Attitude consistency and validity-reliability assumptions — An exercise in methodological self-criticism
✍ Scribed by Pablo Suárez
- Book ID
- 104640586
- Publisher
- Springer Netherlands
- Year
- 1980
- Tongue
- English
- Weight
- 925 KB
- Volume
- 14
- Category
- Article
- ISSN
- 0033-5177
No coin nor oath required. For personal study only.
✦ Synopsis
The aim of this paper is to question some "normal-science" procedures and assumptions in the construction and interpretation of composite indices. The analysis is illustrated by reference to an attempt (Suarez, 1975) to measure fear towards Communism as aroused by the 1970 pre-election propaganda in Chile, known as the "terror campaign" (see, e.g., U.S. Senate, 1975).
The problem was faced by registering the expressions of fear towards Communism the subjects (N = 952) manifested in connection with the issues the campaign agitated most. On the basis of a content analysis of the propaganda, six sentences were worded expressing an anti-Communist statement and other six a pro-Communist one. The subjects were then asked whether they agreed or not with each sentence. The intention was to develop an overall index measuring [ 11 the multidimensional patterning of fear which we assumed as being the result of the multi-sided appeal of the propaganda as well as of previous anti-Communist socialization.
However, in absence of a guiding theory any index is arbitrary. The lack of theory forces the definition of alternative measures but no possible decision can be taken between them. Resulting from the combination of criteria related to the ambiguous interpretability of residual answers, such as "don't know" (cf., Galtung, 1970, pp. 94-102) and the possible wording-bias, fifteen measures were defined as in Table I. These alternative definitions [2] are needed to handle the indetermin-* This paper is a slightly modified version of one presented at the 9th World Congress of Sociology,