Which statistic is commonly used to measure internal consistency reliability?

• A. Inter-rater reliability
• B. Cronbach's alpha
• C. Kappa statistic
• D. Spearman-Brown coefficient

Answer: (B)

Internal consistency reliability gauges whether the items on a scale all measure the same underlying construct and therefore produce cohesive responses. Cronbach's alpha is the statistic used to quantify that coherence, by looking at how closely the items correlate with one another and taking into account the number of items in the scale. A higher alpha indicates stronger inter-item relationships, meaning the items hang together well as a single measure. This makes Cronbach's alpha the standard tool for assessing internal consistency in psychological scales. Other statistics address different ideas: inter-rater reliability checks agreement between different observers; the Kappa statistic also assesses agreement for categorical ratings; the Spearman-Brown coefficient is tied to split-half reliability and predictions about how reliability would change with test length. But for the question of whether items on a scale form a consistent whole, Cronbach's alpha is the most appropriate and commonly used measure.