Coding Results. Surprising as it may seem, the parties' literature offers no clue concerning membership requirements in one quarter of our parties. The situation that obtains in the other three quarters is described in Tables 12.1a and 12.1b. The most typical requirement for membership is a combination of registering as a member (code 1) and paying dues (code 2), which produces a summed code of 3, earned by half of the world's parties. The next more typical situation, accounting for about one-fifth of the parties, consists of no membership requirements at all. This relationship is tainted, however, by the strong correlation of .40 between BV 1101 and AC 1101, which hints at our propensity to deny the existence of membership requirements when the literature is silent on the subject. Although a few parties demonstrate nonscale types (i.e., codes 2 and 5 in the distributions), the vast majority conform to the assumption of cumulative properties in the operationalization. One might note finally that more than 10 percent of the world's parties do not admit just anybody as members; they must first prove their mettle to the party's satisfaction through a probationary period (included in codes 5 and 7) before the affiliation is consummated.

Duverger argued for distinguishing among party members according to their "degrees of participation," stating that "supporters, adherents, militants, propagandists, form a series of concentric circles of ever-increasing party solidarity" (1963, p. 61). A similar continuum based on the degree of participation underlies Milbrath's "hierarchy of political involvement," which classifies political actors as "apathetics," "spectators," "transitionals," and "gladiators" (1965, p. 18). Other writers have referred to differences in degrees of political participation in a similar manner, and differences in participation obviously indicate differences in involvement in party affairs.

Perhaps the most useful set of categories for assessing party participation-as opposed to political participation in general-is that offered by Barnes, who distinguishes among nominal members, marginal members, participants, and militants (1966, p. 351). The basis of Barnes' distinction lies in the members' participation in party meetings and other activities, but these categories are more readily applied to parties with formal membership requirements than to parties without formal requirements. Indeed, there seems to be no way to apply this categorization to parties without formal membership requirements,, except to treat such parties as those which have formal membership requirements but for whom the most common type of member is the "nominal" member. Thus the nominal membership category constitutes the "least involved" category of membership participation.

Operational Definition. We recognize Barnes's set of categories as points along an ordinal scale, and we characterize parties according to the tendency of their members to "center" on one of these points in the continuum. Our choice for a measure of central tendency is the mode, which is the most frequent or most common type of participation by party members. We score the party by entering the code associated with the modal category of membership participation. In the case of a party with a bimodal distribution--for example, most of the party's members are distributed relatively equally across two categories--we assign the party the code located midway between the categories.

go to page 128