📸 SPSS partial correlation menu path
What Is Partial Correlation?
Partial correlation measures the linear relationship between two variables while statistically controlling for the effect of one or more additional variables. When a third variable (confounder) influences both X and Y, the zero-order Pearson r may be misleading — partial correlation removes that shared influence.
Example: The correlation between stress and health outcomes might be inflated because both are influenced by age. Partialling out age gives the true stress-health relationship.
Running Partial Correlation in SPSS
Step 1: Analyze → Correlate → Partial.
Step 2: Move the two variables of interest to Variables. Move the control variable to Controlling for → OK.
📸 Partial correlation output — r changes after controlling for age
APA Reporting
After controlling for age, the partial correlation between stress and sleep quality remained significant, r(96)=-.412, p<.001, though smaller than the zero-order correlation (r=-.531), indicating that age partially accounts for this relationship.