Likert Scales

Likert scale is a technique to measure complex concepts, such as attitudes, styles of media use, or satisfaction for aspects of life, and have become increasingly common for survey research into consumer behavior. Due to simplicity, it is the most well-known and used practice of unidimensional scaling. By scales, we mean “measurement instruments [Page 866]that are collections of items combined into a composite score, and intended to reveal levels of theoretical variables not readily observable by direct means” (DeVellis 2003, 8–9). For instance, it is common to combine a set of variables (exposition to political information on newspapers, exposition to political information on television, knowledge of political facts, knowledge of political leaders, etc.) into a single index to measure the level of political information of a sample of individuals. Political information is an example of a unidimensional scale: combined with others (electoral participation, parties participation, associations participation, etc.), it can originate a multidimensional scale to reveal political participation.

The original and principal application of Likert scale is the measurement of attitudes (Likert 1932). If attitudes are general beliefs impossible to observe, then opinions derived from such attitudes are easier to reveal: based on this principle, Likert scales consist of a set of opinions on which individuals are asked to agree or disagree. For instance, it is possible to define a scale of social dominance from the agreement or disagreement individuals give on a battery of opinions like the following ones: “Some groups of people are simply inferior to other groups,” “It's OK if some groups have more of a chance in life than others,” “If certain groups stayed in their place, we would have fewer problems,” and so on. Likert scale is used in other applications where items change—they are not opinions—while scale analysis procedures remain the same. From the point of view of consumer culture, two other applications of Likert scale are particularly meaningful: the measurement of the frequency and satisfaction for consumption behavior. For instance, it is possible to define an index of cultural extra-domestic consumption asking for the frequency individuals go to cinemas, theaters, libraries, and so forth. Moreover, the satisfaction for the various aspects of these and other consumptions can be investigated.

The procedure to define a Likert scale presents three phases: formulating questions, administering questions, and analyzing scale. Attention is paid here to the first and the last phases that are the more specific to Likert scales. As to the first phase, the usual norms in question wording apply: statements must be clear, concise, straightforward, and unambiguous while the vocabulary must be as simple as possible, according to the cognitive skills of the respondents. Double negatives (e.g., “I am not in favor of corporations stopping funding for antinuclear groups”) or double-barreled statements (e.g., “I support civil rights because discrimination is a crime against God;” an individual could agree with the first part of the sentence disagreeing with the second part) should be avoided. In attitudes measurement applications, it is important that sentences permit a judgment of value (opinions) rather than a judgment of fact: “Two persons with decidedly different attitudes may, nevertheless, agree on questions of fact. Consequently, their reaction to a statement of fact is no indication of their attitudes” (Likert 1932, 44). Opinions can come from different sources: newspapers and magazines, books, addresses and pamphlets, parallel pieces of qualitative research (e.g., focus group), previous research. As to question and response formats, out of the four kinds of questions originally proposed by Likert, the five-point agreement/disagreement scale (strongly approve, approve, undecided, disapprove, strongly disapprove) prevailed in the scaling research following his contribution. On this basis, much methodological research has been developed on three technical points: whether to allow a longer seven-point scale, whether to permit a central neutral option (undecided), whether to allow a residual option (don't know).

...