Skip to main content icon/video/no-internet

Guttman scaling (also called scalogram analysis) is a method of scaling involving items that reflect increasing levels of extremity on a single dimension of interest. This procedure was developed by Louis Guttman and was introduced in response to the concern that prior attitude measures might sometimes tap multiple constructs. For instance, a scale might include (among others) the statements presented in Table 1.

These statements both represent negative views toward obesity. Item 1 is the most extreme, and it seems reasonable to expect that individuals who endorse it might also endorse Item 2. However, it is plausible that an individual might have an attitude toward obesity based solely on the associated health risks. This person might agree with the more extreme (health-related) item, but disagree with the less extreme attractiveness item. A scale containing these types of items would be assessing more than one construct. Therefore, it would be difficult to determine whether two people receiving the same score actually reported the same attitude, or whether their scores were driven by different dimensions. Guttman's approach was designed to correct this potential problem.

Guttman Scale Criteria

A Guttman scale is composed of a set of statements to which respondents indicate their level of agreement. Because these scales are meant to assess a single dimension, items are selected to vary on extremity alone. Therefore, statements chosen must be unipolar, reflecting gradations of either support or rejection. It is expected that individuals will agree with all items leading up to their most extreme endorsement and will disagree with items of higher extremity. With this goal in mind, researchers aim to select items that will fit a stepwise sequence of extremity. In fact, successful Guttman scales often consist of nested statements, such that agreeing with any one item almost necessitates agreement with items of lower extremity. This is a useful strategy because individuals with extreme attitudes might otherwise reject more moderate items on the basis that they denote too weak a stance. For example, the statements presented in Table 2 might be used to examine respondents' attitudes toward the film Casablanca.

Table 1 Obesity Items Reflecting Two Dimensions
Items
Obesity is a serious health threat that leads to early death.
Obese individuals are less physically attractive than people who aren't overweight.
Table 2 Sample Guttman Scaling Items
Items
Casablanca is the greatest movie ever made. (most extreme)
Casablanca is one of the top 5 movies of all time.
Casablanca is an exceptional film, one of the best.
Casablanca is a great film, well above average.
Casablanca is a good movie. (least extreme)
Table 3 Initial Response Matrix (Items by Respondents)
Items
Person # 1 2 3 4 5 Total Score
1 0 0 0 1 1 2
2 0 0 1 1 1 3
3 1 1 1 1 1 5
4 1 0 0 1 1 3
5 0 0 0 0 1 1
6 1 0 0 0 0 1
7 0 0 0 1 1 2
8 0 1 1 1 1 4
9 1 1 1 1 1 5
10 0 1 1 1 1 4

To satisfy the requirements of a Guttman scale, individuals who agree with Item 1 should also agree with all other statements listed because they represent less extreme views. Likewise, respondents who disagree with Item 1 but agree with Item 2 should endorse all remaining items, and so on. This not only ensures that scores obtained are interpretable (because individuals with the same score will have endorsed the same statements), but it also introduces the possibility that knowledge of individuals' scores alone will allow exact replication of their responses. This “reproducibility” is a key element of Guttman scaling.

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Sage Recommends

We found other relevant content for you on other Sage platforms.

Loading