialist toolchain installation a requirement for completing the foundation course.
Browser activities are formative. Their answers and explanations are available to learners; they are not secure examination items. The site does not collect identities or grades. Discuss open-ended answers or compare them with model responses.
Course 1 · Module 1
| Minutes | Activity |
|---|---|
| 0–10 | Check preparation; ask learners to identify editor and console |
| 10–30 | Demonstrate assignment and calculation; use question 1 |
| 30–55 | Guided table inspection and summary; use questions 2 and 3 |
| 55–65 | Break |
| 65–85 | Modify an object; diagnose capitalization with question 4 |
| 85–110 | Independent task, hints, and paired explanations |
| 110–120 | Exit question and restart-from-top check |
Keep the data-creation block supplied. Avoid a detour into every data structure or editor. Ask a learner to explain a line before running it. If setup blocks a learner, pair them with someone whose R works and use the rendered output while troubleshooting.
Listen for: confusing a displayed result with assignment; treating a column as a separate scalar; believing a mean describes every learner; treating head() as deleting rows.
Exit answer: study_hours is a separate object storing one value, classroom$study_hours is a 12-value column, and its mean is one summary value. The independent task should run after restarting R without manually recreating objects.
Course 2 · Module 2
| Minutes | Activity |
|---|---|
| 0–10 | Recall scatterplots and summaries; identify outcome and predictor |
| 10–30 | Demonstrate the formula, model object, coefficients, and question 1 |
| 30–55 | Guided plot inspection, predictions, and question 2 |
| 55–65 | Break |
| 65–85 | Discuss limits and diagnose reversed formula; questions 3 and 4 |
| 85–110 | Independent task and comparison of full/restricted samples |
| 110–120 | Exit question and discussion |
Do not assume learners know regression notation. The required interpretation uses score points per hour. Keep formal inference optional; a p-value cannot turn fictional records into empirical evidence.
Listen for: points confused with percentages; predictions treated as exact outcomes; a positive slope treated as proof of causation; a valid reversed formula treated as the same model.
Exit answer: the prediction does not guarantee an individual’s result, and an observed association does not establish what assigning four hours of study would cause. The records are also constructed rather than evidence about an actual population.
Pilot feedback
The lesson templates are authored; beginner usability has not yet been tested. Before extending all 48 modules, try both pilots with a few intended readers and record:
- Where the learner first needed help and whether it was setup, syntax, or interpretation.
- Whether the learner could modify an example without copying the solution.
- Whether the learner could explain a result with units and a limitation.
- Time spent on each phase and the prompt that caused the longest delay.
Use the feedback worksheet. Keep identifying information out of shared notes. Split an overloaded module before authoring its successors. The remaining 46 modules are curriculum entries, not finished lessons.