Scale

Scale Definition

Scales are systematic methods used in social statistics and psychological measurement to quantify attitudes, beliefs, or complex factors by arranging items into ordered categories or continuous ranges. These methods allow researchers to move beyond simple dichotomous answers to gauge the intensity and distribution of responses across a spectrum, thereby providing quantifiable data on human behaviour and social constructs.

Types of Measurement Scales

Various scales exist, each suited for measuring different types of psychological phenomena. One notable approach is the Guttman scale, which aims to derive a single score representing an underlying factor from a set of related questions. This method involves assessing items—such as those relating to anti-Semitism—by experts and then using their consensus judgments to assign scores. For this method to yield meaningful results, it is generally accepted that a ten per cent margin of error is acceptable; differences exceeding this threshold are required to suggest a genuine divergence rather than mere measurement error.

Likert Scales

Likert scales are designed to capture one-dimensional attitudes by asking respondents to indicate their degree of agreement or disagreement with a series of statements, rather than simple yes/no answers. Respondents typically rate items along a continuum, often using a three, five, or seven-point scale where the middle point represents neutrality. A common five-point structure includes options such as strongly agree, agree, neither agree nor disagree, disagree, and strongly disagree. By analysing the correlation between item scores and overall responses, researchers can construct scales that exhibit high internal consistency.

Thurstone Scaling

Thurstone scaling methods focus on establishing a scale of favourableness for complex concepts by using a process involving multiple statements. The ‘equal-appearing intervals’ method is a common form of this technique. This involves generating a large set of statements describing specific attitudes towards a concept (e.g., AIDS sufferers). A panel of judges rates the favourableness of each statement on a numerical scale. By analysing the distribution of these ratings, researchers can select statements that represent distinct levels of attitude, effectively creating evenly spaced intervals that allow respondents to express their feelings along a continuous spectrum rather than just a binary choice.