Correlation
What r measures and what it does not.
Slides · Dataset
About 30 minutes to complete
Enrol to open the slides and download the data.
What this module covers
what r measures and what it does not; Pearson vs Spearman vs Kendall's tau and the rule for choosing; assumptions, especially linearity - always plot first; the scatterplot gallery showing r = 0 with an obvious relationship; the menu path; reading the correlation matrix; interpreting magnitude, and r-squared as shared variance; partial correlation and what controlling actually does; correlation is not causation, stated with a concrete confounding example; APA reporting.
By the end of it
- What r measures and what it does not
- Pearson vs Spearman vs Kendall's tau and the rule for choosing
- Assumptions, especially linearity - always plot first
- The scatterplot gallery showing r = 0 with an obvious relationship
- Reading the correlation matrix
- Interpreting magnitude, and r-squared as shared variance
- Partial correlation and what controlling actually does
- Correlation is not causation, stated with a concrete confounding example
The dataset
Antenatal care and birth outcome — Sigma teaching dataset
A teaching dataset of 300 cases, generated from a fixed seed so every figure in the module reproduces exactly.
- CodebookIncluded
- Dataset (CSV)Included
- Dataset (CSV)Included
- SPSS syntaxIncluded