Which of the following coefficients is commonly used for Likert-type items?

The use of Cronbach's alpha is well-suited for Likert-type items because it measures the internal consistency of a set of items that are designed to assess the same underlying construct. Likert scales typically consist of multiple items that provide a range of responses (such as from "strongly disagree" to "strongly agree"). By calculating Cronbach's alpha, researchers can determine how closely related these items are to one another; a high value of Cronbach's alpha (generally above 0.70) suggests that the items measure an underlying factor in a reliable way.

This is particularly important in survey research where multiple items are used to form a scale or index, as it helps ensure that the scale is reliably measuring the same concept across different items. The other coefficients listed serve different functions: the KR-20 coefficient is specifically used for dichotomous items, split-half reliability measures consistency between two halves of a test, and test-retest reliability assesses stability over time. Therefore, a focus on internal consistency makes Cronbach's alpha the most appropriate choice for Likert-type items.