year to year, 1950 to 1978. The results of these calculations for the party systems are given in Table 15.12, which ranks the countries according to increasing levels of party instability. Other measures would no doubt give somewhat different rankings, but most would produce comparable results.

The average country in our sample experienced an annual shift of ten percentage points in party representation in the legislature. This figure does not take into account the frequency of elections or the number of parties, but neither of these factors bears any necessary relationship to the party instability score. Systems with many parties and frequent elections can demonstrate a high degree of stability, which occurs in the case of the German Democratic Republic (East Germany). Uganda, on the other hand, had only two legislative elections but sports one of the highest instability scores. The main factor in the calculation of a high instability score is simply large changes in the percentage of seats held from one year to the next. Thus the most stable party systems were found in Eastern Europe; no change at all in party representation was experienced in Albania. The United States showed more stability than the average country, while Canada was ranked among those showing greater amounts of change in party representation from year to year.

The ranking of countries in Table 15.12 invites speculation about the patterns of party change over time in each country and about the causes of party system instability. An inquiry into the causes of party system instability is clearly beyond the scope of this book, but the data contained herein should help advance its study. The patterns of party change in each country over time, while not a concern of this chapter, are considered in detail in Part Two, which treats party continuity and change for each of the 53 countries, graphing changes in party legislative representation over time for each country and reporting its system instability score from Table 15.12. These graphs disclose details of changes in party fortunes and illustrate differences among countries with similar instability scores.

Despite the limited objectives of this analysis of continuity and change in political parties, we can offer the following summary comments about party stability. Depending on the environment in which they were created, political parties vary substantially in their likelihood of survival. Although only 61 percent of the parties in our worldwide sample that were active in 1950-1962 survived until 1979, about 80 percent of the parties persisted in certain regions. Parties in Eastern and Western Europe, Anglo-American countries, and South America were particularly hardy.

This is not to say that the party systems in these regions have not undergone changes. Western Europe, for example, ranks high on party continuity but also dem-

go to page 169