|
|||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||
higher the governmental status; the greater the negative discrimination, the lower the governmental status. The last two variables (like variable 1.07) were eliminated from further consideration on the grounds that they lack face validity as measures of governmental status. "Outside origin" was initially included because of the prominence of Duverger's proposition (1954, pp. xxiv-xxvii) that links the origins of parties to their role in government (see LaPalombara and Weiner 1966, pp. 7-14). But, upon reflection, this is seen to involve hypothesizing about the effects of party origins rather than claiming that "inside" origin manifests governmental status. "Percentage of pages indexed" was excluded on the same grounds argued for "number of pages indexed" in the institutionalization cluster; both were more direct measures of scholarly tendencies than of party attributes. Among the remaining six items, expectations were that they would all be positively and strongly intercorrelated. This was in fact the case for all the variables, but the correlations that involved national orientation tended to be consistently smaller (averaging only .34) than the others. A factor analysis of the correlation matrix for the remaining five variables showed that the first factor explained 70 percent of the variance among the variables. The alpha coefficient of reliability for the ''governmental status'' scale composed of these five items in standard form was .92, which suggests that it is highly reliable in the internal consistency sense. As for its face validity, at present one is left to decide whether parties that are favorably treated by the government, that monopolize government leadership, that continuously occupy cabinet positions, that hold high percentages of legislative seats, and that win high percentages of electoral votes are indeed high on "governmental status." Social Support. There are various ways of conceptualizing the social bases of party support. One common conceptualization links parties with specific social groups and describes them as being supported, to varying degrees, by "workers," "farmers," "business," "Catholics," "educated classes," and so on. Because the particular groups involved in such classifications depend largely upon common social structures in the societies being studied, categories of party support based on this conceptualization are not very rewarding for cross- cultural research. The analysis of party support by specific social groupings is important, but it seems better suited to comparative analysis of parties in similar cultures. The original conceptual framework proposed the concepts of "social aggregation" and "social articulation" for measuring the social bases of party support under the assumption that the process of interest aggregation and articulation (as formulated by Almond and Coleman 1960) followed from the underlying structure of social support. In an effort to keep these social structure concepts distinct from those associated with the process involved in gathering and articulating interests, the more descriptive term "social attraction~~ was subsequently formulated to replace "social aggregation," and "social concentration" replaced "social articulation." Furthermore, these two concepts were subsumed under the broader heading of "social support," which can accommodate other conceptualizations, including those pertaining to specific social groups. As reconceptualized, social support in the ICPP project related to Kirchheimer's idea of the "catchall" party, which deemphasized "the classe gardee, specific social-class of denominational clientele, in favor of recruiting voters among the population at large" (1966, p. 190). The concepts of social attraction and concentration approach the measurement of party support from two different perspectives, asking (1) "How well does the party attract different types of supporters?" and (2) "How much of the total party support is concentrated in a single type of supporter?" During the course of data collection, it became evident that a third question also needed to be asked: (3) How well does the distribution of supporters within the party reflect the distribution of social types within the country? Previous descriptions of "catchall" parties as socially "heterogeneous" or "diverse" rarely specified which perspective |
|||||||||||||||||||||||||||