Classically, a Likert scale was composed of the sum of many Likert items (ordinal ratings of the amount of agreement with a statement), where all the items were equally valid. Today the term sometimes is used synonymously with 'ordinal rating scale' (which may be based on only 1 item).

A Likert scale is a measurement instrument commonly used in questionnaires that is intended to assess the strength of a person's attitude or belief. The original theory pertained to the sum or average of many Likert items (individual questions) that all measured the same construct, and where the responses to each item were a set of ordered categories from "strongly disagree" to "strongly agree" with some number of intermediate response levels in between. Today, the term is often used to refer to any ordinal rating scale.

A prototypical Likert item might be:

I often wonder about ____________.  

        Strongly       Disagree       Neither agree        Agree       Strongly
        disagree                      nor disagree                      agree  

From the perspective of the theory of levels of measurement, there has been a long-running debate regarding whether it is proper to consider Likert data to be interval level, or only ordinal.

history | excerpt history