Guttman scale

from Wikipedia, the free encyclopedia
This article or the following section is not adequately provided with supporting documents ( e.g. individual evidence ). Information without sufficient evidence could be removed soon. Please help Wikipedia by researching the information and including good evidence.

The Guttman scale is a scaling method in empirical social research with which attitudes to a thing (e.g. people, groups of people, behavior) are recorded. It is named after the social researcher Louis Guttman .

To determine values ​​on the Guttman scale, the respondents are presented with several statements that they should affirm or negate. These are arranged in such a way that a very general statement is at the beginning and an extreme one at the end. If you answer in the affirmative, you usually do the same for all of the previous statements. A Guttman scale where this always occurs is called perfect. Generating such a scale is often difficult and only possible with a few statements that need to be evaluated.

example

A variant of the Guttman scale is the social distance scale , which was introduced in 1925 by Emory S. Bogardus . It measures the respondent's distance from a certain group, e.g. B. a social minority . A member of a group is asked for willingness to do this

  1. to be admitted as a visitor to the home country;
  2. to be admitted as fellow citizens;
  3. accept as work colleagues;
  4. to accept as neighbors;
  5. to have as a friend;
  6. to marry into the family.

Whoever answers question 3 with yes should also answer questions 1 and 2 in the affirmative.

Reproduction coefficient

A so-called reproduction coefficient is derived from the ratio between appropriate answers (i.e. consents that build on each other and are not interrupted in their “logical chain”) and the total number of all answers. Inappropriate answers would therefore be those that break this chain:

“Yes, Muslims are allowed to live in this residential area. But no not in this country ”.

This is a measure of how well the set of questions (in this case the measuring instrument) depicts reality, the empirical relative , in a homomorphic manner in a numerical relative (numerical scale of approval / disapproval) and a measure of how well the Question set measures the latent variable of consent.

The reproduction coefficient should be at least 0.9.

See also

literature

  • Raymond Gordon: Unidimensional Scaling of Social Variables: Concepts and Procedures. The Free Press, New York 1977
  • Tilman Betsch: Introduction to the Methods of Psychology University of Erfurt, Erfurt 2009