Often times the terms index and scale are used interchangeably. While these two types of measures do have some characteristics in common, they are two different concepts and have several distinguishing features.
An index is a type of composite measure that summarizes and rank-orders several specific observations and represents some more general dimension. An index is an accumulation of scores from a variety of individual items.
For example, let’s say we are interested in measuring job satisfaction and one of our key variables is job-related depression. This might be difficult to measure with simply one question. Instead, we can create several different questions that deal with job-related depression and create an index of the included variables. Let’s say we have four questions to measure job-related depression, each with the response choices of "yes" or "no":
- "When I think about myself and my job, I feel downhearted and blue."
- "When I’m at work, I often get tired for no reason."
- "When I’m at work, I often find myself restless and can’t keep still."
- "When at work, I am more irritable than usual."
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 take advantage of differences in intensity among the indicators of a variable. The most commonly used scale is the Likert scale, which contains response categories such as "strongly agree," "agree," "disagree," and "strongly disagree." Other scales used in social science research include the Thurstone scale, Guttman scale, Bogardus social distance scale, and the semantic differential scale.
Let’s say, for example, that we are interested in measuring prejudice against women. One way to do that would be to create a series of statements reflecting prejudice ideas, each with the response categories of "strongly agree," "agree," "neither agree nor disagree," "disagree," and "strongly disagree." One of the items might be "women shouldn’t be allowed to vote" while another might be "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 summed for each respondent to create an overall score of prejudice. If we had 5 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.
Scales and indexes have several similarities. First, they are both ordinal measures of variables. That is, they both rank-order the units of analysis in terms of specific variables. For example, a person’s score on either a scale or index of religiosity gives an indication of his or her religiosity relative to other people.
Both scales and indexes are composite measures of variables, meaning that the measurements are based on more than one data item. For example, a person’s IQ score is determined by his or her responses to many test questions, not simply one question.
Even though scales and indexes are similar in many ways, they also have several differences. First, they are constructed differently. A scale is constructed simply by accumulating the scores assigned to individual items. For example, we might measure religiosity by adding up the number of religious events the respondent engages in each month. A scale, on the other hand, is constructed by assigning scores to patterns of responses with the idea that some items suggest a weak degree of the variable while other items reflect stronger degrees of the variable. For example, if we are constructing a scale of political activism, we might score "running for office" higher than simply "voting in the last election." "Contributing money to a political campaign" and "working on a political campaign" would likely score in between. We would then add up the scores for each individual based on how many items they participated in and then assign them an overall score for the scale.
References Babbie, E. (2001). The Practice of Social Research: 9th Edition. Belmont, CA: Wadsworth Thomson.
Babbie, E. (2001). The Practice of Social Research: 9th Edition. Belmont, CA: Wadsworth Thomson.