Kenneth Janda

Performance, leadership, factions and party change: An empirical analysis

This article reports the first empirical findings based on data from a major study of party change. Hypotheses are developed linking party change to both internal and external factors. The data provide support for the conclusion that electoral performance alone is not sufficient as an explanation for parties’ decisions to change, and that new leaders and/or dominant factions may indeed make a difference. This leads the authors to suggest that ‘the burgeoning field of theoretical and empirical work on party change should focus even more attention on internal decision‐making processes’.

Changes in Party Identity : Evidence from Party Manifestos

Political folklore holds that political parties often try to change their images following a disastrous election defeat. This paper inquires into the truth of this common assumption through a systematic analysis of manifestos promulgated by eight parties in Britain, Germany and the USA prior to national elections in the 1950s through 1980s. Each election was classified as triumphal, gratifying, tolerable, disappointing or calamitous from the standpoint of each party. The change in party images for adjacent elections was assessed by correlating the percentages of sentences devoted to standard political themes in the pair of manifestos. We tested the hypothesis that parties were most likely to change their policy images following disappointing or calamitous elections. Our findings suggest that poor electoral performance was not a sufficient condition to produce a major overhaul of party images, but poor performance in the prior election was virtually necessary to produce major change in policy packaging at the next election.

Formalizing and Testing Duverger's Theories on Political Parties

Maurice Duvergers Political Parties, written more than three decades ago, remains the most prominent source of hypotheses on parties and party systems. Although many years have passed since its publication, no one has formalized Duvergers main hypotheses on political parties and subjected them to empirical test. This article identifies Duvergers key concepts on party structure, links the concepts in 19 formal bivariate propositions, operationalizes the concepts using data from a worldwide sample of 147 parties in 53 countries, and tests all 19 propositions. Twelve are supported by the cross-national empirical test. Interrelationships among these 12 bivariate propositions are shown in a causal diagram, and suggestions are made for moving beyond Duvergers bivariate thinking to more powerful multivariate theorizing about the causes and consequences of party structure.

Ecology of Party Strength in Western Europe A Regional Analysis

Following the Lipset-Rokkan cleavage approach, we present an ecological analysis of the electoral outcomes at the regional level of the political parties in 16 European democracies. The search for relationships between voter alignments and the social structure is conducted in terms of a comparative ecology model. Ecological factors at the regional level within each country account for 75% of the variance in support for 93 parties over three elections during the 1970s. More than half of the “regional” variance could be explained by five “structural” properties of the regions: industry, agriculture, affluence, religion, and ethnicity. The impact of these structural properties varied across countries and across party types. Some of the more theoretically interesting variations are discussed for specific countries, individual parties, and party types.

How well does ‘region’ explain political party characteristics?

Abstract Region is presumed to have an ‘effect’ on the characteristics of political parties, causing parties in the same region of the world to be ‘different’ from parties in other regions. The nature and extent of differences in party characteristics within and between regions have not previously been systematically investigated. This paper applies the analysis of variance to 11 organizational characteristics of 147 political parties. The parties came from 53 countries representing a stratified random sample of party systems in 10 cultural-geographic regions of the world. The study finds significant differences between the regional groupings of parties on all characteristics, with region predicting from 11 to 52 per cent of the variance in individual party traits. A separate discriminant analysis of parties grouped into a First-Second-Third World typology shows that 90 per cent of the parties can be correctly classified in ‘their’ world on the basis of their organizational characteristics.

Managing qualitative information and quantitative data on political parties

Abstract Techniques are described for managing information collected for a project to compare 158 political parties from 53 countries during 1950–1962. Information obtained from bibliographic searches and correspondence was stored on microfilm; the MIRACODE system was used for retrieval. Researchers scored the parties examined on thirteen different issues along a scale ranging from +5 (leftist) to − 5 (rightist). Each code assigned was accompanied by a discussion of the coding judgments and a code to indicate adequacy of the information and the researchers degree of confidence. The use of the system is demonstrated with reference to the two major parties in the US and UK, the Communist Party of the Soviet Union and the (now defunct) Portuguese National Union. The accuracy of the coding is evaluated by comparing the ratings of experts from the US and Soviet Union.

Keyword Indexes for the Behavioral Sciences

T ioral sciences, inform a t’ ion retrieval schemes involving tlie use of computers probably appear interesting and ambitious but still of dubious utility for present-day research problems. wliile it is true that much of the really exciting work being done by information retrieval specialists is still experimental, some of their tecliniques have long been operational and have already been applied to practical problems of scientific research. Computergenerated keyword indexes to current resmrch literature, for example, have found standard usage within the physical sciences, at l e s t as a partial solution to the problem of keeping abreast of publications in one’s field. The success of this general method of using computers to prepare keyword indexes to bibliographical material within the physi~al sciences has been amply proven; tlie practical applications of keyword indexing within the behavioral scienccs are no less promising. Keyword indexing in any discipline operates on the assumption that ccrtain tenns or “keywords” constitute useful handles for pulling out or retrieving information of interest to the researcher. An ordinary book index serves as a good example of this assumption. The preparation of any index involves many mechanical tasks: scanning pages for important terms, recording page numbers, cross-listing, and alphabetizing. Familiarity with the subject is undoubtedly an asset in preparing an index, but a g d job can ordinarily be expected from any tliorough, accurate, and efficient resmrch ‘. assistant \vho is instructed in advance about the terms to be included in the index. Computers, which are notoriously thorough, accurate, and efficient, possess tlie requisite virtues for routinized keyword indexing, and need only be instructed liom to do the job. Computer programs providing these instructions have been available now for some time? By mcans of the proper program, a computer can be instructd to read natural language material, search the material for the appevancc of pre-defined keywords, alphabetize the keywords found, and print out tlie alphabetized keywords along with Keyword Indexes

Selective Dissemination of Information A Progress Report from Northwestern University

A computerized system for automatically notifying social scientists of new journal articles that appear to fit their personal interests is now in operation at Northwestern University. Kenneth Janda is Associate Professor of Political Science at Northwestern, and Gary Rader, is a former undergraduate major in the Political Science Department.

Manifestos and the “two faces” of parties: Addressing both members and voters with one document

It is commonplace to see references to parties’ manifestos as their written issue “profiles,” and changes in such documents as constituting changes in the parties’ “images” or “identities,” with the latter terms often used interchangeably to capture the role of platforms. This article argues, however, that projection of a party’s “image” and its “identity” are two different functions for a manifesto, not just one, and that it is important for the building and testing of theory that this distinction be maintained. Parties are, after all, addressing two audiences simultaneously with one document, and the two dimensions provide two alternative objects of change which can be used strategically to please both audiences at once. The article employs existing manifesto-based measures of parties’ relative issue emphases and their positions on a range of issues as indicators of image and identity, respectively, and finds that the two are indeed empirically distinct. Then, an earlier test of the electoral performance hypothesis as applied to emphasis change is replicated with data designed to capture change in issue positions. The test provides evidence for the prudence of maintaining the distinction between emphasis and position as two different dimensions of party profile change.

A Note On Measures of Party System Change

In this symposium Mogens Pedersen argues against measuring party system change by comparing static measures of “fractionalization, ” and proposes a measure based on changes in party strength from time 1 to time 2. Shankar Bose proposes a related measure of party system change that combines changes in strength with changes in party continuity over time. This article compares their measures as applied to data from ten party-systems.