Return to: Search Page or to: Table of Contents Vol. 10, issue 3

Cem Baslevent, Hasan Kirmanoglu and Burhan Senatalar, "Voter Profiles and Fragmentation in the Turkish Party System," Party Politics, 10 (May 2004), 307-324.

First Paragraph:
The general elections of November 2002 were a turning point in Turkish politics for various reasons. With 34 percent of the vote, the Justice and Development Party (Adalet ve Kalkinma Partisi; AKP) won nearly twothirds of the seats in parliament. This was the first time a pro Islamist party had won enough seats to form a single-party government. The centre-left Republican People's Party (Cumhuriyet Halk Partisi; CHP) became the only party other than the AKP to surpass the 10 percent nationwide threshold for representation. This was not only the first time since the 1954 elections that Turkey had had a two-party parliament, but also the first time that the incumbent parties had been totally wiped out of parliament in the general elections.

Figures and Tables:
Figure 1. Effective number of parties in Turkey (1950-2002)
Table 1. Multinomial logit estimation results of the vote intention function
Table 2. Tests of the pairwise comparison of voter profiles by category
Table 3. Tests of the pairwise comparison of voter profiles by variable

Last Paragraph:
(First Paragraph in Conclusion) The purpose of this article was to compare the voter profiles of major political parties in Turkey to test the common belief that fragmentation in the Turkish party system was an 'artificial' rather than a 'natural' one. Restriction tests based on the multinomial logit estimates of an individual vote intention function revealed that there are significant differences between the parties on the centre-right and the centre-left as far as their voter profiles are concerned. The two pro-Islamist parties, on the other hand, are found to have relatively similar voter profiles. However, in all three cases considered, significant differences between voter profiles resulted not only from socio-demographic controls or economic evaluations, but also from what we called the issues and identity variables. Since those variables were meant to represent the dominant cleavages in Turkish politics, we could interpret this finding as evidence against the hypothesis that the party system is artificially fragmented.

updated November 2013