Background Essay on the Campaign Protestant Evangelicals in Politics: Who They Are and COMPAS Inc. Public Opinion and Customer Research June 3, 2004
Introduction Evangelicalism has become a factor in the federal campaign as a result of Liberal strategists introducing doubt about the suitability for national leadership of a party whose leadership may be unduly devout or Evangelical. This mini-report data-mines the National Post/Global/CanWest/ COMPAS national survey (fielded May 14-19, n>1400) to explore public attitudes towards Evangelicalism and the role of Christian believers in politics. This opinion survey was the most extensive psychological and policy study undertaken for the media over the course of the campaign. Most Protestants at Least Somewhat Comfortable with the Evangelical Label Most Canadians Comfortable with a Christian Believer in Office Very few (1%) of English-speaking Canadians voluntarily describe themselves as Evangelical when asked for their religious denomination. That may be because Evangelical is often considered an approach to life rather than an actual denomination. As an approach to life, the term does resonate with sizeable numbers of people, as shown below. After revealing their religious denominations, those respondents in the survey who were Protestant were then asked to what extent they would characterize themselves as Evangelical. About two-fifths reply that they are definitely not Evangelical, one fifth that they are definitely Evangelical, and the remainder in between, as shown in table 1. The categories outlined in table 1 were then analysed statistically to discover relationships with political attitudes and patterns of voting. The statistical analysis revealed two categories of attitudes definite non- 2
Evangelicals vs. all others. In light of this binary pattern, all our reporting below utilizes this binary distinction. Those who say that they are definitely not Evangelical are termed non-identifiers; all others are termed friendlies, as shown in table 2. By conventional scholarship, adherents of the Anglican and United Churches are deemed religious liberals. As apparent, religious liberals, they might not be expected to describe themselves as Evangelical or somewhat Evangelical. But such a sharp pattern does not emerge. To the contrary, there is no great difference in the propensity of Anglicans and United Church members to describe themselves as Evangelical compared to members of other denominations. In practice, 36% of Evangelical friendlies belong to the Anglican and United Churches while 45% of nonidentifiers do. Table 1: Suppose you were asked if you were an Evangelical Christian. Would you say % Definitely yes 21 Probably 12 Probably not 21 Definitely not 42 DNK/Refused 5 Table 2: Evangelical Friendlies and Nonidentifiers 1 % 1 Respondents who classified themselves as Protestants were asked Suppose you were asked if you were an Evangelical Christian. Would you say Respondents who responded definitely yes, probably yes and probably not were classified as Evangelical Friendly while respondents who said definitely not were classified as Evangelical Nonidentifiers. Respondents who classified themselves as Evangelicals were coded as definitely yes and Evangelical Friendly. 3
Evangelical Friendlies 54 Evangelical Nonidentifiers 42 DNK/Refused 5 Table 3: Denominational Profiles of Evangelical Friendlies and Non-Identifiers (Vertical Percents) EVANGELICAL FRIENDLIES EVANGELICAL NONIDENTIFIERS % % Baptist, Lutheran, Other Protestant, Christian 53 49 United Church, Anglican 36 45 Liberal Protestant 7 6 Evangelical 4 0 We reported above that a majority of Protestants are comfortable or non uncomfortable with the label, Evangelical. Given this finding, one might expect a lot of Canadians to be comfortable with the idea of religious individuals running our government. In practice, far more Canadians are comfortable with this possibility than are uncomfortable. Some Canadians believe that there is validity in the criticism that Harper and some of his advisors are Christian Church-goers, who shouldn t be running government, as shown in table 4. However, those Canadians who see no or little validity in this criticism greatly outnumber those who see some or much validity in it. Approximately one in seven Canadians scores high agreement with the criticism (6-7 on the 7 point agree-disagree scale). Agreement with the criticism is higher among Canadians as a whole than among Protestants who do not identify as Evangelical and especially among Evangelical friendlies. The small minority who agree with the preceding criticism of the Harper-led Conservatives is greatly outnumbered by those who see no 4
merit in the criticism. Opposition to the criticism is especially strong among Evangelical friendlies, among whom 52% score 1-2 on the scale. Given the large numbers of Canadians who are comfortable with seeing religious people in high office and given the large number of Protestants who are not uncomfortable with the label of Evangelical, a reasonable conclusion is that Liberals expression of concern about an excessive Evangelical or Christian influence in the Conservative party may have propelled some devout individuals towards the Conservatives. Table 4: Rejection of Anti-Christian Criticism of Harper and Conservatives. Please score each of the following criticisms of the Conservatives on a 7 point scale where 7 means that the criticism is valid and a persuasive reason to vote against the Conservatives and 1, the opposite Harper and some of his advisors are Christian Church-goers, who shouldn t be running government. All Canadians, including RCs, non- Christians Evangelical Friendly Protestants Evangelical Nonidentifying Protestants TOP TWO BOX BOTTOM TWO BOX 7 6 5 4 3 2 1 DNK 16 42 11 5 13 8 10 11 31 12 12 52 9 3 13 7 8 12 40 8 17 42 12 5 14 13 12 11 31 2 5
Evangelical Friendlies Largely But Not Overwhelmingly Conservative Voters As shown in table 5, Evangelical friendlies and nonidentifiers differ in their voting preferences but not sharply. Friendlies lean to the right with 50% voting Conservative and only 15%, NDP. Nonidentifiers lean left with 22% NDP and only 34% Conservative. Below, we report patterns of attitudes on ballot questions relating to social and cultural issues. Evangelical friendlies tend to the right on these issues. These issues make the Conservative party a comfortable home for a large share of Evangelical friendlies (50%). However, the litmus test issues do not generally propel friendlies to the Conservative party any more than they would otherwise go to the party. Evangelical friendlies tend strongly to the Conservative party, and are broadly conservative in their views on social issues. But Evangelical friendlies are not so committed to these socially conservative positions as to increase their voting support for the Conservative party if these socially conservative issues were their only electoral considerations. We provide details below The Conservative Party led by Stephen Harper The Liberal Party led by Paul Martin The NDP led by Jack Layton The Bloc Quebecois led by Gilles Duceppe Table 5: Vote Intention EVANGELICAL EVANGELICAL DNK / FRIENDLY NONIDENTIFIERS REFUSED % % % 50 34 40 31 40 43 15 22 13 1 1 0 Other 2 3 4 6
Evangelical Friendlies Are Social Conservatives But the General Appeal of the Conservative Party is as Important as Most Socon Issues As might be expected, the issue of gay marriage distinguishes Evangelical friendlies from nonidentifiers but the difference is not profoundly greater than the differences in party preferences. As evidenced in table 6, the Conservative vote would go up by 5 percentage points (to 55 from 50%, as shown in table 5) on the basis of the gay marriage issue, as that issue is defined in table 6. Table 6: Gay Marriage and Evangelical Vote Intention (Q11) Suppose the Conservatives said that homosexuals should have a right to a civil union but only a man and a woman should be allowed to marry and adopt children, and suppose the other parties said that homosexuals should have a right to marriage and adoption. If you based your vote on this issue, how would you vote? EVANGELICAL FRIENDLY EVANGELICAL NONIDENTIFIERS DNK / REFUSED % % % The Liberal Party led by Paul Martin 28 39 23 Conservative Party led by Stephen Harper 55 37 77 NDP led by Jack Layton 13 23 0 Bloc Quebecois led by Gilles Duceppe 3 0 0 [UNPROMPTED] Other 2 1 0 7
[UNPROMPTED] Would not vote (excl. from calc.) [UNPROMPTED] DNK/ Undecided/ Refused (excl. from calc.) 1 2 3 16 6 41 Table 7 lists the impacts on the Conservative vote among Evangelical friendlies and non-identifiers of a variety of ballot issues. These issues are listed in descending order according to the magnitude of the difference in impact among these two groups of Protestants. At the top are concerns about gay marriage, terrorism, the gun registry, and health and education. These are the issues that most separate Evangelical friendlies from nonidentifiers. Two important general findings are that Evangelical friendlies remain to the right of nonidentifiers in their voting preferences and that none of the issues truly increases the Conservative vote among friendlies beyond the base in table 5. Table 7: Evangelical Support for the Conservative Party (Q6B) If this were the only factor in your thinking, for which party would you vote? [REPEAT LIST ONLY IF NECESSARY] 8
EVANGELIC AL FRIENDLY 2 EVANGELICA L NONIDENTIFI ERS DNK Gay marriage (Conservative Gap between Friendlies and Non-Identifiers = 25) Conservative 52 27 45 Liberal 31 41 55 NDP 14 30 0 Security against terrorism (Conservative Gap = 23) Conservative 51 28 49 Liberal 33 52 51 NDP 13 18 0 The gun registry (Conservative Gap = 21) Conservative 56 35 38 Liberal 28 36 55 NDP 12 27 7 2 Respondents who classified themselves as Protestants were asked Suppose you were asked if you were an Evangelical Christian. Would you say Respondents who responded definitely yes, probably yes and probably not were classified as Evangelical Friendly while respondents who said definitely not were classified as Evangelical Nonidentifiers. Respondents who classified themselves as Evangelicals were coded as definitely yes and Evangelical Friendly. 9
EVANGELIC AL FRIENDLY 2 EVANGELICA L NONIDENTIFI ERS DNK Healthcare and education (Conservative Gap = 21) Conservative 46 25 53 Liberal 29 44 48 NDP 21 29 0 Getting a change after three Liberal governments (Conservative Gap = 19) Conservative 57 38 53 Liberal 22 31 40 NDP 19 28 7 Other 2 3 0 The economy (Conservative Gap = 18) Conservative 51 33 33 Liberal 33 49 60 NDP 13 17 8 10
EVANGELIC AL FRIENDLY 2 EVANGELICA L NONIDENTIFI ERS Leadership (Conservative Gap = 18) DNK Conservative 47 29 46 Liberal 36 47 54 NDP 14 23 0 Other 2 1 0 Taxes (Conservative Gap = 16) Conservative 51 35 45 Liberal 31 43 48 NDP 16 21 7 Protecting [OWN PROVINCE] in decision-making process in Ottawa (Conservative Gap = 16) Conservative 50 34 46 Liberal 31 41 54 NDP 16 23 0 Bloc 1 1 0 Keeping Quebec in Confederation (Conservative Gap = 16) 11
EVANGELIC AL FRIENDLY 2 EVANGELICA L NONIDENTIFI ERS DNK Conservative 43 27 42 Liberal 40 55 58 NDP 13 15 0 Other 3 3 0 Scandal and waste (Conservative Gap = 15) Conservative 55 40 50 Liberal 26 27 37 NDP 16 32 14 Bringing Western Canada into the decision-making process in Ottawa (Conservative Gap = 12) Conservative 54 42 53 Liberal 28 40 48 NDP 14 17 0 Other 1 1 0 Bloc 3 0 0 Agriculture (Conservative Gap = 7) Conservative 44 37 32 Liberal 32 40 54 12
EVANGELIC AL FRIENDLY 2 EVANGELICA L NONIDENTIFI ERS DNK NDP 21 21 14 Issue Priorities The Paradoxicality Shared by Evangelical Friendlies and All Canadians Table 8 lists in descending order of importance the issues that friendlies claim drive their vote. Paradoxically, the issues that they say are most important (e.g. healthcare and the economy) are similar to the judgements of all Canadians even if friendlies vote more to the right than the general public. Evangelical friendlies share another paradox with the public as a whole the issues that they say are most important appear from statistical analysis to not be the issues actually driving their vote. For example, gay marriage and security against terrorism are major vote drivers because these are the issues that most propel Evangelical friendlies to the Conservatives relative to the voting behaviour of Protestants who do not identify as Evangelical. Yet, Evangelical friendlies name healthcare/education and the economy as the issues most important to them. Table 8: Issue Importance for Evangelical Voters 13
EVANGELICAL EVANGELICAL FRIENDLY NONIDENTIFIERS DNK Healthcare and education Top 2 Box 86 89 82 Bottom 2 Box 1 1 6 Economy Top 2 Box 73 73 65 Bottom 2 Box 1 2 0 Taxes Top 2 Box 68 52 70 Bottom 2 Box 2 6 6 Leadership Top 2 Box 66 66 80 Bottom 2 Box 2 7 0 Protecting [OWN PROVINCE] in decision-making process in Ottawa Top 2 Box 65 55 55 Bottom 2 Box 1 12 0 Scandal and waste Top 2 Box 63 58 56 Bottom 2 Box 6 8 25 Bringing Western Canada into the decision-making process in Ottawa Top 2 Box 62 59 75 Bottom 2 Box 7 9 6 Agriculture Top 2 Box 59 48 83 Bottom 2 Box 5 9 0 Security against terrorism Top 2 Box 58 52 74 Bottom 2 Box 10 11 7 Getting a change after three Liberal governments Top 2 Box 52 45 56 14
EVANGELICAL EVANGELICAL FRIENDLY NONIDENTIFIERS DNK Bottom 2 Box 19 21 26 Keeping Quebec in Confederation Top 2 Box 46 50 56 Bottom 2 Box 17 15 19 The gun registry Top 2 Box 35 23 38 Bottom 2 Box 38 42 13 Gay marriage Top 2 Box 27 20 32 Bottom 2 Box 44 45 44 15