Selections from The Concept of Political Culture.
There are three main problems of comparability of political culture among nations, namely the problem of complexity, the problem of salience, and the problem of indexicality.
The problem of complexity is an exacerbated case of the problem of `many variables, small N', meaning that the number of variables is too large for the sample size to uniquely determine the contribution of each term to the model.
The problem of salience refers to the fact that politics and political objects themselves have varying salience across nations, which means that attitudes towards them cannot be directly compared. A. L. Craig and W. A. Cornelius give two examples of this problem in relation to the measurement of Mexican political culture. One is that Almond and Verba's measures of political knowledge are based on questions that in the Mexican context would be somewhat esoteric, owing to the regional nature of the political conflict there. Another is that measures of ability to influence political outcomes ignores the fact that, in Mexico, influence is usually directed to the `rule application' rather than the `rule making' stage. A point very similar to the latter one has been made in connection with the study of Soviet political culture. F, Barghoorn had characterized Soviet political culture as `subject participatory', a modification of Almond and Verba's typology designed to draw attention to the combination of low levels of `subject competence' combined with `participation directed ultimately from the political center at the top of the [Party] command structure'. Taking issue with this characterization, W. DiFranceisco and Z. Gitelman demonstrate the use of informal channels and means of influence, such as the use of connections and bribery.
The problem of indexicality is a particular severe form of the problem of salience. In philosophical logic, `indexicality' refers to phrases such as `my hat'. whose denotation is indexed to the utterer, and possibly to the time and place of utterance as well. Indexical responses to questions clearly necessitate, on the part of the inquirer, an act of contextualization if they are to be understood. We will use the term `indexicality' in a somewhat extended sense, to refer to responses that require a degree of contextual interpretation that threatens their comparability. For instance, a response that indicated a high level of participation in politics would, as we just saw, need to be contextualized in order to make it comparable; we would need to know whether the decision-making stage or the implementation stage was the more usual target for participation in the context we were or figure from, say, Mexican history. Then the problem becomes one of indexicality. An even larger, potentially prohibitively large, amount of interpretation is necessary in order to render such responses comparable.
The complete text was published in:
The Concept of Political Cultures
Stephen Welch, 1993.