Which statistic is commonly used to assess internal consistency of a multi-item scale?

• A. Pearson correlation.
• B. Cronbach's alpha.
• C. Cohen's kappa.
• D. ANOVA.

Answer: (B)

Internal consistency asks whether a set of items all tap into the same underlying construct. Cronbach's alpha is the standard statistic used to quantify that coherence. It looks at how well each item tends to move with the others and accounts for the number of items in the scale. When items are consistently related, alpha is higher, indicating the scale hangs together well. If items aren’t related, alpha is lower, suggesting the items may not be measuring the same thing. A higher alpha generally means more cohesion, but an alpha that’s too high can signal redundancy among items. In practice, values around 0.7 to 0.9 are often considered acceptable to good, depending on the context. The other options don’t assess internal consistency: Pearson correlation measures the relationship between two variables at a time, Cohen's kappa is about agreement between raters on categorical judgments, and ANOVA compares means across groups.