participation." However, there was no significant correlation between BV204 and AC204.
The extent of a party's representation in the national legislature is often used as an indicator of its governmental status, with status being a function of the proportion of seats held by the party. In reality, governmental status is not a simple function of legislative strength, for governments are sometimes formed by coalitions of parties, and small parties occupying strategic positions on the ideological spectrum are apt to gain power denied their larger competitors. Then, too, legislatures differ in terms of their role in the political system. When power resides elsewhere, party representation in the legislature may serve a purpose other than participation in policy making. These cautions notwithstanding, a party's legislative strength, interpreted as the proportion of seats held in the lower house of the legislature--regardless of its role in the system--is another indicator of governmental status.
Although it would undoubtedly be informative to determine party representation in the upper houses of bicameral legislatures as well as the lower houses, we opt only for the latter to aid cross-national comparisons. According to Banks and Textor, the countries of the world are about equally divided between unicameral and bicameral legislatures, with federalism almost always associated with bicameralism (1963, p. 111). Not only are half the countries without a second chamber, but the composition of the upper houses is often linked to structural features of federal states. There is, therefore, greater variance in selection processes for upper houses in bicameral systems than for lower houses, which commonly employ some form of popular election of members--as in most unicameral legislatures.
Our assessment of party representation is based on the proportion of seats held by the party in the years studied. The proportion is based on the number of years rather than the number of elections at which seats were decided so that loss of' seats through by-elections and defections might be more easily handled and, again, to facilitate cross-national comparisons, where the number of elections may vary considerably.
Operational Definition. Legislative strength is expressed by summing the proportion of seats that the party held in each year and dividing by the appropriate number of years for the time period. For parties and legislatures that have existed continuously during the time period, there is no difficulty in determining the appropriate number of years. But there are two methods that represent themselves when either the party did not participate in the legislature throughout or the legislature itself did not exist.
One method is to sum the proportions of seats that the party held each year during each hall' of the overall time period and then to divide each sum, respectively, by the number of years in that subdivision Another approach is to sum the proportions of seats as above but then to divide only by the number of years in the subdivision during which the party was represented or the legislature operated. Obviously, these methods give identical results under conditions of continuity, but under conditions of discontinuity the second method produces both (I) higher legislative strength scores and (2) lower legislative instability scores. Thus, if the party operated for only two years in a seven-year period (or if the legislature was proscribed for all but two years) and if the party received 40 percent of the seats when it did get the chance to contest legislative elections, the party would receive, under the second scoring procedure, a legislative strength score of .40 and an instability score of .00, based only on two years' participation in the legislature.