0	Nominal members. Most members are members in name only, having done little more than register with the party and pay party dues, if dues are collected. (This code was used for non-membership parties also.)
1	Middle code assigned in cases of bimodal distributions.
2	Marginal members. Most members exhibit at least some interest in the party as evidenced by having attended an occasional meeting or having performed some activity on behalf of the party.
3	Middle code assigned in cases of bimodal distributions.
4	Participants. Most members are regular attenders of party meetings and occasionally perform activities on behalf of the party.
5	Middle code assigned in cases of bimodal distributions.
6	Militants. Most members attend virtually all party meetings and constitute a ready source of manpower for performing party activities, in which they frequently participate.

Coding Results. The participation of party members in party activities is not treated systematically in the literature, and we succeeded in coding barely 60 percent of our parties on BV1102. The picture conveyed by the data in Tables 12.2a and 12.2b is one with many shades of difference. While it is true that more than one-third of the parties claims mostly nominal members, nearly 15 percent are served by eager militants. The spread between these points is level enough to be encouraging about the discriminatory capacity of the variable, and the absence of any relationship between BV102 and AC1102 relieves concern about any bias in our coding due to data quality. Remember, however, that all the nonmembership parties are automatically scored 0 on "membership participation," which builds in a relationship between BV1101 and BV1102. For correlational analysis between these two variables, one may want to filter out the twenty or so parties in each time period that had no membership requirements so that the resulting correlations reflect empirical rather than logical relationships.

The motivating influence of patronage in political parties, especially American political parties, has been the subject of considerable debate among parties' scholars. According to Clark and Wilson, patronage constitutes one form of "material incentive" that can be used to motivate individuals (1961, p. 134). As a class, material incentives are defined as tangible rewards that either have monetary value or can be translated into monetary value. They are to be distinguished from two other types of incentives: (1) solidary incentives, which are basically intangible with no monetary value and derived largely from the act of associating with friends, tend to be independent of the precise ends of the organization and differ in this respect from (2) purposive incentives, which derive explicitly from the stated ends of the organization.

Clark and Wilson recognize that organizations typically rely on all three types of incentives for motivating members, but they suggest that organizations can be distinguished by the form of incentive on which they principally rely. Within this threefold categorization of incentive types, the existence and extent of material and purposive incentives serve to differentiate political parties from one another far more than solidary incentives, which appear to be present in some degree in all parties.

Certainly it can be expected that the types of incentives operative would differ depending on the category

go to page 129