Also found in: Dictionary, Thesaurus, Medical, Legal, Financial, Wikipedia.
Likert scalea technique for measuring the strength of a person's ATTITUDE or predisposition towards a person, object, idea, phenomenon, etc. (Likert, 1932). Likert scales assume that attitudes lie on a simple, dichotomous continuum running from one extreme position through neutral to the other extreme, for example, capitalism/ communism, religion/atheism.
Likert scales are subjective in nature, insofar as they are based on the replies given by individuals to a battery of questions. In constructing such scales a sample of respondents from the target population are presented with a large number of statements thought to have a bearing on the subject. For example, to construct a scale to measure the strength of religious belief respondents may be presented with statements such as: ‘The laws relating to blasphemy are outdated and should be abolished’; ‘We can be almost certain that human beings evolved from lower animals’; ‘Every woman has the right to terminate an unwanted pregnancy if she so wishes’, ‘The miracles in the Bible happened just as they are described there’.
Respondents are asked to indicate to what extent they agree or disagree with each statement, using a three-, five- or seven-point scale. A five-point scale is generally considered to be best. The replies given to each question are then coded (see CODING) so that a high score indicates a strong disposition towards the subject under consideration and a low score indicates its polar opposite. Finally, the Likert scale is constructed using those items whose scores correlate most closely with the overall scores, i.e. the scale has internal consistency and each item has predictability. This final form of the scale can then be administered to the population for whom it is intended.
The main problem with constructing Likert scales is that of ensuring that the individual items in the scale tap one dimension only. In measuring religious attitudes, as discussed above, for example, people's opinions about abortion are determined by many factors of which the individual's religious persuasion is only one. Various statistical techniques, such as FACTOR ANALYSIS have been devised which enable researchers to calculate the internal consistency of their scales. See also ATTITUDE SCALE AND MEASUREMENT, GUTTMAN SCALE.