Schedule (topic flow)

A 15-week pacing plan — Blackboard is authoritative for dates and deadlines

Important

This is a pacing plan, not a calendar. It shows the order in which the methods unfold and roughly when, so you can read ahead and see how the pieces connect. It is not the authoritative calendar: exact dates, deadlines, and any mid-term adjustments live in Blackboard (the LMS), which is authoritative. A few fixed semester dates are noted below for orientation only — confirm everything against Blackboard.

The course runs fifteen weeks in four parts. Every week is one pass through the same analysis blueprint — ask the question, read the data structure, choose the method, check assumptions, report the estimate with its uncertainty, and draw a conclusion that keeps statistical significance, practical significance, and causation distinct. The methods change week to week; the six-step move does not. Each part below lists its weeks, the theme for each week, and the throughline — the question that carries the week. The theme links to that week’s note.

This is a draft course site: the example numbers in the linked notes come from one synthetic world (the Cypress Ridge College Student-Success Study, generated with set.seed(35203)) and are provisional and unverified — they illustrate the reasoning, not real findings.

Part I — Questions, structure, and the estimate (Weeks 1–3)

We set the frame the whole course runs on: turn a vague question into a sharp statistical one, read the data structure that decides the method, and learn to report an estimate with its uncertainty instead of a bare verdict. No named test is the point yet — the blueprint is.

Week	Theme	The throughline
1	Statistical questions, data structure & applied workflow	Are we comparing, explaining, or predicting — and what is the unit, the response, and the design?
2	Exploratory analysis & graphical comparison	What do we see in the data — its shape, spread, and group differences — before we fit anything?
3	Estimation, uncertainty & practical significance	What does the analysis estimate, how uncertain is it, and is the effect big enough to matter?

Note: Labor Day (Mon, Sep 7) falls in Week 3 — no class that Monday, so the week runs Wednesday and Friday (compressed).

Part II — Comparing groups: means, factors, and assumptions (Weeks 4–9)

The core of the course. We work through comparisons of growing structure — paired, two-group, many-group, and two-factor — and on the way learn to check what each method assumes, control error rates across multiple comparisons, and read an interaction. Every method is one instance of the blueprint, never a standalone recipe.

Week	Theme	The throughline
4	One-sample & paired comparisons	When each unit is its own control, how does pairing remove between-unit variation — and what do we estimate?
5	Two-group comparisons	With independent groups, what is the mean difference, its interval, and why is this only an association?
6	Many-group comparisons & one-way ANOVA	When several group means differ, how does ANOVA test them together, and what does \(F\) actually compare?
7	Assumptions, diagnostics & the midterm	What does the model assume, how do we check it, and what do we do with an outlier we shouldn’t auto-delete?
8	Multiple comparisons & planned contrasts	Once ANOVA is significant, which means differ — and how do we ask without inflating the error rate?
9	Two-way ANOVA & interaction	With two factors, when is an effect conditional — and why do we read the interaction before the main effects?

Note: the midterm is Friday, Oct 9 (in class), during Week 7. It covers the applied workflow, exploratory analysis, estimation and uncertainty, practical significance, and the comparisons through assumptions and diagnostics — Weeks 1–7. No graded content appears on this public site — see Blackboard.

Part III — Models, covariates, and categorical outcomes (Weeks 10–13)

We turn to models that explain and adjust. Regression introduces the partial (adjusted) slope; ANCOVA brings the same adjustment idea to group comparisons; then the outcome itself goes categorical — contingency tables and logistic regression for a binary result. The recurring thread is confounding → adjustment → association vs causation: adjusting for a covariate changes the estimate, and observational data still buy only association.

Week	Theme	The throughline
10	Simple & multiple regression review	What does a slope estimate, and why does it change once we hold other variables fixed (the partial slope)?
11	ANCOVA & adjustment	How do we compare group means at the same baseline by adjusting for a covariate — and what must hold?
12	Categorical outcomes & contingency tables	When the outcome is a category, how do we measure and test association — risk difference, relative risk, odds?
13	Logistic regression for binary outcomes	How does a model on the log-odds scale give an odds ratio and a predicted probability — and why isn’t OR an RR?

Note: fall break is Nov 22–28 (no classes), falling between Weeks 13 and 14.

Part IV — Reporting and synthesis (Weeks 14–15)

We bring it together: take one dataset end-to-end through the blueprint and write an honestly-bounded report, then step back and see the whole course as one repeated move across the five data structures.

Week	Theme	The throughline
14	Applied analysis report workshop	How do you carry a real analysis from question to conclusion and report the estimate, the uncertainty, and the limits?
15	Applied methods synthesis & review	How did one blueprint generate the paired \(t\), ANOVA, regression, ANCOVA, chi-square, and logistic regression?

Note: the last class is Mon, Dec 7 (the Week 15 synthesis meeting), the consultation day is Dec 8, and the final-exam window is Dec 9–15 (the exact block is set via Blackboard).

Note

Again: this page is a pacing plan to help you read ahead and see where the course is going. For the real calendar — every date, every deadline, the midterm and final blocks, and any adjustment during the term — Blackboard (the LMS) is authoritative. If this page and Blackboard ever disagree, follow Blackboard.