The same arguments made in behalf of the frequency of local meetings as an indicator of the degree of party organization apply to frequency of national meetings as an organizational variable. The more frequent the meetings at the national level, the greater the potential for party organization.

In conceptualizing this variable, we limit our concern to "governing" agencies of the "national committee" and "council" types. (See the distinction between governing and administrative agencies and the discussion of types of organs in each category as presented for variable 8.01.) This means that we count only the meetings of these national party organs. We do not count meetings of the "inner" executive committees, "politburos," or secretariats, which meet more often.

Operational Definition. A party is scored with the highest appropriate code along the following scale of frequency of meetings:

0	There either are no local party organs or the local units do not meet.
1	Basic party units do meet, but rarely and irregularly.
2	Basic units meet during campaigns only.
3	Basic units meet once or twice a year.
4	Basic units meet three to six times a year.
5	Basic units meet seven to eleven times a year.
6	Basic units meet once a month or more.

Coding Results. The parties' literature is more apt to comment on the frequency of national than local meetings, but the difference is not enough to promote notably greater success in coding BV805, which was omitted for about half of our parties. The data that were obtained are reported in Tables 9.5a and 9.5b, which display considerable variance in the frequency of national party meetings. Code 3 emerges as the modal category by a small margin, establishing the fact that about one-fifth to one quarter of the national committees meet two or three times a year. Although the coders were specifically instructed to limit their coding to national committees as defined above, they sometimes lapsed from the straight and narrow and seized upon meetings of "executive committees" and even party conferences on occasion. I hope that these mistakes have all been caught and corrected so that the final codes are faithful to our operationalization as intended.

This variable contains three basic components that, taken together, constitute a broad conception of maintaining records, an activity thought to be indicative of structural differentiation. We take the most important indicator of record keeping to be maintaining lists of party members or party activists. The second component of record keeping is the preparation and publication of party propaganda--for example, speeches and policy statements--for either internal or general distribution. The third component is the maintenance of

---

*Gilbert Rotkin assisted in writing this section.

**Donald Sylvan assisted in writing this section.

go to page 105