Likert Scale

The Likert scale, named for Rensis Likert (pronounced “lick-urt”) who published a seminal report describing its use, possibly is the most widely employed form of attitude measurement in survey research. Similar to nearly all psychometric scale measures, the Likert scale consists of multiple items that typically are summed or averaged to produce a more reliable measure than could be obtained by use of a single item.

The Likert scale is a special type of the more general class of summated rating scales constructed from multiple ordered-category rating items. Its distinguishing characteristics are as follows:

• Each item uses a set of symmetrically balanced bipolar response categories indicating varying levels of agreement or disagreement with a specific stimulus statement expressing an attitude or opinion (e.g. Ripe cherries are delicious).
• The response category points for each item are individually labeled (e.g. Strongly Agree, Agree, Disagree, Strongly Disagree).
• The descriptive text of these labels is chosen so that gradations between each pair of consecutive points seem similar.

This sense of equidistance often is reinforced by a set of consecutive integers (e.g. 1, 2, 3, 4) used to label each alternative along the continuum of choices. Although in technical terms, the response format used for Likert scale items is at the ordinal level of measurement, researchers traditionally have used parametric statistics (which assume at least an interval level of data) to analyze Likert scales.

It is commonplace for researchers to use the term Likert scale incorrectly. The term often is used to refer to an individual item or the response choice set featured by the items. These usages are improper. Rather, a Likert scale is a multi-item measure, each item in a Likert scale is known as a Likert item, and the response categories of a Likert item are known as a Likert response set.

Likert response sets may include four or more points, though five categories are traditional. Typical wording labels for the five categories are Strongly Agree, Agree, Neither Agree Nor Disagree, Disagree, and Strongly Disagree, though certainly other descriptors indicating varying levels of agreement and disagreement are used. Though a five-category set is most frequently employed, many psychometricians advocate using response sets of seven, nine, or even eleven points. Others prefer an even number of response choices, eliminating the neutral alternative to force a positive or negative expression of attitude. Use of four or six categories is the norm when an even number is offered.

Choosing the number of points featured by Likert items should not be driven by personal preference or simply whether one judges it appropriate to prevent fence-sitting by using a forced-choice response set with an even number of response options. Rather, the reliability of Likert scales can be affected by the number of response categories. In situations where low scale score variability is expected, reliability generally can be improved by employing greater numbers of categories. In contrast, when opinion toward the topic is widely divided, scale reliability is largely independent of the number of categories.

Although true Likert items use a symmetrically balanced agree-disagree continuum for individually labeled response choices, several other types of ordered-category items are often and incorrectly referred to as Likert scales or Likert items.

...