Welsh Journals

Search over 450 titles and 1.2 million pages

Some Geographlcal Obøervatloaa on the 1983 General Election In Wales and ito Implicat_ons for the County's Fntnre Political Pattern. HAROLD CARTER University College of Wales, Aberystwyth (Received August 1983) It is. perhaps, only barely tenable to justify a paper on the basis of the availability of data. Even so the publication by the OPCS of the Parliamentary Constituency Monitor, 1983 Boundaries, Wales (CEN 81 PCM 25) provides a unique circumstance where an array of 1981 socio-economic data on a parliamentary constituency basis is avail- able very near to the time of a general election which provides evi- dence of voting patterns (The Times, 11.6.83). Although, therefore, the motivation is opportunist, an analysis of the data can provide an effective contemporary assessment of the correlates of party strength and permit comment on both the socio-economic and political geo- graphy of Wales. The scale is, of course, ecological with all its well known disadvantages. But at least the interpretive comment which can be derived, if, as will probably be apparent, never value-free or loosed from the constraints of prejudice, has at least an extra array of factual data with which to contend. The procedure which this paper adopts is to begin by setting out the analysis in the form first of a tabulation of those variables (a full list is given in the appendix) which show statistically significant relationships with voting for each of the four main political parties (Table 1), and second of a multivariate analysis where all the vari- ables, including voting, are subject to factor analysis with a varimax rotation (Table 2). In addition, constituency scores on the first three abstracted factors have been mapped (Figures 1, 2 and 4) as well as the distribution (percentage) of Conservative votes (Figure 3). In all these tabulations and maps no attention is paid directly to which party won any seat since the 'first past the post' system while critical to Parliamentary representation has little relevance to the present analysis, except in so far as it may have induced tactical voting and the distortion of allegiances.