386 Guide to Journal Articles

Kevin Featherstone, 'Elections and Parties in Greece' , Government and Opposition, 17: 2, Spring 1982, pp. 180 194. The aim of this article is to examine the contemporary Greek party system and the results of the 1981 elections after a consideration of the philosophy and leadership of the two main p a r t i e s - - P A S O K and New Democracy.

William H. Flanigan and Nancy H. Zingale, 'Western European and Anglo-American Party Systems: A Dimensional Analysis' , Comparative Political Studies, 14: 4, January 1982, pp. 481 515. The article looks at the major directions that spatial analysis of party systems has taken. A technique for determining the locations of parties-in-the-electorate in a social space is introduced and applied to 14 Western European and Anglo-American party systems.

William Fraeys, 'Les +lections l+gislatives du novembre 1981. Analyse des r&ultats ' , Res Publica, XXIV: 1, 1982, pp. 129 149. Fraeys examines the results of the 1981 general election in Belgium at the national and regional levels, both in terms of party vote and seats in the Chamber, Senate and Communal and Regional Councils.The main trends were a heavy defeat for the Social-Christians and F D F / R W and dramatic gains for the Liberals and (to a lesser extent) the Volksunie. Fraeys offers the tentative conclusion that the Social-Christian losses were mainly from its right wing---those discontented with socio-economic policies shifted to the Liberals across the country and to the UDRT in Brussels and Wallonia, those in disagreement with its communal policy switched to the Volksunie in Flanders. But Fraeys warns against making premature predictions of an onset of left-right polarization in Belgian politics. See also: entry for Res Publica.

Daniel Gaxie, 'Mor t et r&urrection du paradigme de Michigan. Remarques sur quelques r&ultats r&ents de la sociologie des comportements politiques aux Etats-Unis' , Revue

fran(aise de science politique, 32: 2, April 1982, pp. 251-269. The author refers to recent works (particularly those of Nie, Verba and Petrocik) which suggest that the American voter has changed and that the Michigan paradigm was only valid for one specific period of American history. This article argues that old and new results are not necessarily incompatible and that they can be integrated into a model taking into account both the individual characteristics of the social agents and the working of the political field.

Donald A. Gross, 'Units of Analysis and Rae's Fractionalization Index', Comparative Political Studies, 15: 1, April 1982, pp. 85 98. The author examines the relationship between Rae's fractionalization index and the units of analysis to which it is applied. The analysis is based on data pertaining to US congressional elections between 1824 and 1978. The level of aggregation is found to affect the value of Rae's index.

Patricia A. Hurley, 'Collective Representation Reappraised' , Legislative StudiesQuarterly, 7: 1, February 1982, pp. 119 136. This paper presents a critique and an extension of Weissberg's theory of collective representation. The author redefines collective representation as the extent to which policy outputs reflect nation-wide public preferences, Data used are the SRC/CPS election surveys of 1978 along with roll-call data from the 95th Congress.

International Political Science Review, 2: 4, October 1981, special issue devoted to 'Experimentation and Elections', edited by J. A. Laponce.