Bivariate correlation (Pearson, Spearman)
Bivariate correlation measures the strength and direction of the association between two variables: Pearson's r for linear relationships between scale variables, Spearman's rho for ranked or skewed data.
Menu: Analyze → Correlate → Bivariate…
When should I use Bivariate correlation?
- Two scale variables (Pearson) or ordinal/skewed variables (Spearman).
What are the assumptions?
- Pearson: linear relationship, no extreme outliers, roughly normal variables. Check the scatterplot.
- Missing values are excluded pairwise (each pair uses all its complete cases).
How do I run it in StatLab?
- Choose two or more variables.
- Tick Pearson and/or Spearman; with exactly two variables a scatterplot with fit line is added.
How do I read the output?
| Output | Meaning |
|---|---|
| Correlation | r or rho from −1 to +1. |r| ≈ .1 small, .3 medium, .5 large. |
| Sig. (2-tailed) | Tests whether the correlation is zero (t distribution with n − 2 df). |
| N | Pairs used. |
| * / ** | Significant at .05 / .01. |
Worked example
Correlate age and score in the bundled sample dataset (File → Sample dataset: 150 survey respondents with gender, programme group, age, pre/post test scores, five satisfaction items, income, smoking and education): r = .37, p < .001, N = 150.
How would I report it in APA style?
Age was significantly positively correlated with wellbeing, r(148) = .37, p < .001.
Numbers above were computed by StatLab’s validated engine from the sample dataset when this page was built.
What are common mistakes?
- Reading correlation as causation.
- Using Pearson on a curved relationship — look at the scatterplot first.