📸 Phi and Cramer's V in SPSS Crosstabs → Statistics
Why Chi-Square Alone Is Insufficient
Chi-square tells you if an association is significant, not how strong it is. Two studies can show identical chi-square values but very different sample sizes — meaning entirely different practical importances. Phi (φ) for 2×2 tables and Cramer's V for larger tables provide the effect size missing from chi-square.
Interpretation Benchmarks (Cohen, 1988)
- φ or V = .10: Small effect
- φ or V = .30: Medium effect
- φ or V = .50: Large effect
Running in SPSS
Step 1: Analyze → Descriptive Statistics → Crosstabs.
Step 2: Statistics → check Phi and Cramer's V → Continue → OK.
📸 Phi coefficient — φ=.341, medium effect size
APA Reporting
Chi-square analysis revealed a significant association, χ²(1, N=120)=13.93, p<.001, with a medium effect size, φ=.341.