Schedule
Public version. A generic week-by-week topic map, kept in “Week 1, Week 2, …” form so it stays useful across terms.
The table below shows the order the course builds ideas in, and which weekly note page goes with each topic. Each note page is the main reading for that week, with outbound pointers to OpenIntro IMS and ISLBS for a second voice.
Week-by-week
| Wk | Topic | Notes |
|---|---|---|
| 1 | Data, evidence, and statistics | Week 1 notes — cases, variables, types; reading a small dataset honestly |
| 2 | Study design, bias, and causality | Week 2 notes — populations and samples, observational vs experimental, confounding |
| 3 | One-variable summaries | Week 3 notes — center, spread, shape; tables and graphs for a single variable |
| 4 | Comparing groups | Week 4 notes — conditional proportions, group differences, choosing the right display |
| 5 | Association | Week 5 notes — scatterplots, correlation as a descriptive number, the linear-only caveat, association vs causation |
| 6 | Confounding and multivariable thinking | Week 6 notes — confounders, stratification, Simpson’s paradox, alone vs after accounting for |
| 7 | First-half synthesis | Week 7 notes — Units 1–6 walkthrough on one connecting study; a six-pass routine for reading any study; a single bounded forward-pointer to next week’s regression line |
| 8 | Simple regression | Week 8 notes — fitted line, slope and intercept in context, residuals and the residual-plot reading, least squares, R², extrapolation, light outliers and leverage |
| 9 | Multiple / logistic regression interpretation | Week 9 notes — adding predictors; the holding constant coefficient reading; adjusted vs unadjusted coefficients; categorical predictors; adjusted R² as a reading; interpretation-only logistic regression (direction + predicted probability) |
| 10 | Probability as risk and diagnosis | Week 10 notes — probability as risk; conditional probability and P(A | B) vs P(B | A); diagnostic two-way tables; sensitivity, specificity, predictive values; false positives/negatives; base-rate effects |
| 11 | Simulation-based inference | Week 11 notes — randomization tests, the null distribution, simulated p-values as evidence, bootstrap distributions, percentile confidence intervals, and reading StatKey-style output |
| 12 | Classical hypothesis testing | Week 12 notes — null and alternative hypotheses, the test statistic and p-value from a normal/t model, confidence intervals, reading one-proportion z and one/two-mean t output, and Type I/II errors with significance level and power |
| 13 | Categorical outcomes | Week 13 notes — comparing rates and proportions: reading a 2×2 table; risk, the risk difference, relative risk, and the odds ratio; why case-control studies report an odds ratio; expected counts under independence and chi-square logic with the p-value as a right-tail area; and association versus causation |
| 14 | Meta-analysis and forest plots | Week 14 notes — how do studies become evidence? Single study versus systematic review versus meta-analysis; reading a forest plot (each study’s estimate and confidence interval, the line of no effect at RR = 1, square size as weight, and the pooled diamond); heterogeneity as cross-study disagreement; and a cautious read of evidence strength. An instructor-original, source-gap week with schematic (not real-study) figures |
| 15 | Final review | Week 15 notes — what claim can we responsibly make? The whole-course synthesis: the arc from data and design through inference, categorical outcomes, and evidence strength, plus a responsible-claim checklist and one cumulative scenario |
How to use this schedule
- Read by row. Each row gives the week’s topic and links to its weekly notes page. All fifteen weeks are live.
- Read ahead, then back. The topics are cumulative: each week assumes the one before it, and Weeks 7 and 15 are deliberate synthesis points that gather what came earlier.
- Weeks, not dates. This map is deliberately generic, so it describes the sequence of ideas rather than one term’s calendar.
Section details and dates are set in the learning management system (LMS) and the official institutional syllabus. See the Syllabus page for what this site does and does not cover.