Which statement about Cronbach's alpha is a known limitation?

• A. It is unaffected by the number of items in the scale.
• B. It measures test-retest reliability.
• C. It directly indicates validity.
• D. It can be inflated by more items and assumes unidimensionality.

Answer: (D)

Cronbach's alpha gauges internal consistency, meaning it looks at how closely the items on a scale hang together to reflect a single underlying construct. A key limitation is that alpha can be inflated simply by having more items, especially if those items are similar or redundant. Adding items tends to boost average inter-item correlations and the overall alpha, even if the new items don’t genuinely improve measurement of the construct. At the same time, alpha assumes unidimensionality—the idea that one factor underlies all items. If a scale actually measures more than one dimension, alpha can misrepresent reliability, either overstating or understating it depending on how the items group. Because of these issues, Cronbach's alpha isn’t a flawless stand-alone index of reliability; it’s important to examine the scale’s dimensional structure (e.g., with factor analysis) and consider alternative reliability estimates when multidimensionality is present.