Week 13 · Suggested 90 minutes · First-pass draft
More than one predictor or group
See where an omnibus ANOVA, logistic regression, and a missing-data audit differ from ordinary straight-line fitting.
Applied assignment, not a stored grade. Instructions and model checks are provided for self-study or instructor adaptation. Work is not transmitted to this site.
What you will do
- Compute one-way ANOVA sums of squares and F for the fictional groups [2,3,4], [4,5,6], [6,7,8].
- Use logistic regression to translate a log-odds value of −1 into probability.
- Inspect missingness for R's historical airquality dataset and document complete-case selection.
- Describe one assumption or design limitation associated with each of the three procedures.
What to submit
A worked F statistic, a logistic probability/odds explanation, and an R missingness audit.
Self-check before finishing
ANOVA F=12 with df=(2,6); logistic(−1)≈.269; airquality has 153 records and 37 missing ozone values.
Assigned reading
Study ANOVA, logistic regression, causal designs, and missingness.
- Lesson 39: analysis of variance
- Lesson 40: logistic regression introduction
- Lesson 41: causal inference study design
- Lesson 42: missing data
Instructor discussion guidance
Insist on the distinction between odds and risk, and between descriptive association and causation.
Print a predict–experiment–explain worksheet → · Open browser R →