SCOTTISH ELECTIONS RESEARCH MAY-JUNE 2003 PREPARED FOR: THE ELECTORAL COMMISSION

Transcription

1 SCOTTISH ELECTIONS RESEARCH MAY-JUNE 2003 PREPARED FOR: THE ELECTORAL COMMISSION

2 REPORT PREPARED BY: Martin Boon Head of ICM government research John Curtice Professor of politics University of Strathclyde 2

3 1. Table of contents Page 2. SUMMARY ON A PAGE 4 3. METHODOLOGY 5 4. BACKGROUND TO THE SURVEY 8 5. THE SCOTTISH POLITICAL MAP Desired political landscape in Scotland The location of political power in Scotland The difference that the Scottish Parliamen 11 has made National identity INTEREST & UNDERSTANDING Level of interest Level of understanding Does it makes a difference who wins? ATTITUDES TO THE VOTE Agreement with statements VOTE BEHAVIOUR The decision to turnout Counter-factual voting decisions Reasons why non-voters did not vote Reasons why voters did vote WHY TURNOUT FELL INFORMATION & MEDIA THE POLITICAL PARTIES IN SCOTLAND Level of trust The party platforms Campaigning tactics NEW METHODS OF VOTING New voting mechanisms Postal voting in STATISTICAL MODELS APPENDIX 51 3

4 2. Summary on a page Turnout in the Scottish Parliamentary elections fell from 59% in 1999 to 49% in We offer two main reasons for this. Firstly, the Scottish Parliament and the outcomes of elections to it have declined in importance over the last four years in the public s mind. Secondly, any differences between the parties are perceived to have narrowed yet further, with party leaders unable to reach out and fill the policy void through personality politics. Our survey shows that over one in three (37%) members of electorate felt that they had insufficient information of the party leaders a substantial rise on a UK equivalent measure asked in We estimate that 10% of the Scottish electorate are Serial Non Voters. A further 21% are seriously disinterested but retain something of an allegiance to the political party of their choice, which is occasionally sufficient to drag them to a polling booth. For the remainder, circumstances more often than not explain their failure to turnout. Circumstantial reasons for not voting were indeed provided by one in five (22%) non-voters, and account for nearly half of those regular voters who didn t vote on this occasion. Voter apathy, disinterest and dislike of politicians were rationale provided by the serial non-voters. The Scottish Parliament is failing to motivate the electorate. In 1999, 41% believed that it had the most influence over the way Scotland is run, but now the proportion has fallen to 24%. This is due to the perception that the Parliament has changed little, with 57% of the electorate (and 67% of nonvoters) saying that it has made no difference to the way Scotland is governed. Those people who do vote are largely motivated by a civic duty. Three in four (74%) members of the public, and 94% of all voters on May 1 st felt they had a civic duty to vote. This is the glue that holds together a solid core of regular voters, but even this comes with a warning many of these same people are elderly, and this duty is not so deeply embedded among the young or the middle aged. Had various mechanisms that make the physical act of voting more accessible been available on May 1 st, many non-voters say they would have been more likely to turnout. Voting by phone or mobile phone and voting by post would have made a difference to almost half of all non-voters, with Internet voting and weekend voting more likely to have made a difference to one in three and one in four respectively. Statistical segmentation suggests that 15% of all Scottish electors fall into a Technology Friendly group, whose busy lives are likely to be made easier by the introduction of these kinds of new voting methods. 4

5 3. Methodology ICM Research interviewed a random sample of 1,100 people aged 18+ by telephone on 2 nd -5 th May A boost of 100 interviews among year olds is included within the sample size. All interviews were conducted across Scotland and the results have been weighted to the profile of all Scottish adults. The boost interviews were down-weighted into the overall profile. The geographical spread of interviews is in proportion to the distribution of the Scottish population (weighted). That profile is shown below: CENTRAL SCOTLAND INTERVIEWS GLASGOW INTERVIEWS HIGHLANDS & ISLANDS INTERVIEWS LOTHIANS INTERVIEWS FIFE & MID SCOTLAND INTERVIEWS NE SCOTLAND INTERVIEWS SOUTH OF SCOTLAND INTERVIEWS WEST OF SCOTLAND INTERVIEWS TOTAL (SCOTLAND)... 1,100 INTERVIEWS It should be remembered at all times that a sample and not the entire population has been interviewed. Consequently, all results are subject to sampling tolerances, which mean that not all differences are statistically significant. We can, however, predict the variation between the sample results and the true values (if everyone in the population had been interviewed) from knowledge of the size of the samples on which the results are based and the number of times answers are given. The confidence with which we can make this prediction is usually chosen to be 95% - that is, the chances are 95 times out of 100 that the true value will fall within a specified range. The table below illustrates the predicted ranges for different sample sizes and the percentage results at the 95% confidence level. 5

6 SAMPLE SIZE SAMPLING TOLERENCES APPLICABLE TO % S AT OR NEAR 10% OR 90% + / - 30% OR 70% + / - 50% + / interviews 5.88% 8.98% 9.8% 250 interviews 3.72% 5.68% 6.2% 500 interviews 2.63% 4.02% 4.38% % 2.84% 3.1% interviews 1,100 interviews 1.77% 2.71% 2.95% For example, with a sample size of 1,100 interviews where 50% (the worst case scenario as far as tolerances are concerned) give a particular answer, we can be 95% certain that the true value will fall within the range of 2.95% from the sample result. When results are compared between separate groups within a sample (say, between men and women), different results may be obtained. The difference may be real or it may occur by chance (because a sample rather than the entire population has been interviewed). To test if the difference is a real one, i.e. if it is statistically significant, we again have to know the size of the samples, the % giving a certain answer and the degree of confidence chosen. If we assume the 95% confidence level again, the differences between the results of two separate groups must be greater than the values given in the table below: SAMPLE SIZES TO BE COMPARED DIFFERENCES REQUIRED TO BE STATISTICALLY SIGNIFICANT AT OR NEAR 10% OR 90% + / - 30% OR 70% + / - 50% + / and % 12.7% 13.9% 200 and % 8.9% 9.8% 500 and % 5.7% 6.2% 1000 and % 4.0% 4.4% 6

7 SOCIAL CLASS DEFINITIONS Most market research projects classify the population into social grades, usually on the basis of the Market Research Society occupational groupings (MRS, 1991). They are defined as follows: Professionals such as doctors, solicitors or dentists, chartered people like architects; fully qualified people with a large degree of responsibility such as senior civil servants, senior business executives A. and high ranking grades within the armed forces. Retired people, previously grade A, and their widows. B. C1. C2. D. E. People with very senior jobs such as university lecturers, heads of local government departments, middle management in business organisations, bank managers, police inspectors, and upper grades in the armed forces. All others doing non-manual jobs, including nurses, technicians, pharmacists, salesmen, publicans, clerical workers, police sergeants and middle ranks of the armed forces. Skilled manual workers, foremen, manual workers with special qualifications such as lorry drivers, security officers and lower grades of the armed forces. Semi-skilled and unskilled manual workers, including labourers and those serving apprenticeships. Machine minders, farm labourers, lab assistants and postmen. Those on the lowest levels of subsistence including all those dependent upon the state long-term. Casual workers, and those without a regular income. 7

8 4. Background to the survey ICM Research were commissioned by the Electoral Commission to undertake a quantitative research exercise that sought to understand Scottish people s behaviour, attitudes and perceptions toward the 2003 Scottish Parliament and local council elections. Given that the main objective of the Electoral Commission is to increase public confidence in electoral processes and encourage people to take part in elections, recorded declining levels of turnout in elections across Scotland (and other areas of the UK) is clearly a cause for serious concern. Indeed the 2003 elections to the Scottish Parliament reinforced the notion that declining turnout is a clear trend rather than isolated or cyclical phenomena. The facts speak for themselves: Turnout in the 2003 election was 49.4% compared to 59% in the 1999 equivalent on 1 st vote. This figure also compares unfavourable to turnout in the 2001 General Election itself a low point of 58% in Scotland. The table below shows how turnout varied across Scotland between elections on 2 nd vote: REGIONAL 2 nd VOTE TURNOUT Region 2003 turnout 1999 turnout Change Central Scotland 48.61% 66.5% % Glasgow 41.45% 48.0% -6.55% Highlands & 52.22% 61.43% -9.21% Islands Lothian 50.52% 60.88% % Mid Scotland & Fife % -9.78% NE Scotland % -6.43% South of Scotland % -9.53% West of Scotland % -8.45% Total Scotland % -9.71% Source: SPICe briefing 6/5/03 Within this context, there are a number of key questions that the survey attempts to answer; some of which relate to levels of turnout in this election compared to recent others across Scotland, and others which relate to the conduct of the democratic process in Scotland. These questions include: 1. Why are people not voting in general? 2. Why is 2003 turnout lower than 1999 turnout for Scottish Parliament election? 3. How does the mindset of younger voters aged under 34 differ from older voters? 4. How does the level of interest in Scottish politics and parties vary over time? 8

9 There are two other issues that the reader should be aware of. Firstly, psephologists have long realised that there is often a discrepancy between actual turnout in elections and recalled turnout (and recall of past voting for candidates or parties for that matter) on subsequent opinion surveys. The reasons for this are themselves contested, but probably relate to a failure to interview a politically (rather than demographically) representative sample, or the reluctance of nonvoters to admit to being so. This survey is no different whilst the differential in actual turnout and recalled turnout is rather smaller than perhaps has been the case on other surveys for previous elections, recalled turnout remains higher 57% on the survey than the reality of 49%. We can see from the table below how this varies across the Scottish regions: Region 2003 actual turnout 2003 recalled turnout Central Scotland 48.61% 58% Glasgow 41.45% 57% Highlands & 52.22% 61% Islands Lothian 50.52% 54% Mid Scotland & Fife % NE Scotland % South of Scotland % West of Scotland % Total Scotland % Secondly, previous studies have indicated that as far as Scotland is concerned; age is by far the most significant discriminator in determining voter turnout*. ICM boosted the number of interviews with year olds in order to better understand this issue. As the table below shows, at first sight there does indeed appear to be some correlation between age and turnout: Age 1999 turnout 2001 turnout 2003 turnout % 43% 42% % 49% 41% % 67% 51% % 78% 59% % 79% 66% % 83% 75% However, statistical modelling of the data rather de-bunks this proposition. The survey does not provide consistent evidence that that fall in turnout is particularly concentrated amongst younger people. Quite simply, there is no statistically significant evidence to support the argument. 9

10 5. The Scottish Political Map 5.1 Desired political landscape in Scotland The establishment of a devolved Scottish Parliament was considered by many to be the settled will of the Scottish electorate, and to some extent it has proven to be so. When the Labour government made legislative proposals in 1997, the proportion that supported the concept rose from an average of 42% to a mark around 54%, and remains in that area today despite perceived failings on the part of the Parliament itself, which many people think has achieved little. Clearly, the performance of the Parliament has not done anything to counter the view that its inception was good idea in principle. The desire for an independent Scotland has consistently fallen in line with this, support for independence now stands at just under a quarter (24%). However, at the other end of the scale, there appears to be a rise in support for a return to predevolution Scotland, probably motivated by the failings of the Parliament, which are not considered or wanted to be solved via the independence route. Evidence for this argument is supported by the data, with 49% of Conservative voters in the Parliamentary elections in 2003 wanting a return to Westminster rule. In fact, party politics maybe the only factor involved in this, with voters and non-voters showing few discriminating characteristics on this particular measure. DESIRED POLITICAL LANDSCAPE FOR SCOTLAND ICM Research - Scottish Elections Research May Q42. For the running of Scotland as a whole, which of the following would you most like to see? Base: All respondents: 1, Now Devolved Scotland Independent Scotland No devolution 10

11 5.2. The location of political power in Scotland The political landscape of Scotland may have changed incontrovertibly with the introduction of the Scottish Parliament in 1999, but you would not necessarily know it given that only a quarter (24%) of the public think that the Parliament has the most influence over the way Scotland is run. It is true that slightly more voters (28%) than non-voters (19%) think that power is located more in the new Parliament than in the Westminster alternative, but it conclusively remains the case that twice as many members of the public (50%) still think that Westminster has the most say over how Scotland is run. There appears to be no divergence in the views of voters (50%) and non-voters (51%) on this, or indeed when it comes to the European Union (9% of voters vs. 10% of non-voters) or local Scottish councils (8% vs 9% respectively). The difference with the Scottish Parliament is largely made of up non-voters who do not know where most power is located. There are few conclusions to be drawn on this, with the possible exception that differential voting cannot be explained by the diffusion of power in Scotland, except in a small minority of cases The difference that the Scottish Parliament has made More than half (57%) of the electorate do not think that the Parliament has made any difference to the way Scotland is governed. The proportion rises to 67% among non-voters and falls to 50% among those who did turn out in the Parliamentary elections in May. When it comes to improving the way Scotland is run, three times as many voters (10%) think it has improved it a lot compared to non-voters (3%) while twice as many voters (27%) think it has improved it a little compared to non-voters (15%). When it comes to actually making the running of Scotland worse, there are no observable differences between voters and nonvoters, with one in ten (11%) saying that the Parliament has made running Scotland either a lot or a little worse. 5.4 National identity Nine in ten members of the Scottish public think themselves to be somewhere in between equally Scottish and British or Scottish not British. There has been a real convergence in the proportions of people who place themselves on one of the three points on this part of the national identity scale, with 32% saying they are More Scottish than British, and 30% saying they are Scottish not British AND Equally Scottish and British. There is some evidence to suggest that non-voters are more likely than voters to think of themselves as being Scottish and not British, but few other real differences emerge on this measure. Q43. Which, if any of the following best describes how you see yourself? Voters Non-voters Scottish, not British 25% 33% More Scottish than British 31% 27% Equally Scottish and British 31% 24% More British than Scottish 3% 3% British not Scottish 8% 10% 11

12 6. Interest & Understanding 6.1 Level of interest The Scottish electorate can be relatively easily sub-divided into three neat compartments when it comes to their interest in politics and political institutions. As the chart shows, just about one in three (33%) show a great deal or quite a lot of interest in politics, in the Scottish Parliament (28%) and their local council (31%). LEVEL OF INTEREST IN ICM Research - Scottish Elections Research May % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Politics The Scottish Parliament 26 Your local council Q1-3: How much interest do you generally have in what is going on in.? Base: All respondents: 1,100 None at all Not very much Some Quite a lot A great deal The second group show a lower level of interest and are probably best described as indifferent. In each case, just over one in four fall into this category 28% show some interest in politics per se, 27% in the Parliament and 26% in their council. The third group are rather or completely disinterested, representing about four in ten Scottish adults (39% disinterested in politics, 44% in the Parliament and 42% in their local council). As one might expect, voters (43% net interest saying a great deal or quite a lot ) are (statistically) significantly more interested than non-voters (19%) when it comes to politics in general, to the Parliament (40% vs. 14%) and their local council (42% vs. 18%). 12

13 The demographic differences that political observers would expect to see do emerge. Men show a greater level of political interest than women, the older the respondent, and the higher the social classification, the greater the level of interest. 6.2 Level of understanding Similar conclusions can be made about Scottish voters levels understanding of political institutions and systems (as they were for their interest in politics). Three groups are observable those who think they have depth to their understanding (about one in three), those who have some (about a quarter), and those who have hardly any at all (just under a half). LEVEL OF KNOWLEDGE ABOUT.. ICM Research - Scottish Elections Research May % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% House of Commons Scottish Parliament Local council Voting system for Scottish Parliament Q4: How much do you feel you know about what. does/how voting system for the SP election works? Base: All respondents: 1,100 None at all Not very much Some Quite a lot A great deal Scots do think their knowledge of their local council is very slightly higher than it is for the House of Commons and for both the Scottish Parliament and its proportional voting system. Overall, 32% say they know a great deal (8%) or quite a lot (24%) about their council, with a further 28% having some knowledge and 40% knowing not very much or nothing at all. The chart above shows how the findings distribute across the other political institutions and processes, but clearly those differences are very marginal across the board. Again, voters display significantly higher levels of understanding compared to nonvoters. Looking at the table below, the level of interest is seen to vary in direct proportion to number of times that respondents have voted in each of the four 13

14 elections covered within the context of this survey. Of course, this is no sensation it is intuitive and reasonable to assume that those people who show a higher level of interest in and understanding of political processes are more likely to participate in those processes. % who say they have a great deal/quite a lot of knowledge about.. VOTER COMMITMENT Number of times voted* House of Commons 12% 25% 29% 30% 38% Scottish Parliament 7% 13% 17% 20% 39% Local council 14% 15% 31% 28% 44% SP voting system 11% 20% 24% 33% 42% *Elections include: Scottish Parliament 2003, Local Council 2003, General Election 2001, Scottish Parliament 1999 However, while the proportion of Scottish people who do not know much about the Scottish Parliament is slightly lower than for Westminster and local councils, (as one would probably expect given it s only five years into its political life), it is not low enough (relatively) for us to argue that turnout has declined because the electorate does not know what the Parliament is doing. This is a crucial message it does not appear to be the case that low turnout can be explained by the fact that either the Scottish Parliament is actually not doing anything, or because it has failed to communicate what it is doing successfully. 6.3 Does it make a difference who wins? Westminster election winners are considered to have more of a chance of making a difference than Scottish Parliament or local council election winners, but at least one in three think it winning any election will not make much or indeed any difference at all. The electorate are rather unsure about whether either the Parliament or their local council has a better chance of improving things, with 42% saying that the Parliament can make a great deal or quite a lot of difference, compared to 41% who think this is the case with the local council. Of course, there are at least two factors that must be taken into account when considering this issue. Firstly, if many Scots have the impression that the Parliament doesn t or can t actually do anything and 37% cited it as a reason why they didn t vote - then the Scottish Parliament is bound to suffer in comparison to Westminster. Secondly, if there are no perceived differences between the policies of the major parties in Scotland, or in Westminster for that matter, then this also can potentially also have an impact on turnout. Indeed, over half (51%) of nonvoters do think that it doesn t make very much difference, or no difference who wins the Scottish Parliament election because things will go on much the same. 14

15 HOW MUCH OF A DIFFERENCE DOES IT MAKE WHO WINS. ICM Research - Scottish Elections Research May 2003 House of Commons elections Scottish Parliament elections Local council elections Q35. How much of a difference do you think it makes who wins.house of Commons/ Scottish Parliament/local council elections? Base: All respondents (1100) 10 A great deal Quite a lot Some Not very much None at all 15

16 7. Attitudes Toward Voting 7.1 Agreement with statements Many people do think they have a civic obligation to go and vote, normally grounded in the knowledge that people have died to protect the franchise for future generations, or else they feel they have no right to complain about things thereafter. People are often embarrassed to admit they have not voted in a particular election, a fact that opinion polls have struggled to deal with for many years given their tendency to overestimate turnout on a regular basis. AGREEMENT WITH STATEMENTS: ICM Research - Scottish Elections Research May 2003 I felt it was my duty to vote It was obvious who was going to win There wasn't much evidence an election was taking place There wasn't much difference between the parties Q19. I am going to read out a number of statements some people have made about the Scottish Parliament election. For each one, please tell me whether you Base: All respondents (1100) Agree strongly Tend to agree Neither agree nor disagree Tend to disagree Disagree strongly As the chart shows, over half (54%) agree strongly that they had a duty to vote, with a further one in five (20%) tending to agree. A similar number (22%) tended to disagree or disagreed strongly. Agreement increases with age, with 85% of those aged 65+ net agreeing compared to only 63% net agreement among year olds. Voters are twice as likely to agree as non-voters, with 94% of voters in both the Scottish Parliament election and the Scottish council election agreeing, compared to only 47% and 49% respectively among non-voters. The correlation between actual turnout and having a duty to vote can be further demonstrated through analysis of the voter commitment variable. The table below shows the extent to which voters feel they have a duty to vote by the number of times they have actually voted in the main four elections covered by the survey. Clearly, the 16

17 link is strong, with almost all of those who regularly turn out believing there to be a responsibility to vote. % who agree strongly or tend to agree it was their duty to vote VOTER COMMITMENT Number of times voted* Net agreement 28% 56% 63% 94% 95% Net disagreement 63% 33% 32% 6% 5% *Elections include: Scottish Parliament 2003, Local Council 2003, General Election 2001, Scottish Parliament 1999 Opinion poll evidence has also previously suggested that people who think that the result of an election is obvious beforehand are less likely to vote in that election. This has particularly applied to General Elections but evidently does NOT apply to these Scottish elections. While it is the case that net agreement (54%) outweighs disagreement (39%) on the statement that it was obvious who was going to win, it is more useful to once again look at the splits between voters and non-voters. In this case, more voters (57%) than non-voters (48%) felt that the result was obvious. Of course, there are a number of separate factors at play in the Scottish Parliament election that are not in any General Election. The most important is the system of proportional representation used for Holyrood, which some people find difficult to understand (qualitative evidence is available to support this) and could make the result less clear, especially if no overall majority is thought to be achievable. Less important, but still apparent is the apathetic factor some people did not particularly realise that an election was taking place, and many more probably did not care. Indeed, the chart above shows 40% net agreement on the statement there wasn t much evidence that an election was taking place, (although more 54% disagreed with this). A further factor that may have had an effect on turnout was the perceived similarity between the major parties, with 48% agreeing this was the case and 42% disagreeing. However, it is again the case that voters were more likely to agree (51%) this was the case than non-voters (42%). The chart below shows some of the other factors that were involved in voting in the Scottish Parliament and other elections. Firstly, there is a hardcore of people who say they will never vote, with 8% agreeing strongly that this is the case as far as they are concerned. However, one in seven of these hardcore non-voters also say they did actually turnout in the Scottish Parliament election, so some caution is advisable. The Scottish Parliament itself is not yet thought to be a powerful political institution, with 86% disagreeing with the statement that it has too much power. Two in three (62%) also disagree that the campaign was interesting, while a similar number (61%) disagree that the Scottish Parliament elections are more important than Westminster elections. Over half (54%) disagreed with the statement that 17

18 none of the parties stood for the policies I believe in, with voters for each of the main four Scottish parties displaying almost identical levels of strong disagreement. AGREEMENT WITH STATEMENTS: ICM Research - Scottish Elections Research May 2003 I never vote at elections and never will The Scottish Parliament has too much power It was an interesting election campaign Scottish Parliament elections are more important than UK general elections None of the parties stood for the policies I believe in Q19. I am going to read out a number of statements some people have made about the Scottish Parliament election. For each one, please tell me whether you Base: All respondents (1100) Agree strongly Tend to agree Neither agree nor disagree Tend to disagree Disagree strongly 18

19 8. Voting Behaviour 8.1 The decision to turnout Two in three voters (57%) made up their mind well before polling day that they were going to vote. Indeed, 63% of those voters who have turned out in every recent major election made their decision to vote in this one before April in other words they probably decide to vote in all elections no matter what. TIMING OF DECISION ABOUT WHETHER TO VOTE. ICM Research - Scottish Elections Research May Election day During the week before 15 9 Sometime earlier in April Q12/Q15. When did you decide not to vote/you were going to vote? Base: All non-voters (except those not registered (408)) OR all voters (624) Before that Voters Non Voters A large proportion of non-voters (31%) also admit that they made the decision not to vote long before polling day. This figure rises to 55% among those who have not voted in any of the four major elections, suggesting that a core of serial non-voters probably around 6% of the total Scottish population - are fairly secure in their intention not or perhaps never- to vote. Many non-voters, however, say they are victims of circumstance. Four in ten (39%) made the decision not to vote on Election Day itself, suggesting that factors conspired to prevent them from getting to their polling station. These people are more likely to be female (44%) than male (33%) and slightly lower down the social classification scale. The youngest cohort (40%) is likely to have made the decision on the day, but then so are years olds (48%). 19

20 8.2 Counter factual voting decisions A number of questions were asked about people s decision to vote if alternative elections were or were not actually being held on the day of the Scottish Parliament and Scottish council elections, in order to identify whether inclination to vote depends upon the particular political institution being voted for. In total, 71% say they would have voted if the election had been for Westminster. However, 73% recall voting in last 2001 General Election when the actual turnout was 58%, so we might consider that this particular finding is somewhat on the high side. VOTED IF WESTMINSTER ELECTION INSTEAD OF SCOTTISH ELECTION? ICM Research - Scottish Elections Research May Voters 7 47 Non-voters Q18. Talking to people we have found that a lot say they would not have voted if a General Election to the UK House of Commons had been held on May 1 st instead of a SP election. How about you, do you think you would have voted in a House of Commons election on May 1 st or do you think you would not have voted? Base: All respondents (1100) Yes No 51 As the chart shows, if the election were actually a General Election to the Westminster Parliament, the vast majority (90%) of (Scottish Parliament) voters would also have turned out for that election. Only 7% say they would not have voted in a Westminster election. Among this 7%, only one in four (27%) voted for the SNP in the Scottish Parliament election, so we cannot assume that these potential abstentions are primarily based on a refusal to vote in a UK based election on grounds of Scottish independence. Non-voters are fairly equally split on whether they would have voted had the election been for Westminster. Just under half (47%) say they, in fact, would have voted in a General Election, and there is probably substance to this. Of the 47% who did not vote in the Holyrood election but say they would have voted if it were a Westminster election, almost three quarters of them (70%) do recall voting in the 2001 Westminster election. 20

21 The other half (51%, representing 22% of the total sample) would still not have voted if the election had been for Westminster. We can surmise that many of these are solid non-voters, who rarely turn out irrespective of the election that was taking place. A further question was asked about the local council elections taking place on the same day, with the vast majority of voters (87%) saying that they would have voted in the council elections even if they had not been held on the same day as the Scottish Parliament election. Only one in ten (12%) would not have turned out had this been the case. VOTED IN COUNCIL ELECTION IF NOT ON SAME DAY AS SCOTTISH PARLIAMENT ELECTION? ICM Research - Scottish Elections Research May Voters Non voters Q22. Talking to people we have found that a lot say they would not have voted in the local council elections if the SP election had not been held on the same day. How about you, do you think you would have voted in the local council elections if the SP election had not been held on the same day, or would you not have voted? Base: All who voted in local council election (609) Yes No 66 Perhaps rather surprisingly, one in three (33%) non-voters in the Scottish Parliament election say they would have voted in the council election had it NOT been on the same day as the Parliament election! It is difficult to rationalise why so many non-voters say this, but we can speculate that there are two potential arguments. Firstly, that many did not know that the local council elections were taking place on the same day because the Parliament election received all the publicity, or secondly that the decision not to vote was a deliberate abstention that reflected their opposition to the existence of the Scottish Parliament. 21

22 83 Reasons why non-voters did not vote Market research more often than not attempts to rationalise attitude and behaviour rather than inaction. While respondents find it relatively easy to identify the reasons why they do something, they can find it difficult to rationalise why they DON T. However, within the context of this survey we asked people to tell us why they didn t go to vote. Spontaneously, one in five (22%) members of the Scottish electorate say they were working, had other things to do, or were simply too busy to vote. There is a clear age correlation, with younger people most likely to tell us this, older people least likely. The proportion rises to 30% among those who are working full time. REASONS WHY NON VOTERS DID NOT VOTE (UNPROMPTED) ICM Research - Scottish Elections Research May 2003 Working/busy Politicians don't keep promises Not registered Not interested/not bothered Waste of time/makes no difference Didn't know who to vote for Illness/old age Away/abroad Lost polling card Polling station too distant Don't agree with Scot Parl Labour always win Other No reason Q10. People have many different reasons for not voting in elections. Why did you not vote in the SP election on May 1 st? Base: All non-voters: (476) % A further 10-20% provide other circumstantial reasons for not tuning out. One in 15 (7%) were too old or ill to get to their polling station; one in 20 (5%) was away, and small numbers either lost their polling card (4%) or thought their polling station was too far away (2%). Many people also gave deliberate reasons for not voting, the most important being a cynical impression of politics and politicians who are thought not to be trusted because they don t keep their promises (19%). Cynicism usually develops with age, as is the case here. While only 9% of year olds adopt this position, the proportion rises to 24% of year olds and 28% of those aged years. One in ten (11%) also say that voting is a waste of time, or makes no difference as the Scottish Parliament has no power. 22

23 Apathetic reasons for not turning out were also admitted by almost one in four. A total of 13% say they are not interested in politics and could not be bothered to vote, and 10% simply did not know who to vote for (the incidence of saying this declines with age). Non-voters were then provided with a list of prompted reasons that may rationalise why they didn t vote, responses to which are provided in the chart below. Interestingly, a deliberate reason for not voting not trusting politicians to keep their promises (57%) came top. Those over the age of 55 years were most likely to say this (rising to 70% among those 55-64); while a sizable majority (68%) of those people who did not vote in any of the four elections covered by the survey also said this. PROMPTED REASONS WHY PEOPLE DID NOT VOTE. ICM Research - Scottish Elections Research May 2003 You just can't trust politicians to keep their promises I really intended to vote but circumstances on the day prevented me There was no point in voting because all the parties are all the same There was no point in voting because the SP can't do very much I'm just not interested in politics There was no point in voting because it was obvious who would win The voting system was too confusing None of these Q11. I am going to read out a number of reasons that other people have given us for not voting in the SP election. For each one, could you please say whether or not it was a reason Why you did not vote in the SP election on May 1 st? Base: all non voters except those not registered (408) % A circumstantial reason came second, with 45% saying they intended to vote but could not do so on the day for one reason or another. A host of apathetic reasons were also provided by one in three in each case. Only one in ten (13%) say that they did not vote because they find the proportional voting system too confusing for them. Many of these non voters are over the age of 65 years, with 47% of this cohort agreeing that this was a reason why they did not turn out. Council renters (24%), those people not working and not seeking work (30%) and DE s (23%) also disproportionately gave this as a reason (although we might expect considerable cross-over between these demographic variables). 23

24 8.4. Reasons why voters did vote Voters in the Scottish elections provide two main sets of reasons to explain why they cast their vote. The first set encapsulates the political dimension, with the single most important reason supporting or opposing the party of their choice - mentioned more often than any other (29%). Other more rarely mentioned political reasons include hoping for a change/better future (10%), to get Scottish independence (4%)and not liking the ways things are currently run (3%). The second set of reasons can be generalised as being principled. One in four (28%) voted out of duty, 22% wanted to voice their opinion, 21% simply wanted to exercise their right to vote and 13% think that if you don t vote then you don t have a right to complain about the government. REASONS WHY VOTERS DID VOTE? (UNPROMPTED) ICM Research - Scottish Elections Research May 2003 Support/oppose party Out of duty Want to voice opinion Exercise my right to vote No vote - no right to complain Hoping for change Always vote To get Scottish independence Don't like government Other Don't know Q14. People have many different reasons for voting in elections. Why did you vote in the SP election on May 1 st? Base: All voters: (624) %

25 9. Why Turnout Fell In this section we examine why turnout in the 2003 Scottish Election was as much as nine points lower than it was in both the last Scottish Parliament election in 1999 and the last UK general election in Does this drop reflect yet another increase in voter apathy, or was it occasioned by the fact that this particular election secured voters interest to an even lesser degree than did the relatively low turnout elections of 1999 and 2001? We adopt a two-fold approach to answering this question. First we compare the results of our survey with that of previous Scottish and British surveys that carried the same or similar questions. If, for example, the low turnout reflects growing voter apathy then we should find that fewer voters in our survey say they are interested in politics or believe that people have a duty to vote than was the case in surveys undertaken in 1999 or Then, we report the results of some multivariate analysis of our survey with a view to identifying the factors that appear to have dissuaded those who said they voted in either 1999 or 2001 from doing so in We begin by looking at the time trends in a number of questions that tap voter apathy and disengagement. The first is the simplest of all - the proportion of people who say they are interested in politics. As the next table shows, there is in fact no evidence that interest in politics in Scotland is on the decline. Indeed, at 33%, the proportion of people who in our survey who express a great deal or quite a lot of interest in what is going on in politics is somewhere higher than the 27% who expressed that view immediately after the 1997 UK general election, even though turnout was more than 20 points higher at that election than it was in TRENDS IN POLITICAL INTEREST % 1999 % 2001 % 2003 % Great deal 9% 6% 7% 11% Quite a lot 18% 17% 19% 22% Some 35% 36% 33% 28% Not very much 30% 31% 27% 25% None 7% 9% 13% 14% Sources: 1997: Scottish Election Survey; 1999 and 2001: Scottish Social Attitudes Survey 25

26 Another indicator of voter engagement is whether or not they believe they have a duty to vote, something which has long been found to be strongly associated with whether people vote or not. Here too, we can find little evidence of disengagement. The 70% of people who in our survey said that it s everyone s duty to vote was very similar to the 69% who gave the same response in the 2001 Scottish Social Attitudes survey. And although this question was not asked after the 1999 election in Scotland, our reading is also well in line with readings of between 64% and 68% obtained by the British Social Attitudes survey across Great Britain as a whole on four occasions between 1991 and 1998 (Bromley and Curtice, 2002). But if voters are no less interested in politics and no less likely to believe that they had a duty to vote, they are less trustful nowadays of politics and the political system. Only one in four people in Scotland now say that they trust governments to put the needs of the nation above the interests of their party at least most of the time, whereas over one in three took that view in the mid 1980s. However, the decline in trust in government appears to date from the mid 1990s. As the following table shows trust in government is certainly no lower now than it was two years ago, while there is no reason to believe from the data that are available that it is lower now than it was at the time of the first Scottish Parliament election. So even if scepticism about politicians may make it rather more difficult to persuade voters to go to the polls, there is little reason to believe that this is any more the case now than it was in 1999 or TRENDS IN TRUST IN GOVERNMENT % trust governments of any party to place the needs of the nation above the interests of their own political party just about always / most of the time Mid Mid s 1990 s 37% 29% 29% 13% 27% 25% Sources: 1980s: 1987/87 British Social Attitudes survey (Scottish respondents only); 1990s: 1994/1996 British Social Attitudes survey (Scottish respondents only); 1997: Scottish Election Study; 2000, 2001: Scottish Social Attitudes survey. Note that these earlier surveys asked about British governments. So we can find little evidence of growing voter apathy or disengagement. Our attention thus turns naturally to what features about this election might have discouraged voters from voting. Here it is useful to make a distinction between two rather different aspects of the election that might have been responsible. The first possibility is that voters reluctance to vote is a reflection of their attitudes towards or evaluations of the Scottish Parliament. The second consideration is that voters may have found the choices put before them at the election by the political parties unattractive or unimportant. The first of these two explanations suggests that the fall in turnout at this election was primarily the result of the public s reaction to devolution and does not necessarily have implications for other kinds of elections. The second, in contrast, is something that could also replicate itself in other elections, including those for the House of Commons. 26

27 We have already shown in section 5 that attitudes towards devolution itself have changed little in recent years. Just 22% say that Scotland should not have any kind of Parliament, little different from the 18% who took that view when ICM asked the same question a few months before the 1999 Scottish election. So the fall in turnout cannot be blamed on a significant growth in hostility towards the existence of a Scottish Parliament. But while people in Scotland may still support the principle of having their own Parliament, this does not necessarily mean that they think that it is proving to be an important institution in practice. TRENDS IN PERCEIVED INFLUENCE OVER HOW SCOTLAND IS RUN Scottish 41% 13% 15% 24% Parliament UK Government 39% 66% 66% 50% Local councils 8% 10% 9% 9% European Union 4% 4% 7% 10% The question wording was as follows:- 1999: Which do you think will have most influence over the way Scotland is run? : Which do you think has most influence over the way Scotland is run? Source: ; Scottish Social Attitudes survey As the next table shows, at the time of the first Scottish Parliament election in 1999, slightly more people thought that the Scottish Parliament would have the most influence over the way Scotland was run in future than thought that the UK government would. Now, perceptions are very different. Twice as many people think that the UK government has most influence over the way that Scotland is run as think that the Scottish Parliament does. True, there does appear to have been an increase in the perceived importance of the Parliament as compared with perceptions during the first two years of its life, but it still means that the relative importance of the Scottish Parliament in voters eyes was lower at the time of the 2003 election than it was when they were first asked to vote for the body four years previously. This finding is corroborated by the results of a second question on the perceived importance of the outcome of elections, whose results are shown in the next table. After the 2003 election just 42% said that they thought that who won a Scottish Parliament election made a great deal or quite a lot of difference, well down on the 56% who took that view four years ago. Of course this perception could reflect a belief that there was not much difference between the parties as well as a perception that the Scottish Parliament is relatively un-influential. However, the proportion who believe that the outcome of House of Commons elections makes a difference is, at 52%, both higher than it is for Scottish Parliament elections and only a little down on the 55% who took that view four years ago. It thus looks likely that a significant proportion of the decline in the perceived importance of the outcome of Scottish Parliament elections is a consequence of a decline in the perceived importance of the institution itself. 27

28 TRENDS IN PERCEIVED DIFFERENCE WHO WINS AN ELECTION MAKES Perceived Difference Scottish Parliament Elections House of Commons Elections Great deal 28% 15% 23% 27% 16% 29% Quote a lot 28% 28% 19% 28% 29% 23% Some 19% 22% 19% 18% 19% 15% Not very 18% 25% 24% 22% 25% 22% much None at all 5% 8% 13% 4% 8% 10% Source: 1999, 2001: Scottish Social Attitudes survey At the same time, however, many voters did have the impression that there was little difference between the parties. Moreover this impression was somewhat more widespread in 2003 than it was four years earlier, if not necessarily as compared with the position at the time of the last UK general election. Certainly, as the next table shows, even the difference between Labour and the SNP is thought to be much smaller now than the difference between Labour and the Conservatives was thought to be just a decade ago. TRENDS IN PERCEIVED DIFFERENCE BETWEEN THE PARTIES Conservative and Labour Great 53% 32% 30% 21% 24% Some 28% 44% 35% 39% 36% Not much 17% 20% 32% 38% 37% SNP and Labour Great Na Na 41% 33% 37% Some Na Na 40% 42% 37% Not much Na Na 15% 19% 23% Source: 1992, 1997: Scottish Election Study; 1999, 2001, Scottish Social Attitudes survey Further evidence that voters did not perceive much difference between the parties, even as compared with other recent elections, come from the responses to another question in our survey. As many as 49% indicated that they agreed with the statement, There wasn t much difference between the parties, whereas two years ago only 38% of respondents across the UK as a whole agreed with a very similar statement, There was very little difference between what the main parties were offering, that was included in MORI s survey for the Electoral Commission. But however similar or different the parties messages are, in modern elections all parties rely heavily on their leaders to communicate their message. It appears that here too the parties had difficulties in engaging the interest of the electorate. As many as 37% said that felt they had received too little information about the party leaders, whereas only just under a quarter of people across Britain as a whole said 28