In social sciences, especially when working with survey data, factor analysis is frequently used to understand relationships between variables. However, before conducting this analysis, it’s essential to test whether the data is suitable for it. This is where scale suitability tests come into play. In this article, we’ll explain two key tests—KMO (Kaiser-Meyer-Olkin) and Bartlett’s Test—in a clear and simple way.

1. What Is a Scale Suitability Test?

Scale suitability tests are preliminary checks to determine whether your data is appropriate for factor analysis. These tests assess whether there are sufficient correlations among variables.

KMO (Kaiser-Meyer-Olkin) Test
The KMO test measures whether the correlations among variables are adequate for factor analysis. It yields a value between 0 and 1.

Interpretation:

Tip: If the KMO value is below 0.60, consider reviewing your variables before proceeding with factor analysis.

Bartlett’s Test of Sphericity
This test checks whether there are significant correlations among variables. Its main purpose is to determine whether the correlation matrix is an identity matrix (i.e., all variables are independent).

Interpretation:

• p < 0.05 → Significant relationships exist; factor analysis is appropriate
• p > 0.05 → Insufficient relationships; factor analysis is not recommended

Note: Bartlett’s test should be significant, and the KMO value should be high. These two results should be evaluated together.

2. How to Run KMO and Bartlett’s Test in SPSS

• Go to: Analyze > Dimension Reduction > Factor
• Select your variables
• Under “Descriptives,” check the box for “KMO and Bartlett’s test of sphericity”
• Click “Continue” and then “OK”
• The output will display both the KMO value and Bartlett’s test result

3. How to Report in Your Thesis

“Before conducting factor analysis, scale suitability tests were performed. The KMO value was found to be 0.81, and Bartlett’s Test was significant (χ² = 456.78, df = 66, p < 0.001). These results indicate that the data is suitable for factor analysis.”

4. Conclusion

KMO and Bartlett’s tests are essential tools for checking the prerequisites of factor analysis. Skipping these tests can lead to statistically weak and unreliable results. Including them in your thesis demonstrates both methodological rigor and analytical competence.