skip to main content
Democrats' Election Strength Evident Across Voter Segments

Democrats' Election Strength Evident Across Voter Segments

Independent support a key to Democratic success

by Frank Newport, Jeffrey M. Jones, Lydia Saad and Joseph Carroll

GALLUP NEWS SERVICE

PRINCETON, NJ -- An analysis of Gallup's final pre-election poll data shows that the Democratic victory in Tuesday's House elections was because of a rising Democratic tide that lifted support in almost all key subgroups. In addition to solid support from their core constituent groups such as liberals, nonwhites, women, urban residents, and older Americans, Democrats also owe a significant debt to independent voters, who tilted strongly in their direction. Whites and those who are married -- groups that usually favor the Republican Party -- were evenly divided in their vote. Democrats did better among rural voters, a change from previous voting patterns.

Opposition to the Iraq war appears to have helped the Democratic cause. Although supporters and opponents of the war voted about equally for the Republican and Democratic parties, respectively, the fact that opponents outnumbered supporters made Iraq a net-plus for Democrats. In general, it appears that Democratic efforts to get out the vote played a significant part in the win, because those who were contacted by Democrats and urged to vote for Democratic candidates strongly supported Democrats, even if they were also contacted by Republicans.

The following breaks down the vote in the 2006 congressional elections among likely voters in Gallup's Nov. 2-5 poll, according to key political, issue, and demographic groups.

Vote by Party ID

A major factor in the Democratic Party's strength this election was the solid support for its candidates among political independents. According to Gallup's final pre-election poll, a 55% majority of independents (who comprise 27% of the "likely voter" pool) planned to vote for the Democratic candidate in their district, while only 38% planned to support the Republican. This represents a change from the last midterm election four years ago, when independents were more closely divided in their preferences: 46% voted Democratic and 43% voted Republican.

Republicans and Democrats were nearly unanimous this year in saying they would vote for their own party's candidate for president. The rate was 95% among Democrats and 93% among Republicans, similar to the pattern seen in 2002.

Generic Ballot by Party ID
Based on likely voters

Democratic
candidate

Republican
candidate

%

%

Republicans

5

93

Independents

55

38

Democrats

95

3

Vote by Ideology

Similar to the pattern by party identification, Gallup finds sharp differences in vote preferences according to voters' political views. Three-quarters of self-described conservatives (77%) supported the Republican in their district, while self-described liberals were even more solidly united behind Democratic candidates (91%). Political moderates tilted strongly Democratic this year, with 67% backing Democrats and only 29% voting for Republicans.

Comparing these figures with 2002, there was virtually no change in the party preferences of conservatives, but the preferences of liberals and moderates have shifted toward the Democrats. In 2002, the percentage of each supporting Democrats was 18% among conservatives, 57% among moderates, and 84% among liberals.

Generic Ballot by Political Ideology
Based on likely voters

Democratic
candidate

Republican
candidate

%

%

Conservative

18

77

Moderate

67

29

Liberal

91

3

Vote by Party/Ideology

Combining likely voters' party identification with their political ideology, we see that the vast majority of Democrats this year, regardless of ideology, planned to vote for the Democratic candidate in their district: 98% of liberal Democrats and 93% of conservative/moderate Democrats. By contrast, while conservative Republicans were nearly unanimous in supporting their Republican candidates (97%), only 81% of moderate-to-liberal Republicans did the same. A sizable minority of nonconservative Republicans (10%) said they were voting Democratic, while another 9% said they were voting for a candidate of another party, or were undecided.

This slippage among the Republican base, combined with the near-perfect unity among the Democrats, could provide an important window into the political dynamics behind the change of power in Congress.

Generic Ballot by Party/Ideology
Based on likely voters

Democratic
candidate

Republican
candidate

%

%

Liberal Democrats

98

2

Conservative/Moderate Democrats

93

4

Moderate/Liberal Republicans

10

81

Conservative Republicans

3

97

 

 

Vote by Candidate Supported in 2004 Presidential Election

Those who voted for John Kerry in 2004 were much more loyal to the Democratic Party's 2006 congressional candidates than 2004 Bush voters were to 2006 Republican Party candidates. Kerry voters supported Democratic House candidates over Republican candidates by a 94% to 4% margin, while Bush voters supported Republicans over Democrats by a smaller 83% to 14% margin. This suggests that most of the party switching from the 2004 to the 2006 elections went from the Republicans to the Democrats.

 

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Voted for John Kerry in 2004

94

4

430

Voted for George W. Bush in 2004

14

83

503

Party contact seems to make a difference in vote choice, but that is partly because Republicans and Democrats may be targeting known supporters of their parties. Among those contacted only by Democrats, 74% planned to vote for the Democratic candidate and 24% for the Republican. Among those contacted only by Republicans, 65% supported the Republican candidate and 34% the Democrat. Democratic candidates fared better among voters who were contacted by both sides, with 55% supporting the Democrat in their local district and 36% the Republican. Voters who were not contacted by either party were evenly divided in their vote.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Contacted by both Republicans/Democrats

55

36

245

Contacted only by Republicans

34

65

143

Contacted only by Democrats

74

24

141

Not contacted by either party

47

49

446

Issues and the Vote

Voters who said national issues were most important to their vote supported the Democratic candidate by 52% to 44%. If Republicans were hoping that local issues would be most important and thus help them avert a big Democratic victory, that hope was not realized. Voters who said local issues were most important to their vote supported Democratic candidates by 61% to 32%.

Republicans fared best among voters who said candidate character and experience mattered most to them, with these voters preferring the Republican candidate by nearly 2 to 1, 61% to 31%. Voters who said party affiliation was the most important factor in their voting calculus strongly supported Democratic candidates, 66% to 34%.

Democratic
candidate

Republican
candidate

Sample size

%

%

 

National issues most important to vote

52

44

392

Local/State issues most important to vote

61

32

185

Character/Experience most important to vote

31

61

232

Candidate's political party most important

66

34

132

Iraq

Much has been made this year of the fact that the Iraq war has been a major issue in voters' decisions. Gallup surveys over the past three-and-a-half years have established that support for or opposition to the war is highly partisan. Republicans tend to support the Bush administration's policies in Iraq, while Democrats oppose them.

It is therefore not surprising to find a strong correlation between likely voters' positions on the war and their congressional votes. The data reflect an extraordinary mirror image. Eighty-one percent of those who said Iraq was a mistake said they would vote for the Democratic candidate. Eighty-three percent of those who believed the Iraq war was not a mistake were planning on voting for the Republican candidate.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Mistake to send troops to Iraq

81

15

548

Not a mistake to send troops to Iraq

13

83

420

This correlation alone does not prove causation. In other words, because Democrats have opposed the war for years and Republicans have supported it, the relationship between voting and position on the war may be an artifact of these pre-existing relationships. Most Democrats vote for the Democratic candidate regardless of specific issues and most Republicans vote for the Republican candidate (as has been demonstrated elsewhere in this article). For these partisans, the Iraq war may have simply reinforced prior convictions. But for Democrats, this type of reinforcement may have been a powerful factor behind their better-than-usual enthusiasm for turning out to vote.

About one out of three individuals in the "likely voter" sample are independents. The data show that 59% of these independents said Iraq was a mistake (and we know that the overwhelming majority of those who said it was a mistake planned on voting Democratic). Sample sizes are small, but 76% of independents who said the war was a mistake said they were voting Democratic, while 75% of independents who said the war was not a mistake said they were voting Republican.

Demographics

There was a significant gender gap in the 2006 vote. Male voters were more likely to support the Republican candidate in their district, while female voters were more inclined to support the Democrat. Among male voters, 52% opted for the Republican candidate and 43% the Democratic candidate; among female voters, 59% opted for the Democrat and 36% the Republican.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Men

43

52

492

Women

59

36

509

Whites divided evenly in their vote preference, 48% to 48%, while nonwhites showed a 3-to-1 margin in favor of the Democratic candidate in their local district. In recent elections, whites have tended to vote more Republican than Democratic.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

White

48

48

853

Nonwhite

66

22

154

Older voters were more likely than younger voters to support the Democratic candidate in their congressional district, while younger voters were more divided in their choices. Half of voters aged 18 to 39 told Gallup they would vote for the Democrat in their district, while 45% would vote for the Republican. Among voters in the 40 to 54 age group, 49% preferred the Democrat and 48% the Republican. And among voters aged 55 and older, 53% opted for the Democratic candidate in their district, while 39% opted for the Republican candidate.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

18- to 39-year-olds

50

45

159

40- to 54-year-olds

49

48

304

55 years and older

53

39

538

Married voters were divided in their vote choices, with 48% opting for the Republican candidate in their district and 47% for the Democratic candidate. This is a significant departure from recent election patterns, in which married voters showed strong support for Republicans. Unmarried voters exhibited their usual support for the Democratic Party, preferring Democratic candidates over Republican candidates, 58% to 37%.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Married

47

48

641

Not married

58

37

344

Interestingly, married men and unmarried men differed in their support for the congressional candidate in their district. Married men preferred the Republican candidate (53%) rather than the Democratic candidate (42%); unmarried men were divided, with 49% choosing the Republican and 47% the Democrat. Both married and unmarried women indicated they would vote for the Democratic candidate in their district, though unmarried women were more inclined than married women to say this.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Married men

42

53

337

Unmarried men

47

49

150

Married women

53

41

304

Unmarried women

67

28

194

Voters who have children under age 18 were evenly divided in their vote for Congress -- 51% supported the Republican and 46% the Democrat. Voters without children supported the Democrat, 53% to 41%.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Have children under age 18

46

51

278

Do not have children under age 18

53

41

713

Urban, Suburban, and Rural Voting

Gallup's final pre-election poll shows that there was not a great deal of difference in vote choice by type of community in which the voter lives. A majority of 55% of voters in both urban and rural areas voted Democratic. Voters in suburban areas split their vote essentially equally between the two parties.

Democratic
candidate

Republican
candidate


Sample size

%

%

 

Urban

55

39

271

Suburban

47

48

539

Rural

55

39

191

This is a different pattern from what has been noticed in previous elections. In 2002 (a year in which the Republicans won the national popular vote), Gallup data suggested that voters in rural areas were significantly more likely than urban voters to vote for the Republican candidate.

There was a great deal of discussion this year about the potential loss of support for the Republicans among one of their most loyal constituency groups: religious whites (defined as whites who attend religious services frequently). This discussion arose after the scandal involving former Congressman Mark Foley and his relationships with House pages, and continued even in the last week of the campaign when the evangelical preacher and leader Ted Haggard (who was frequently in touch with the White House) admitted to sexual immorality in response to accusations by a male prostitute. (Haggard was fired by his church and left his post as head of the National Association of Evangelicals.)

A Gallup analysis based on data collected in early October did suggest that the Republicans had lost some of their support from religious whites.

But subsequent analysis showed that this key group had begun to return to the Republican fold by the time of Gallup's Oct. 20-22 poll. Gallup's final pre-election poll shows that any impact of campaign events on loyalty from this group was apparently short-lived. Religious whites indicated a voting preference for Republicans over Democrats, by a 60% to 36% margin. By way of contrast, whites who are not religious indicated a voting preference for Democrats over Republicans, by a 60% to 35% margin. Nonwhites planned to vote overwhelmingly for the Democratic candidate, by a 65% to 20% margin.

Survey Methods

Results are based on telephone interviews with 1,007 likely voters, aged 18 and older, conducted Nov. 2-5, 2006. For results based on the total sample, one can say with 95% confidence that the maximum margin of sampling error is ±4 percentage points.

Likely voters are identified according to a series of questions measuring current voting intentions and past voting behavior. Based on past voting history in United States midterm elections and current interest in the election, turnout is assumed to be 40% of the voting-age population.

In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.


Gallup https://news.gallup.com/poll/25399/Democrats-Election-Strength-Evident-Across-Voter-Segments.aspx
Gallup World Headquarters, 901 F Street, Washington, D.C., 20001, U.S.A
+1 202.715.3030