📸 Canonical correlation in SPSS — run via MANOVA syntax
What Is Canonical Correlation?
Canonical correlation analysis examines the maximum linear relationship between two sets of variables. One set might contain academic variables (GPA, test scores, attendance) and another behavioral variables (motivation, self-efficacy, anxiety). Canonical correlation identifies synthetic linear combinations of each set (canonical variates) that are maximally correlated.
Running in SPSS (Syntax Required)
Step 1: Canonical correlation isn't available in SPSS menus. Open File → New → Syntax.
Step 2: Type:
MANOVA y1 y2 y3 WITH x1 x2 x3 /PRINT=SIGNIF(EIGEN DIMENR).Step 3: Run → All. Output shows eigenvalues, canonical correlations, and Wilks' Lambda for each canonical dimension.
📸 Canonical correlation output — two significant canonical dimensions
APA Reporting
Canonical correlation analysis revealed two significant canonical dimensions. The first canonical correlation was rc=.712, Wilks' Λ=.412, F=8.41, p<.001, and the second was rc=.481, Wilks' Λ=.768, F=3.12, p=.024.