Institute of Governance > Online Articles & Papers, by Author > Online Articles & Papers, by Date Published Online > Article: Sources of support for the SNP by Lindsay Paterson

Sources of support for the SNP

by Lindsay Paterson

School of Education, University of Edinburgh, Holyrood Road, Edinburgh. EH8 8AQ

Tel: 0131 651 6380 Fax: 0131 651 6678 Email: lindsay.paterson@ed.ac.uk

draft: September 2004

to appear in Has Devolution Delivered? The New Scotland Four Years On,
edited by Catherine Bromley, John Curtice, David McCrone and Alison Park, 2005.

• Home
• News & Events
• About the Institute
• Consultancy & Research
• Publications
• Parliament News
• Online Articles & Papers
• by author
• by date online
• Internship Programme
• Site Map
• Search
• Contact

• Scottish Affairs web site

Introduction

The Scottish National Party is in an odd position. On the one hand, it has achieved more in the last eight years than in the previous sixty: a parliament is now well-established in Edinburgh, the party has been confirmed in two elections as the main challenger to Labour, and independence is now a serious option for Scotland, with something approaching a consensus for strengthening the parliament's powers and with growing although still minority support for separate statehood (McCrone and Paterson, 2002). And yet the party seems becalmed. It, like Labour, has fallen victim to Scotland's new multi-party politics, such that its performance at the elections of 2003 and 1999 - and at the UK general election of 2001 - was judged to be rather disappointing by most disinterested observers. The party has seemed to be reluctant to claim the setting up of a home rule parliament as any kind of victory, so totemic is its preference for independence. Unlike its counterparts in Catalonia or Quebec, it has yet to learn to live with mere devolution.

The fortunes of individual parties may seem to be of rather arcane interest, but that of the SNP cannot but command attention for at least two reasons. One is simply that it is indeed Scotland's opposition party, still the only one capable of beating Labour in its heartlands. If there is ever to be any alternation of power in Scotland, then the fate of the SNP will be central to that - either as the core of a new government, or as the reason why no other party than these two seems likely to be in a position to provide such a focus. That reason to be interested in the party would be common to any multi-party system, but there is a second reason peculiar to Scotland. The SNP has, historically, been the main motor of the debate about Scotland's constitution, however much its opponents might like to pretend otherwise. Its enthusiastic participation in the referendum of 1997 was the main reason why such a clear majority was achieved: its supporters voted by no less than 76 to 1 in favour, ahead of Labour's 66 to 7 and the Liberal Democrats' 45 to 32 (Surridge and McCrone, 1999, p. 43). It was the SNP's revival in the mid-1980s that began to focus Labour minds again on a Scottish assembly, although Donald Dewar himself did not need such prodding (Dewar, 1988). It was SNP success in 1973-4 that persuaded the Wilson government to legislate on a Scottish assembly, and it was the party's winning of the Hamilton by-election of 1967 that persuaded Wilson to set up the Royal Commission on the Constitution whose report remained influential on the settlement we now have (Paterson, 1998, p. 226). Throughout the period from its founding in 1934 until 1996, the party's ephemeral electoral successes also nudged Conservative and Labour governments into strengthening the administrative powers of the old Scottish Office (Finlay, 1994, 1997; Kellas, 1984; Lynch, 2002). The SNP's capacity to attract votes from elsewhere has always concentrated the minds of other parties' leaders.

Nevertheless, despite that, the Scotland of the early twenty-first century is a very different place from the country which the SNP seemed to be about to take by storm thirty years ago. Some of these changes might have been expected to have favoured the party, given what is known about its social base in the 1970s (Kendrick, 1983; McCrone, 1992, pp. 146-73). Scotland is now more affluent and better educated, and has an economy now based on services rather than manufacturing; all these changes entailed a relative growth in those social groups which provided the core of the party's support in 1974: it benefited from the same kinds of social change as underpinned the success of Thatcher's Conservatives in England. But the ideological difference between these parties also points in a paradoxical direction for the SNP. It moved left while the country was becoming more middle class. That in itself would not necessarily have been an ideological obstacle: the growing Scottish middle class remained quite firmly social democratic, rather more akin to its counterparts in Scandinavia than to the middle class in southern England (Paterson, 2002a,b). But the ideology was a cautious communitarianism, which Labour has been adept at leading, and which the leftist rhetoric of some leading strands of opinion in the SNP has rather missed.

During the recent leadership contest in the SNP, there was much public discussion of the party's immediate prospects and current sources of support. This chapter therefore stands back from the fray, and examines changes in the party's base over a long time-scale, in some respects since the 1970s and, in greater detail, since the pivotal year of 1997. There are four main sections. The first examines the social base of SNP support from 1974 until 2003: why has social change apparently not strengthened the party's base? The second looks briefly at the relationship between support for the party and people's views about Scotland's constitution: why has the party not benefited from the strengthening popular belief that Scotland needed a parliament, and indeed a parliament with more substantial powers than it currently has? The third section concentrates on the party's capacity to build up a core of loyal supporters, of the kind which underpins Labour's dominance of Scottish political preferences. These three sections are about broad support for the party - analysis of an amalgamation of survey responses to two questions, one on whether the respondent is a supporter of a party, and the other (for those who decline to call themselves a party supporter) on whether they feel closer to one party than to the others; that measure is often called 'party identification' by political scientists, but it is referred to here simply as 'party support'. The fourth section then looks at the SNP's effectiveness at translating vague support into votes. The data used in the analysis come from a series of surveys dating back to 1974; these are summarised briefly in the Appendix.

Social base of SNP support

Five major developments in the social base of SNP support may be detected. The first concerns age, shown in the top segment of Table 1. Older people (aged 65 and older) always have tended to give lower support to the party than younger groups, but the change is in the comparison of those aged under 35 with those older than that. In the 1970s, the youngest clearly gave strongest support, but that has not been the case since the early 1990s. Indeed, there is some evidence here that, as the young people of the 1970s have aged, they have taken quite high levels of support for the SNP with them, while new generations of young people have been somewhat more sceptical of the party. Thus the level of support in the age groups 45 and older was higher in the recent surveys than in 1974, whereas in the younger age groups it was clearly lower. Another way of expressing this is that the average age of SNP supporters in 1974 was 38.5 years, more than six years younger than the population average of 44.8 years; by 2003, the SNP average had grown to 46.9 years, whereas the population average had grown only to 47.7 years. The party's main rivals, the Labour party, has to some extent shifted in the opposite direction. In 1974, it had under 30% support among people aged under 35, but 35-40% among people in the groups aged over that. By 2003, its support was around one third in all age groups.

The second notable change is in relation to gender. It has long been noted that the SNP seemed more attractive to men than to women, and the surveys from 1974 to 1999 confirm that (Table 1): in the first and last of these, the level of support among men was about one half greater than among women. The party leadership around John Swinney after 2000 - prompted by the quite high levels of female representation in the Scottish Parliament's SNP group - tried to appeal more directly to women, and the results may be seen in the surveys of 2001 and 2003, where the gap was closed and then in fact reversed. (The attribution to Swinney's leadership from autumn 2000 is made more plausible when we note that the proportions in the 2000 Scottish Social Attitudes Survey from spring 2000 were similar to those in 1999 in Table 1, at 22% among men and 17% among women, whereas those for 2002 were similar to the figures for 2001 in Table 1, at 17% among men and 16% among women.)

The third change concerns social class and social mobility. The bottom segment of Table 1 shows that the party has never had much appeal to people in professional jobs. In the 1970s, however, it did have quite high support in intermediate groups, an advantage it had probably lost thirty years later, by when its support was concentrated in working-class groups (the bottom three in the table). This shift from a somewhat mixed class base to a more clearly working-class one might seem consistent with the party's shift to the left in its rivalry with Labour, but the problem for the SNP is that the working class is shrinking and the middle class growing (Paterson et al, 2004). Nevertheless, the story is actually more complex because it is partly about how people arrived at their current class position. The party of the 1970s appealed to upwardly mobile people - people whose own jobs were further up the social hierarchy than their parents' had been. An illustration of the situation in Scotland is in Table 2 (further detailed analysis is in Kendrick, 1983, summarised by McCrone, 1992, p. 166). The table shows the proportions supporting each of four parties according to the broad occupational class the respondent was in and to whether they had been socially mobile to get there; note that the social mobility was defined in terms of the full set of six classes shown in Table 1, not only in terms of the dichotomy shown in Table 2, so that, for example, being upwardly mobile in the upper segment could mean moving up within the non-manual class. The SNP in 1974 seemed to rely on two main groups: those who had been upwardly mobile to the non-manual class (22% support), and also people who were stable in the manual class or had been downwardly mobile to the manual class (28% and 20%); that combination represented the class coalition which the party had assembled then. By 1999, however - as illustrated in Table 3 - only one of these two concentrations of support remained: the party no longer had any advantage among the upwardly mobile, and even seems to have come to appeal particularly to the downwardly mobile. The contrast with Labour is rather stark: from a profile not dissimilar to the SNP's in 1974 (although with much higher levels of support in the manual class), it had become quite definitely associated with upward mobility by 1999.

The fourth social change is about religion. The SNP used to have a problem with Catholics: in 1974, only 12% of them supported the party, in contrast to 27% of people with no religion and to 19% of adherents of the Church of Scotland and of other presbyterian churches. In 2003 the differential was weaker, essentially because support was lower among the non-religious (19%) and among the presbyterians (17%) but had barely changed among Catholics (13%). Much fuller analysis of this slow secularisation of Scottish politics is provided by Rosie and McCrone (2000, pp. 213-5), McCrone and Rosie (1998, pp. 86-90) and Seawright and Curtice (1995).

The fifth and final point about social change concerns national identity. Most people in Scotland (over nine out of ten) choose 'Scottish' or 'British' as the label which best defines their national identity. Here the story is in one respect of no change. The SNP has always had much stronger support among the 'Scottish' than among the 'British': 22% to 5% in 2003, 15% to 2% in 1979. But there is underlying change here, because the size of the 'Scottish' group has been growing and that of the 'British' group has been declining (Paterson, 2002c, p.32): the 'Scottish' proportion was 56% in 1979 and 72% in 2003, while the 'British' proportion fell from 38% to 20%. So the SNP is commanding a constant share of a growing group. Further analysis of this question of identity and the SNP is provided by Bond (2000) and Bond and Rosie (2002, pp. 42-5).

In summary of this evidence on the social base of SNP support the most striking contrast may be summed up through an over-simplified contrast. The SNP used to be the party of the aspirational - the young, upwardly mobile, male children of the welfare state, groups which gave Margaret Thatcher her victories in England. The SNP is now the party of the downwardly mobile and is no longer the party particularly of the young.

That rather crude contrast might make depressing reading for the party, and there are, however, other ways of reading the data which point to a rather more optimistic potential. The SNP seems to have broadened its appeal in some ways even while losing the support of the upwardly mobile. It has apparently stopped alienating Catholics. It has swum with the tide of rising Scottishness. And, for the first time, it has appealed more to women than to men (although its strategy was presumably not to do so at the expense of its male support, which seems to have been what has happened). There is also a further way in which its loss of upwardly mobile support need not be disastrous: there is in fact less upward mobility among younger age groups than among people who grew up in the 1950s and 1960s, the main reason being that young people today are the children of these earlier generations, and opportunities for middle-class careers have not been expanding rapidly enough to absorb them at the rate that the parents could achieve. Thus some of the sons and daughters of people who were upwardly mobile in the 1950s and 1960s have had to settle for jobs at the lower end of the middle-class spectrum. (Further details are in Paterson et al, 2004). As this tide advances as well, perhaps the SNP will find itself carried along with it just as it was by the rising upward mobility of the 1960s.

SNP support and constitutional change

It has long been noted that support for the SNP and support for independence are quite distinct (see, for example, Bond, 2000). Table 4 shows the data that have formed the basis of that conclusion, and takes the picture down to 2003. The party has always been much stronger among people who favour independence than among others, but even there the party's support has dropped to barely two in four as the proportion of people supporting independence has grown (McCrone and Paterson, 2002, p. 57). The same has broadly been true since 1999 of the more general question of strengthening the Scottish parliament's powers, as Table 5 shows: in 2003, among all those who 'agree' or 'agree strongly' that the parliament should be strengthened (59% of the sample), only 26% support the SNP, in contrast to 35% supporting Labour.

The SNP also has stronger support among those who are most enthusiastic in their evaluation of the parliament's performance, and so the perception that the parliament has been lacklustre may have limited the SNP's support: see the upper part of Table 6. For example, among people who, in 2003, agreed that the parliament was increasing the say which ordinary people have in government, 20% supported the SNP; the figure was 16% among those who believed the parliament was making no difference. The levels of SNP support on a question about the parliament's impact on the quality of the health service were 22% among those who believed that it was improving, but only 14% among those who believed it was not changing, and on the performance of the economy the proportions were 24% and 13%. On a question about education there was little difference in analogous levels of SNP support (19% compared to 18%). Labour's share, by contrast, was only a few percentage points higher among those with a positive view of the parliament's performance than among those who believed it had had little impact (details not shown in the table).

In some ways this is surprising. One interpretation of the data on party support tabulated against constitutional option (Table 4) is that it shows SNP popularity to be higher among people who rejected the old Scottish Office than among those who favoured that way of governing Scotland, and yet those who are unhappy with the new status quo are not, on the face of it, so inclined to support the party. The key to resolving the paradox is in how people react to their disappointment at the parliament's performance, as the lower part of Table 6 shows. Among those who want the parliament's powers to be increased, there is little or no difference in the levels of SNP support between those who have a positive evaluation of the parliament's performance and those who don't: on 'say in government', the proportions are 27% and 25%, on the health service they are 26% and 23%, and on education they are 24% and 27%. Only on the economy did the difference remain, but the gap had narrowed from 11 points to 8 (29% against 21%). Moreover, on this and on the question about the health service, the highest level of SNP was found among the small minority who believed that the parliament had had a harmful effect. It would appear from this that the parliament's performance is not directly the explanation of levels of SNP support: what does explain these are the ways in which people are responding to that performance by thinking about alternative constitutional arrangements.

In that sense, the SNP's approach to the sense of disappointment with the parliament has been astute, arguing as it does that the way of improving matters is to increase the parliament's powers: the percentages in the lower panel of Table 6 suggest that only by expanding the group of people who agree with that (already a majority) can the party benefit from the disappointment. In reacting in this way it has also implicitly modernised its very long-standing role as the motor of the constitutional debate. However, unless it can extend its support by this means sufficiently to form a government, it may well - as in the past - see the mantle of further constitutional reform being stolen from it by others, notably Labour.

Stability of SNP support

In some respects, cross-sectional surveys taken at one point in time under-estimate the amount of change in allegiance and attitudes. This has been shown for attitudes to the Scottish constitution, where, in the five years following the 1997 election, around one half of people favoured independence at some point, even though only about one quarter favoured it at any point in time (McCrone and Paterson, 2002, p. 69). Similar findings have been reached concerning the British electorate (Clarke et al, 2004, pp. 175-216). We can similarly use the panel studies of 1992-6 and 1997-2001 to examine the stability of party support in Scotland: as noted in the Appendix, these surveyed the respondents to the 1992 and 1997 election surveys every year or two, and thus allow us to track change in individuals' views over time.

Table 7 summarises the results from the panel which ran every spring from 1997 to 2001. There were thus five waves, and so a wholly consistent person would choose the same party every time. It may be seen that the SNP had only 10% of the electorate as consistent supporters in that sense, similar to the Conservatives and much less than Labour's 35%. Moreover, 60% of people chose Labour on at least one of the five occasions; thus even excluding the 35% who always supported Labour, there were still 25% who were not averse to supporting it, not much less than the SNP's 29%. That stability of Labour's core, and appeal beyond that core, is a measure of the SNP's problem. It (and the other parties) are in effect competing for the two thirds of people who sometimes don't support Labour, but Labour itself is almost as often able to attract these as is the SNP.

Table 8 then compares the two panels. Because the 1992 panel asked the relevant questions only in 1994 and 1996, the 1997 panel has here been restricted to the analogous years of 1997, 1999 and 2001. Since its British victory in 1997, Labour has probably consolidated its core vote, the proportion supporting that party on all three occasions rising from 28% to 38%. The SNP, by contrast, has at best made no progress in this respect: it is no closer to having a core base than it was before a Scottish parliament was in prospect.

Where do these disloyal SNP supporters go? Now returning to the five-wave version of the 1997 panel, we find that 47% of people who supported the SNP at least once moved to Labour at least once, by contrast to 13% to the Liberal Democrats and 6% to the Conservatives. The problem for the SNP is, again, that the favour is not reciprocated: among people who supported Labour at least once, only 23% ever supported the SNP. Indeed, among people who supported Labour at least once, 59% always supported Labour, much more than the 36% sticking with the SNP among people who supported that party at least once. Because Labour support in any given year is more than twice the SNP's, similar absolute numbers of people flow from Labour to the SNP as in the other direction, but that merely then has the effect of consolidating Labour's dominance.

This impression of a rather small core support for the SNP may be confirmed by more subtle statistical analysis of the panel studies. The details are not given here, but the main point is that in the 1997 panel around one fifth of the electorate was very willing to experiment with supporting the party, but one quarter of them deserted between waves. The other four fifths of the electorate rarely supported the party, and around 90% of those who did so deserted between waves. In this latter group, the desertion rate rose to over 95% among people who never supported independence.

SNP votes

Thus far we have been looking only at general support for the SNP. This approach has the advantage of allowing party support to be tracked between elections, and also probably revealing underlying attitudes as distinct from the possibly tactical motives that influence people at elections. However, ultimately parties have to gather votes, and so this section looks at the relationship between votes and support over the same period of time we have been studying so far. We deal first with votes at UK general elections in Scotland, and then turn to the rather different and more complex issues which arise in the proportional electoral system for the Scottish parliament.

Table 9 shows, first, that, except in 1974, the SNP has not been good at attracting votes from people who identify with other parties. Second, it used to attract some votes from people who identify with no party at all, but that capacity had almost vanished by 2001. Above all, however, the SNP used to be able to mobilise its own supporters to vote for it (85% doing so in 1974), but spectacularly lost that capacity by 2001 (the proportion being down to 57%).

It is true that all parties are becoming less effective at mobilising their supporters, as Table 10 shows, but the drop for the SNP has been greater. In particular, comparing the party again with its main rival, Labour, whereas they mobilised similar proportions of their supporters from 1974 to perhaps as late as 1992, the SNP fell behind thereafter. This deficiency is partly explained by different tendencies of each parties' supporters to abstain altogether, as Table 11 shows: SNP abstention rose from 10% in 1974 to 30% in 2001, a somewhat larger rise than Labour's from 10% to 25%. For the SNP it is also explained by a greater amount of desertion of supporters to other parties: in 1974, only 4% of SNP supporters voted for other parties, whereas in 2001 the proportion was 12%; the analogous proportions for Labour were 7% in 1974 and 6% in 2001. The SNP's vote has also suffered from the fact that people who identify with no party were, by 2001, nearly all abstaining.

A similar story emerges from the shorter history of voting at Scottish parliament elections. Table 12 shows the SNP attracting few of the supporters of other parties. In 1999, it was moderately successful at gaining the votes of some people who did not identify with any party, almost indeed back at the 1974 levels in this respect (Table 9). But that collapsed in the election of 2003. The decline between 1999 and 2003 in the SNP's capacity to mobilise its own supporters was greater than for the three other large parties, as is seen in Table 13: that was because it was, in 1999, much better at mobilising its own supporters, but the advantage was mostly lost in 2003. Indeed, we can calculate that, if the SNP had attracted in 2003 the same share (70%) of its supporters as in 1999, its overall vote share would probably have risen by about 4-5 percentage points, cancelling out most of its loss in total share of the vote between these two elections.

A similar picture emerges if we compare the votes on the two ballots in the Scottish parliament elections (as opposed to comparing votes in each ballot with general party support). For example, whereas in 1999 the SNP retained on the list ballot 83% of those who voted for it in the constituency ballot, by 2003 this had fallen to 71%. The change for Labour was less severe, from 78% to 73%. One notable feature here, as of Table 13, was the success of the Conservatives at mobilising their supporters. In 2003, we see from Table 13 that they attracted 60% of their supporters in the list ballot, more even than the SNP, and these list voters included 85% of those who had supported the party on the constituency ballot, 12 points more than the next highest, Labour. This feature of the Tory vote might be interpreted with cautious optimism by the SNP, because some media commentators have suggested that the Tories might be about to overtake the SNP as the second party in Scotland. These figures suggest that that is unlikely: the Conservatives did moderately well in 2003 mainly because they were excellent at mobilising and retaining their quite small body of support. The SNP failed in these respects, but it continues to operate from a somewhat larger if more diffuse base, as Table 7, for example, showed. If the Tories had been able to mobilise on the list vote a share of its supporters only at the level achieved by Labour or the Liberal Democrats (45-50% instead of 60%, from Table 13), its share of the total vote would have fallen by about 4 percentage points, leaving it with a total of around 14 instead of 18 seats, and hence in fourth place behind the Liberal Democrats.

Conclusions

Analysing the social basis of a party's support, and the flows of votes between it and other parties, is only one aspect of understanding why it is or is not successful, and the data offered here therefore provide only one way of understanding the SNP's dilemmas. But however much party strategists might want to supplement such an analysis, the lessons from it do seem to offer both problems and opportunities for the party.

On the negative side first, we have concluded that the SNP is no longer the party of the young and the upwardly mobile. The party's shift to the left may have been understandable as an attempt to take votes from Labour in the 1980s and 1990s, but it seems now to have lodged the party in a shrinking working class, and in an ageing segment of the population. If the SNP was once the vehicle by which the aspirational sought to modernise Scottish society, it may now be the resort of the disgruntled, the downwardly mobile. It also has a declining capacity to mobilise its own supporters, and to attract votes from other parties and from those who support no party. The support it does attract is more transient than that of its main rival, Labour. It has not benefited much from the growth in support for home rule since the 1970s, nor even very much from the recent growth in support for strengthening the powers of the Scottish parliament or for independence. It is clear from this, further, that Scotland could easily over the next decade or so reach a position of majority support for something like independence with the SNP still in a minority.

But these are not the only conclusions warranted by the data, and there are countervailing tendencies that promise rather better prospects for the party. It has broadened its social base, appealing more successfully to women and to people of a variety of religious persuasions that at any time in the past thirty years. If it has come to attract the support of the downwardly mobile more than those moving in the opposite direction, that is in fact in keeping with an important social trend. There are more downwardly mobile people among young people today because the opportunities to move up have not been expanding rapidly enough (a trend common across Britain, and probably other developed societies). If the SNP is a party of the disgruntled, then that might in due course resonate with a disgruntled society.

The SNP is not the only party that is coming to be less effective at mobilising its supporters, and it seems likely that, in due course, Labour will suffer from similar problems. If the only way of avoiding that is to concentrate on a small but beleaguered core support, as the Tories seem to have done, then perhaps both these parties would prefer the alternative of more pluralistic competition. Differential turnout of supporters is, after all, as important in elections as absolute levels of support.

The main reason why party activists ought to be cautiously optimistic, however, is a rather perverse conclusion from the statistics. The SNP's base of support since the Scottish parliament was set up is clearly higher than it has been at any time since the mid-1970s. Indeed, the level and social profile of support in 1999 resembles that in 1974 fairly closely, as may be seen from Table 1, the main exceptions probably stemming from the points about downward mobility that have ambiguous implications. The results of neither the 2001 nor the 2003 elections were anywhere nearly as poor as those of 1979. When we consider that 1999 was regarded as somewhat disappointing for the party, following its very high ratings in opinion polls after the 1997 referendum, and that 1974 was its best ever performance, we have a measure of how far it has come. The very fact that it can be doing so well, in historical terms, despite all the problems sketched in this chapter should be a warning to those who would dismiss its prospects out of hand. The problem remains, however, that after 1974 the party had to wait until the middle or late 1980s before it started to recover again. Dating its most recent high point as 1997-8, and its current problems from 1999, that would place the serious revival of its electoral fortunes at around the time of the Scottish parliament elections of 2011.

Appendix: surveys

The chapter uses the Scottish Election Surveys of 1974, 1979, 1992 and 1997, the Scottish Social Attitudes Surveys of 1999, 2001 and 2003, the British Election Panel Studies of 1992-6 and 1997-2001, and the British Household Panel Study of 1999-2001. The cross-sectional surveys were multi-stage cluster samples, stratified at the cluster level, and drawn from the electoral register until 1992 and from the postcode address file from 1997. Data were collected by face-to-face interviews in respondents' homes, computer-aided from 1997. Full details of the sampling design etc are reported in the appendices of Paterson et al (2001) and Curtice et al (2002), and in the appendix of the present book. The response rates were at least 60% and usually between 65% and 70%. The Scottish Social Attitudes Survey achieved samples of 1482 in 1999, 1605 in 2001 and 1508 in 2003. The Scottish Election Survey achieved samples of 1175 in 1974, 729 in 1979, 957 in 1992 and 882 in 1997. Among the 957 respondents to the election survey in 1992, the subsequent panel study achieved response rates of 61% in 1994 (588 respondents) and 44% in 1996 (418 respondents). Likewise, among the 882 respondents to the 1997 election survey, the panel study achieved response rates of 76% in 1998 (672 respondents), 71% in 1999 (626 respondents), 66% in 2000 (586 respondents), and 68% in 2001 (596 respondents). Although the panel studies thus had quite high levels of attrition, they do have the unique virtue of letting us study the ways in which individuals do or do not change their minds in their political attitudes. Moreover, the attrition did not vary strongly by relevant attitudes (as measured in 1997 or 1992): it was much the same among supporters of various constitutional options for Scotland, of various political parties, and in various demographic groupings defined by social class, gender and age. The only statistically significant variation in response rates were that, in both the 1992 and the 1997 panels, older people (aged over 54) were less likely to respond than younger, and that, in the 1992 panel only, people who had supported the Conservatives and Liberal Democrats in 1992 were more likely to respond than others. The main effect of this variable attrition would be to exaggerate somewhat the Conservative and Liberal Democrat support in the 1992-96 columns of Table 8. This would not affect the main comparison we have looked at, that between the SNP and Labour. These surveys have been funded mainly by the UK Economic and Social Research Council and its predecessors, and have been run mostly by the National Centre for Social Research (formerly SCPR).

The chapter also uses the British Household Panel Survey, an annual panel survey of adults which, for Britain as a whole, has been running since 1991; it is funded by the ESRC and is carried out by the Research Centre on Micro-Social Change at Essex University. Since 1999 the Scottish sample has been enhanced to about 1,500 households and about 3,000 adults, including the surviving people in Scotland from the original sample. The new sample included full representation of the area north and west of the Great Glen, although the original (1991) sample did not. Further information is available from the survey's web site, at www.iser.essex.ac.uk/bhps/index.php.

References

Bond, R. (2000), 'Squaring the circles: demonstrating and explaining the political "non-alignment" of Scottish national identity', Scottish Affairs, no. 32, pp. 15-35.

Bond, R. and Rosie, M. (2002), 'National identities in post-devolution Scotland', Scottish Affairs, no. 40, pp. 34-53.

Clarke, H. D., Sanders, D., Stewart, M. C. and Whiteley, P. F. (2004), Political Choice in Britain, Oxford: Oxford University Press.

Curtice, J., McCrone, D., Park, A. and Paterson, L. (eds) (2002), New Scotland, New Society?, Edinburgh: Edinburgh University Press.

Dewar, D. (1988), Williamson Lecture, Stirling University, 21 October, extracts reprinted in Paterson (ed.) (1998), pp. 169-73.

Finlay, R. J. (1994), Independent and Free, Edinburgh: John Donald.

Finlay, R. J. (1997), A Partnership for Good?, Edinburgh: John Donald.

Kellas, J. (1984), The Scottish Political System, Cambridge: Cambridge University Press, third edition.

Kendrick, S. (1983), Social Change and Nationalism in Modern Scotland, PhD thesis, Edinburgh University.

Langeheine, R. and van de Pol, F. (1994), 'Discrete-time Markov latent class models', in Dale, A. and Davies, R. (eds), Analysing Social and Political Change, London: Sage, pp. 170-97.

Lynch, P. (2002), The History of the Scottish National Party, Cardiff: Welsh Academic Press.

McCrone, D. (1992), Understanding Scotland, London: Routledge, first edition.

McCrone, D. and Paterson, L. (2002), 'The conundrum of Scottish independence', Scottish Affairs, no. 40, pp. 54-75.

McCrone, D. and Rosie, M. (1998), 'Left and liberal: Catholics in modern Scotland', in Boyle, R. and Lynch,. P. (eds), Out of the Ghetto?, Edinburgh: John Donald, pp. 67-94.

Paterson, L. (ed.) (1998), A Diverse Assembly: the Debate on a Scottish Parliament, Edinburgh University Press.

Paterson, L. (2002a), 'Governing from the centre: ideology and public policy', in Curtice, J., McCrone, D., Park, A. and Paterson, L. (eds), New Scotland, New Society?, Edinburgh: Edinburgh University Press, pp. 196-217.

Paterson, L. (2002b), 'Scottish social democracy and Blairism: difference, diversity and community', in Hassan, G. and Warhurst, C. (eds), Tomorrow's Scotland, London: Lawrence and Wishart, pp. 116-29.

Paterson, L. (2002c), 'Is Britain disintegrating? Changing views of "Britain" after devolution', Regional and Federal Studies, 12, pp. 21-42.

Paterson, L., Brown, A., Curtice, J., Hinds, K., McCrone, D., Park, A., Sproston, K. and Surridge, P. (2001), New Scotland, New Politics?, Edinburgh: Edinburgh University Press.

Paterson, L., Iannelli, C., Bechhofer, F. and McCrone, D. (2004), 'Social class and social opportunity', in Paterson, Bechhofer and McCrone, Living in Scotland: Social and Economic Change since 1980, Edinburgh: Edinburgh University Press, pp. 80-104.

Rosie. M. and McCrone, D. (2000), 'The past is history: Catholics in modern Scotland', in Devine, T. M. (ed.), Scotland's Shame?, Edinburgh: Mainstream, pp. 199-218.

Seawright, D. and Curtice, J. (1995), 'The decline of the Scottish Conservative and Unionist Party 1950-92: religion, ideology or economics?', Contemporary Record, 9, pp. 319-42.

Surridge, P. and McCrone, D. (1999), 'The 1997 Scottish referendum vote', in Taylor, B. and Thomson, K. (eds), Scotland and Wales: Nations Again?, Cardiff: University of Wales Press, pp. 41-64.

Vermunt, J. K., (1997), LEM: A General Program for the Analysis of Categorical Data, available from the web site of the University of Tilburg at http://www.tilburguniversity.nl/.

Table 1: Support for SNP (%) by class, gender and age, 1974-2003

*Class measured by Registrar General scheme, 1992-2003, and by social grade in 1974 and 1979.

Omits people with no information on class or on age. Don't know and not answered on party identification included in the base (N).

Percentages weighted in 1997-2003; all sample sizes unweighted.

Sources: Scottish Election Surveys of October 1974, 1979, 1992 and 1997; Scottish Social Attitudes Surveys of 1999, 2001 and 2003.

return to reference in text

 Table 2: Party support (%) by social mobility and broad current class, 1974

Current class and mobility are defined in terms of social grade of father: see text.

Omits people with no information on own class or father's class. Don't know and not answered on party identification included in base (N).

Source: Scottish Election Survey of 1974

Table 3: Party support (%) by social mobility and broad current class, 1999

Current class and mobility are defined in terms of Goldthorpe classes, combining both parents: see text.

Percentages weighted; sample sizes unweighted.

Omits people with no information on own class or parents' class. Don't know and not answered on party identification included in base (N).

Source: respondents in Scotland to British Household Panel Survey of 1999.

 Table 4: Support for SNP (%) by preferred constitutional option, 1979-2003

* The two kinds of independence were not distinguished in the 1979 survey.

** Strong domestic parliament was referred to in 1979 as 'Scottish Assembly which would handle most Scottish affairs', and from 1997 onwards as 'Scottish Parliament within the UK with some taxation powers'. Weak domestic parliament was referred to in 1979 as 'Scottish Assembly which would handle some Scottish affairs and would be responsible to Parliament at Westminster', and from 1997 onwards as 'Scottish Parliament within the UK with no taxation powers'. No distinction was made in 1992.

Omits people with no view on the constitution; don't know and not answered on party identification included in the base (N).

Percentages weighted in 1997-2003; all sample sizes unweighted.

Sources: Scottish Election Surveys of 1979, 1992 and 1997; Scottish Social Attitudes Surveys of 1999, 2001 and 2003.

Table 5: Support for SNP (%) by whether want Scottish parliament to have more powers, 1999-2003

Omits people with no view on the powers of the parliament. Don't know and not answered on party identification included in the base (N).

Percentages weighted; sample sizes unweighted.

Sources: Scottish Social Attitudes Surveys of 1999, 2001 and 2003.

Table 6: Support for the SNP (%), by evaluations of the Scottish Parliament, 2003

Omits people with no views on the questions about evaluation.

The columns are defined as follows:

• 'say in government': 'improved' is those who believed the parliament was giving people more say, and 'declined' is those who believed it was giving people less say.
• 'health service' and 'economy': 'improved' is those who believed the parliament would make the standard of the health service or the performance of the economy 'a lot better' or 'a little better', and 'declined' is those who believed it would make these 'a lot worse' or 'a little worse'.
• 'education': 'improved' is those who believed that the parliament was increasing the standard of education, and 'declined' is those who believed it was reducing the standard of education.

Percentages weighted; sample sizes unweighted.

Source: Scottish Social Attitudes Survey of 2003.

Table 7: Frequency of support for various parties (%), 1997-2001

Other parties, don't know and not answered included in the base.

Percentages weighted; sample size unweighted.

Source: respondents in British Election Panel Study who were resident in Scotland in 1997, waves in spring 1997, spring 1998, spring 1999, spring 2000 and spring 2001.

Table 8: Frequency of support for various parties (%), 1992-6 and 1997-2001*

* Three waves from each panel: 1992, 1994, 1996, 1997, 1999, 2001.

Other parties, don't know and not answered included in the base.

Percentages weighted for 1997-2001 panel; all sample size unweighted.

Source: respondents in British Election Panel Study who were resident in Scotland in 1992, waves in spring 1992, spring 1994 and spring 1996; and respondents in British Election Panel Study who were resident in Scotland in 1997, waves in spring 1997, spring 1999 and spring 2001.

 Table 9: Vote for SNP (%) at UK general elections, by party identification, 1974-2001

Omits people who identify with other parties. Did not vote, don't know and not answered on voting question included in the base (N).

Percentages weighted in 1997 and 2001; all sample sizes unweighted.

Sources: Scottish Election Surveys of October 1974, 1979, 1992 and 1997; Scottish Social Attitudes Survey of 2001.

Table 10: Proportion of party identifiers who voted for party at UK general elections, 1974-2001

Omits people who identify with other parties. Did not vote, don't know and not answered on voting question included in the base (N).

Percentages weighted in 1997 and 2001; all sample sizes unweighted.

Sources: Scottish Election Surveys of October 1974, 1979, 1992 and 1997; Scottish Social Attitudes Survey of 2001.

Table 11: Proportion of party identifiers who did not vote at UK general elections, 1974-2001