Bo Särlvik

Partisan Dealignment in Britain 1964–1974

Britain enjoys a textbook reputation as the historic home and model representative of a stable two-party system. From the factors most frequently cited by way of explanation – the electoral system, the absence of cross-cutting social cleavages – it is implied that this uncommon state of affairs is a natural and permanent part of British politics. This reputation is, in fact, somewhat exaggerated. At no time have MPs or parliamentary candidates in Britain been confined to two parties only (in contrast to the United States); and for most of the period since the introduction of the majority male franchise and the beginning of mass parties in 1884 the configuration of party forces in the Commons would be best described as multi-party (1884–1922), three-party (1922–31), or dominant one-party (1931–45). Britains experience of a stable two-party politics has therefore been both recent and relatively short-lived; it is only since the Second World War that two parties – Conservative and Labour – have alternated in exclusive incumbency of government office on the basis of an evenly balanced duopoly of electoral support and parliamentary seats.

Decade of dealignment : the Conservative victory of 1979 and electoral trends in the 1970's

Preface Note on documentation Note on tables Part I. Political Context and Electoral Change Ivor Crewe: 1. The flow of events 2. The flow of the vote 3. The lockgates on the vote Part II. Issues, Opinion and Party Choice in the 1979 Election Bo Sarlvik: 4. Opinions on political issues and voting - the directions of the enquiry 5. Why the parties were liked and disliked 6. Managing the economy - the Labour governments record and the Conservative alternative 7. Responses to social and cultural change 8. Nationalisation and social welfare policies - a rightward shift in electoral opinion 1974-1979 9. Policy alternatives and party choice in the 1979 election 10. Ambiguity and change in party positions: three special cases 11. The impact of the issues - an overall account 12. The electorate and the party system 13. The making of voting decisions: political opinions and voting intentions during the campaign Part III. A Turning-Point?: 14. At the end of a decade Postscript: realignment in the 1980s? Appendix. Constructing the flow-of-the-vote tables Notes Index.

Angels in Plastic: The Liberal Surge in 1974

Before the Liberal surge in 1974, survey research stressed that the Liberal vote was electorally volatile, socially representative, and negative in character. Data from 1974 indicates that the volatility of the Liberal vote owes more to the absence of a large core of stable Liberal voters than to any difference among parties in their ability to retain the votes of recent converts. Moreover, the small core of regular Liberal voters is unusually middle-aged and middle-class, socially very different from the larger and socially representative body of occasional Liberal voters. In the eyes of the electorate, the Liberal Party continues to have a diffuse image, largely devoid of any specific policy content. The Party benefited from dissatisfaction with the state of the country, but there is no evidence that an image of classlessness contributed to its electoral success. Moreover, while most Liberal voters did so for some positive reason, many of their reasons had more to do with style than policy, and the personalities of party leaders appear to have had much to do with moving people to consider—if not actually vote for—the Liberal Party.

Individual differences scaling and group attitude structures: British party imagery in 1974

Our purpose in this paper is to investigate the applicability of multidimknsional scaling techniques to problems of summarisation and representation of data generated in large surveys. What we are concerned with is largely the exploratory and descriptive stage of data analysis, and in particular the summarisation of data patterns: while coefficients of central tendency and dispersion, and frequency distributions, are in general satisfactory to describe the properties of one variable in isolation, there is frequently a need to describe the behaviour or “investigate the structure” of several variables at a time. The use of multidimensional scaling techniques in this area has become increasingly common (Coxon and Jones, 1974a) quite reasonably, as these techniques handle data patterns of greater complexity than do unidimensional scaling techniques, while allowing the investigator to make less rigorous assumptions about his data than is the case with factor analysis. The novelty of the approach we discuss in this paper means that some of the results presented must be treated with caution, and as exploratory. What we shall do is the following: (1) present the results of a non-metric multidimensional scaling of a number of disparate items from a survey and argue that the dimensions recovered are interpretable; (2) show that different subsamples use (weight) these dimensions differently in directions consistent with the definition of these subsamples; (3) to validate (2), show that when subgroups are geierated randomly or according to criteria that one would not expect to produce differences, differences in weightings of comparable magnitude do not occur.

European elections: Sweden

A preliminary count of the votes of the election on 17 September 1979 gave the Social Democrats and Communists a majority in the unicameral Riksdag. But with the counting of postal votes (especially those from abroad) it soon became apparent that the three non‐Socialist parties in the Riksdag—the so‐called ‘bourgeois’ parties— would maintain their hairsbreadth parliamentary majority. The final count tipped the balance, and the bourgeois parties won a majority of a single seat (compared with one of eleven in the previous parliament). The electoral system is designed to reflect an exact proportionality of representation for parties whose support exceeds four per cent of the total vote. The turnout of around 91 per cent was slightly lower than in 1976, but remained at the high level characteristic of the 1970s.