A scale is a type of composite measure that is composed of several items that have a logical or empirical structure among them. That is, scales make use of differences in intensity among the indicators of a variable. For example, when a question has the response choices of "always," "sometimes," "rarely," and "never," this represents a scale because the answer choices are rank-ordered and have differences in intensity. Another example would be "strongly agree," "agree," "neither agree nor disagree," "disagree," "strongly disagree."
There are several different types of scales. We'll look at four commonly used scales in social science research and how they are constructed.
Likert scales are one of the most commonly used scales in social science research. They offer a simple rating system that is common to surveys of all kinds. The scale is named for the psychologist who created it, Rensis Likert. One common use of the Likert scale is a survey that asks respondents to offer their opinion on something by stating the level to which they agree or disagree. It often looks like this:
- Strongly agree
- Neither agree nor disagree
- Strongly disagree
Within the scale, the individual items that compose it are called Likert items. To create the scale, each answer choice is assigned a score (for instance, 0-4), and the answers for several Likert items (that measure the same concept) can be added together for each individual to obtain an overall Likert score.
For example, let's say that we're interested in measuring prejudice against women. One method would be to create a series of statements reflecting prejudiced ideas, each with the Likert response categories listed above. For example, some of the statements might be, "Women shouldn't be allowed to vote," or "Women can't drive as well as men." We would then assign each of the response categories a score of 0 to 4 (for example, assign a score of 0 to "strongly disagree," a 1 to "disagree," a 2 to "neither agree or disagree," etc.). The scores for each of the statements would then be totaled for each respondent to create an overall score of prejudice. If we had five statements and a respondent answered "strongly agree" to each item, his or her overall prejudice score would be 20, indicating a very high degree of prejudice against women.
Bogardus Social Distance Scale
The Bogardus social distance scale was created by sociologist Emory S. Bogardus as a technique for measuring the willingness of people to participate in social relations with other kinds of people. (Incidentally, Bogardus established one of the first departments of sociology on American soil at the University of Southern California in 1915.) Quite simply, the scale invites people to state the degree to which they are accepting of other groups.
Let's say we are interested in the extent to which Christians in the U.S. are willing to associate with Muslims. We might ask the following questions:
- Are you willing to live in the same country as Muslims?
- Are you willing to live in the same community as Muslims?
- Are you willing to live in the same neighborhood as Muslims?
- Are you willing to live next door to a Muslim?
- Are you willing to let your son or daughter marry a Muslim?
The clear differences in intensity suggest a structure among the items. Presumably, if a person is willing to accept a certain association, he is willing to accept all those that precede it on the list (those with lesser intensities), though this is not necessarily the case as some critics of this scale point out.
Each item on the scale is scored to reflect the level of social distance, from 1.00 as a measure of no social distance (which would apply to question 5 in the above survey), to 5.00 measuring maximize social distance in the given scale (though the level of social distance could be higher on other scales). When the ratings for each response are averaged, a lower score indicates a greater level of acceptance than does a higher score.
The Thurstone scale, created by Louis Thurstone, is intended to develop a format for generating groups of indicators of a variable that have an empirical structure among them. For example, if you were studying discrimination, you would create a list of items (10, for example) and then ask respondents to assign scores of 1 to 10 to each item. In essence, respondents are ranking the items in order of the weakest indicator of discrimination all the way to the strongest indicator.
Once the respondents have scored the items, the researcher examines the scores assigned to each item by all the respondents to determine which items the respondents agreed upon most. If the scale items were adequately developed and scored, the economy and effectiveness of data reduction present in the Bogardus social distance scale would appear.
Semantic Differential Scale
The semantic differential scale asks respondents to answer a questionnaire and choose between two opposite positions, using qualifiers to bridge the gap between them. For instance, suppose you wanted to get respondents' opinions about a new comedy television show. You'd first decide what dimensions to measure and then find two opposite terms that represent those dimensions. For example, "enjoyable" and "unenjoyable," "funny" and "not funny," "relatable" and "not relatable." You would then create a rating sheet for respondents to indicate how they feel about the television show in each dimension. Your questionnaire would look something like this:
Very Much Somewhat Neither Somewhat Very Much
Enjoyable X Unenjoyable
Funny X Not Funny
Relatable X Unrelatable