Path: Janda: Political Parties, Home Page > Part 1: Table of Contents > Chapter 14
Chapter 14: Validating the Conceptual Framework (pp. 135-161), p. 160
(you can navigate to other pages by clicking on page numbers below)
p. 135
p. 136
p. 137
p. 138
p. 139
p. 140
p. 141
p. 142
p. 143
p. 144
p. 145
p. 146
p. 147
p. 148
p. 149
p. 150
p. 151
p. 152
p. 153
p. 154
p. 155
p. 156
p. 157
p. 158
p. 159
----
p. 161
 
TABLE 14.9
Lack of Discriminate Validation among Indicators for Degree of Organization and Involvement
Concepts and Indicators
Mean Correlations with
All Other Indicators for
Degree of
Organization

Involvement

Degree of Organization

8.01

Structural Articulation

0.36

0.24
8.02

Intensiveness of Organization

0.45
<
0.47
8.03

Extensiveness of Organization

0.41

0.17
8.04

Frequency of Local Meetings

0.42
<
0.43
8.06

Maintaining Records

0.47

0.33
8.07

Pervasiveness of Organization

0.44

0.41

Mean =

0.42

0.34

Involvement

11.01

MembershipRequirements

0.39
>
0.35
11.02

Membership Participation

0.39

0.48
11.03

Material Incentives

0.18

0.34
11.04

Purposive Incentives

0.32

0.49
11.05

Doctrinism

0.41
=
0.41

Mean =

0.34

0.41

two organizational indicators are less obvious. While one might expect involvement to be a consequence of intensive organization and frequent local meetings, the linkage appears more theoretical than conceptual.

This problem with measuring degree of organization and involvement remains unresolved at present and signals a difficulty with the conceptual framework in discriminating between degree of organization and involvement. Although the indicators within each scale tend to intercorrelate more highly among themselves than they do across the scales, this is not true of particular indicators, and one should be alert to this in using the data to test theory employing these concepts. Although the scales of degree of organization and involvement did not satisfactorily pass the test of discriminate validation, the other pairs did, signifying that otherwise the concepts in the framework are analytically distinct from one another although empirically correlated to varying degrees.

Theoretical Utility

Although the proposed conceptual framework has theoretical aspects in its predictions of relationships among the basic variables subsumed under each major concept, it is conceptual rather than theoretical because it does not elaborate on expected relationships among the major concepts. It is distinctly pretheoretical, however, in that the formulation of an explicit conceptual framework is a necessary beginning for successful theory construction. The utility of the conceptual framework will never be convincingly demonstrated until it is used to construct valid and fruitful party theory. With the empirical applicability and measurement validity of the conceptual framework established to a reasonable degree of satisfaction, attention will next be given to using its concepts to explain the origins and characteristics of political parties and the effects of political parties on political systems.

As examples of theoretical concerns that can be illuminated with use of these concepts, measures, and data, consider (1) the relative effects of "system" and "party" factors on the characteristics of political parties--as explored by Duverger--and (2) the contributions of social diversity to electoral success--as proposed by Kirchheimer. Duverger, for instance, proposed explanations for the degree of centralization within a party utilizing both "system" factors (e.g., existence of federalism, nature of electoral system, separation of powers) and "party" factors (e.g., ideological orientation, conditions of origin, social composition). Since Duverger's theoretical explorations (1954), explanations of party centralization have rarely risen above the level of bivariate propositions, and even these have not been adequately tested with reliable measures of centralization of power or other party characteristics. The concepts, measures, and data of the ICPP project, however, provided the ingredients for a comprehensive test of propositions advanced to explain party centralization that have arisen in the literature to complement Duverger's. Using ICPP data, Harmel (1977) found that 68 percent of the variance in centralization of power within parties across the world was associated with "system" characteristics peculiar to given countries, leaving relatively little left to

go to page 161