GALLUP NEWS SERVICE
PRINCETON, NJ -- Thirty-three U.S. Senate seats are up for election this year, although most analysts agree that fewer than 10 are truly competitive. Polls -- including the latest six USA Today/Gallup polls reported elsewhere on galluppoll.com today -- provide classic horse race figures that help pinpoint where the races stand at the time of the poll. But understanding what is behind voters' decisions in key states to support one or the other candidate is often more difficult to ascertain.
The recent USA Today/Gallup Senate polls included a question designed to help remedy that situation. After poll respondents indicated for whom they would vote, they were asked: "Why would you say you are voting for this candidate?" Gallup interviewers recorded the results verbatim, and they were later coded into categories that grouped similar responses.
The overall results for the six states -- Missouri, Rhode Island, Virginia, Tennessee, Maryland, and New Jersey -- are presented at the bottom of this article. For each state, the tables display the overall responses for all registered voters who made a candidate choice, and then the results split out among registered voters who chose each candidate.
There are wide varieties of ways in which these data can be approached, and as a result a wide variety of conclusions that can be reached. Gallup Poll editors developed several important points about the reasons behind voters' choices for the Senate.
But in this situation, we thought some readers might have different approaches or insights, and that as a result it would be of interest to allow readers to suggest their own thoughts on these data, and to share those with us here at The Gallup Poll by responding to Talk to the Editors.
We will read all suggested responses, and post the most telling and insightful here on galluppoll.com next Wednesday.
Here are several conclusions from Gallup Poll editors based on voters' self-reported reasons for supporting Senate candidates in six key battleground states:
Party-Line Voting
The top category of explanations for voting for a particular candidate in each state is party-line voting. The percentage of registered voters saying that they are voting for a given candidate because they usually vote for that party's nominee is highest in Maryland (36%), where 41% of Ben Cardin voters say they are voting for him because he is a Democrat and they usually vote along party lines.
Of some interest is that in every state, party-line voting is higher for the Democratic candidate than for the Republican candidate.
Incumbent Referendum
It's not surprising to find that a segment of voters in the four states that have an incumbent running for re-election explain their votes as being either a positive referendum on the incumbent's performance or a negative referendum with the concomitant desire to put someone new in office.
For example, 37% of Lincoln Chafee voters in Rhode Island explain their vote by saying that Chafee is doing a good job and deserves re-election. At the same time, 28% of Sheldon Whitehouse voters in Rhode Island say they are voting because of the need for a change, or because Chafee is doing a poor job.
In Virginia, 29% of George Allen voters cite a positive referendum on his performance as senator, while 32% of Jim Webb voters say they are voting against Allen.
In New Jersey, where incumbent Bob Menendez has only served for a few months, 17% of Menendez voters cite the incumbency factor, while 11% of Tom Kean Jr., voters say they are voting because Menendez has done a poor job and needs to be replaced.
In Missouri, 22% of Jim Talent voters say it is because of his performance in office, while 23% of Claire McCaskill voters say they are voting for her because they disapprove of Talent's performance or are voting for a change in office.
Issues
Only a small minority of voters mention specific issues in explaining their vote for Senate. A much larger percentage mention "issues" more generally, stating that they are voting for their candidate because they agree with his or her positions on the issues, or because they agree with the candidate's platform or views.
This is true even in a state like Missouri, where the issue of stem cell research funding has been one of the consuming issues of the campaign. Two percent of Talent's voters (Talent opposes a high-profile constitutional amendment in Missouri that would allow early term stem cell research) say they are voting for him on this basis, and 2% of those voting for McCaskill (who supports the amendment) say it is on the basis of this issue.
To be sure, in each state, a variety of issues are mentioned by small percentages of the voters. These include abortion, healthcare, taxes, education, immigration, the economy, and government corruption. But even when the percentage of voters who mention any of these specific issues are combined in a given state, the total is quite low, suggesting that there are relatively few "single-issue voters."
Instead, the data suggest that the effect of issues on the Senate vote is more generalized or tied together in voters' minds, becoming part of the overall impression of support or opposition they have for a candidate.
Missing: Bush and Iraq
There has been a great deal of discussion this year about the effect of the national political environment on individual House and Senate races. In particular, many observers have assumed that a high-profile president and a high-profile war in Iraq would become pivot issues that would affect races above and beyond the more typical local concerns and the characteristics and backgrounds of the specific candidates involved.
The open-ended data being reviewed here do not give explicit support to that hypothesis. Very few voters mention either President George W. Bush or the war in Iraq as a reason why they are voting for a specific Senate candidate. No more than 3% mention the war in Iraq in any of the six states, and only in one state -- New Jersey -- do as many as 5% mention that their vote is directly related to support or opposition to Bush.
It is certainly true that over-time analysis shows that the outcome of midterm elections is related to the level of the president's popularity. An unpopular president is likely to see his party lose seats and vice versa. Thus, it may be possible that the war in Iraq and the president are affecting these races more indirectly. For example, explanations for the vote that focus on liking a candidate's platform/position on the issues or liking the candidate's party affiliation could be indirectly based on position on the war or the president.
In Virginia, for example, the Democratic candidate, Webb, has explicitly criticized the Bush administration and the war in Iraq. Only 4% of Webb voters, however, say that their vote for Webb is based on his position on Iraq, and only another 2% mention Bush. But 28% support Webb because they say they usually vote Democratic, 12% say they are voting for a change from the incumbent Allen, and 9% say they like Webb's positions on the issues. All these could, in turn, be based on a dislike for the administration or the Republicans' position on the Iraq war.
Following are the detailed results for each state:
Results for Maryland |
All |
Cardin |
Steele |
|
RV |
RV |
RV |
||
% |
% |
% |
||
Usually support that party/Voting along party lines |
36 |
41 |
29 |
|
Like his platform/views/stand on issues |
17 |
15 |
22 |
|
Like him |
13 |
11 |
16 |
|
Better qualified/Past experience/Good track record |
11 |
12 |
11 |
|
Voting for change |
6 |
5 |
6 |
|
Don't like opponent |
5 |
6 |
2 |
|
Honest/Moral/Has Integrity |
3 |
2 |
5 |
|
Prefer the more conservative candidate |
2 |
1 |
4 |
|
Better for/Sympathetic to the common man/average American |
3 |
2 |
4 |
|
Moral values issues |
2 |
1 |
3 |
|
Iraq war issue |
1 |
2 |
1 |
|
Abortion issue |
2 |
2 |
2 |
|
Education issue |
1 |
1 |
1 |
|
Tax issue |
1 |
* |
2 |
|
Opponent dishonest/Immoral/Lacks integrity |
1 |
1 |
* |
|
Intelligent |
1 |
* |
1 |
|
Good leader/Leadership qualities |
* |
1 |
* |
|
Healthcare issue |
* |
1 |
-- |
|
Prefer the more liberal candidate |
* |
1 |
* |
|
Vote to oppose President Bush |
2 |
3 |
-- |
|
Dislike opponent's platform/views/stand on issues |
* |
1 |
* |
|
Opponent running bad campaign |
1 |
1 |
1 |
|
Lesser of two evils |
1 |
1 |
* |
|
Terrorism/National security issue |
* |
* |
* |
|
The economy |
* |
* |
-- |
|
Running a good campaign |
1 |
1 |
2 |
|
Vote to support President Bush |
1 |
* |
1 |
|
Immigration issue |
* |
-- |
* |
|
Stem cell research issue |
* |
* |
1 |
|
|
|
|
||
Other |
3 |
3 |
2 |
|
None |
3 |
3 |
2 |
|
No opinion |
2 |
2 |
1 |
|
|
McCaskill |
|
|
RV |
RV |
RV |
||
% |
% |
% |
||
Usually support that party/Voting along party lines |
24 |
28 |
20 |
|
Talent doing a good job/Satisfied with performance |
14 |
8 |
22 |
|
Like his platform/views/stand on issues |
13 |
12 |
14 |
|
Talent doing a poor job/needs to be replaced |
8 |
13 |
1 |
|
Don't like opponent |
7 |
4 |
10 |
|
Voting for change |
5 |
10 |
* |
|
Like him/her |
7 |
9 |
5 |
|
Honest/Moral/Has Integrity |
6 |
4 |
9 |
|
Better qualified/Past experience/Good track record |
4 |
6 |
2 |
|
Opponent dishonest/Immoral/Lacks integrity |
3 |
2 |
4 |
|
Prefer the more conservative candidate |
3 |
* |
5 |
|
Dislike opponent's platform/views/stand on issues |
2 |
1 |
3 |
|
Abortion issue |
2 |
1 |
3 |
|
Stem cell research issue |
2 |
2 |
2 |
|
Moral values issues |
2 |
-- |
5 |
|
Opponent running a bad campaign |
2 |
1 |
2 |
|
Healthcare issue |
2 |
4 |
-- |
|
Better for/Sympathetic to the common man/average American |
3 |
4 |
2 |
|
Vote to support President Bush |
1 |
1 |
1 |
|
Tax issue |
1 |
1 |
1 |
|
Education issue |
1 |
1 |
-- |
|
Immigration issue |
* |
* |
1 |
|
Vote to oppose President Bush |
1 |
2 |
-- |
|
Iraq war issue |
1 |
1 |
1 |
|
Good leader/leadership qualities |
* |
* |
* |
|
Intelligent |
* |
* |
* |
|
The economy |
1 |
* |
1 |
|
Terrorism/National security issue |
* |
-- |
1 |
|
Prefer the more liberal candidate |
1 |
* |
* |
|
Lesser of two evils |
1 |
1 |
1 |
|
Running a good campaign |
1 |
1 |
1 |
|
|
|
|
||
Other |
3 |
3 |
2 |
|
No reason in particular |
2 |
1 |
3 |
|
No opinion |
4 |
3 |
5 |
|
|
Menendez |
|
|
RV |
RV |
RV |
||
% |
% |
% |
||
Usually support that party/Voting along party lines |
30 |
38 |
24 |
|
Like his platform/views/stand on issues |
11 |
7 |
16 |
|
Menendez doing a good job/satisfied with performance |
11 |
17 |
5 |
|
Like him |
9 |
11 |
7 |
|
Opponent dishonest/Immoral/Lacks integrity |
5 |
3 |
8 |
|
Menendez doing a poor job/needs to be replaced |
5 |
1 |
9 |
|
Vote to oppose President Bush |
4 |
7 |
* |
|
Voting for change |
4 |
7 |
2 |
|
Don't like opponent |
5 |
4 |
3 |
|
Iraq war issue |
3 |
5 |
1 |
|
Honest/Moral/Has Integrity |
3 |
1 |
5 |
|
Better qualified/past experience/good track record |
3 |
3 |
3 |
|
Prefer the more conservative candidate |
3 |
1 |
5 |
|
Better for/Sympathetic to the common man/average American |
2 |
4 |
1 |
|
His family/background |
3 |
1 |
6 |
|
Prefer the more liberal candidate |
2 |
1 |
2 |
|
Dislike opponent's platform/views/stand on issues |
1 |
1 |
1 |
|
Vote to support President Bush |
1 |
-- |
2 |
|
Moral values issues |
1 |
* |
3 |
|
Tax issue |
1 |
1 |
1 |
|
Good leader/leadership qualities |
1 |
1 |
1 |
|
Running a good campaign |
2 |
2 |
1 |
|
Lesser of two evils |
* |
-- |
1 |
|
Immigration issue |
* |
* |
-- |
|
Government corruption issue |
* |
-- |
* |
|
Stem cell research issue |
* |
* |
-- |
|
Opponent running bad campaign |
* |
* |
* |
|
Health care issue |
1 |
1 |
-- |
|
Terrorism/National security issue |
* |
-- |
* |
|
Abortion issue |
1 |
* |
-- |
|
Intelligent |
* |
* |
* |
|
Education issue |
* |
* |
-- |
|
|
|
|
||
Other |
3 |
3 |
2 |
|
No reason in particular |
3 |
2 |
3 |
|
No opinion |
6 |
3 |
8 |
|
|
Whitehouse |
|
|
RV |
RV |
RV |
||
% |
% |
% |
||
Chafee doing a good job/satisfied with performance |
19 |
8 |
37 |
|
Usually support that party/Voting along party lines |
23 |
31 |
14 |
|
Like his platform/views/stand on issues |
10 |
11 |
10 |
|
Like him |
11 |
11 |
12 |
|
Chafee doing a poor job/needs to be replaced |
9 |
13 |
3 |
|
Voting for change |
9 |
15 |
3 |
|
Honest/Moral/Has Integrity |
5 |
3 |
7 |
|
Lesser of two evils |
3 |
1 |
5 |
|
Iraq war issue |
2 |
4 |
1 |
|
Don't like opponent |
4 |
2 |
7 |
|
Prefer the more conservative candidate |
2 |
* |
4 |
|
Better for/Sympathetic to the common man/average American |
1 |
1 |
2 |
|
Vote to oppose President Bush |
1 |
1 |
2 |
|
Better qualified/Past experience/Track record |
1 |
1 |
1 |
|
His family/background |
2 |
* |
4 |
|
The economy |
* |
1 |
-- |
|
Opponent dishonest/Immoral/Lacks integrity |
1 |
1 |
* |
|
Tax issue |
1 |
1 |
* |
|
Prefer the more liberal candidate |
* |
* |
1 |
|
Vote to support President Bush |
* |
* |
-- |
|
Health care issue |
* |
* |
-- |
|
Intelligent |
* |
1 |
-- |
|
Moral values issues |
1 |
1 |
* |
|
Opponent running bad campaign |
1 |
2 |
1 |
|
Immigration issue |
* |
1 |
-- |
|
Education issue |
1 |
1 |
-- |
|
Running a good campaign |
1 |
* |
1 |
|
Abortion issue |
* |
-- |
* |
|
Stem cell research issue |
* |
-- |
* |
|
Government corruption issue |
* |
* |
-- |
|
|
|
|
||
Other |
5 |
4 |
6 |
|
No reason in particular |
5 |
5 |
1 |
|
No opinion |
3 |
3 |
2 |
Results for Tennessee |
All |
Ford |
Corker |
|
RV |
RV |
RV |
||
% |
% |
% |
||
Like his platform/views/stand on issues |
19 |
19 |
19 |
|
Usually support that party/Voting along party lines |
25 |
30 |
18 |
|
Better qualified/Past experience/Good track record |
12 |
13 |
10 |
|
Like him |
7 |
9 |
5 |
|
Honest/Moral/Has Integrity |
7 |
8 |
6 |
|
Don't like opponent |
6 |
4 |
8 |
|
Prefer the more conservative candidate |
7 |
1 |
14 |
|
Moral values issues |
4 |
2 |
6 |
|
Opponent dishonest/Immoral/Lacks integrity |
5 |
4 |
5 |
|
Voting for change |
3 |
3 |
3 |
|
Better for/Sympathetic to the common man/average American |
5 |
7 |
3 |
|
Immigration issue |
3 |
3 |
3 |
|
Opponent running a bad campaign |
3 |
4 |
2 |
|
Vote to oppose President Bush |
2 |
3 |
* |
|
Abortion issue |
2 |
* |
5 |
|
Lesser of two evils |
2 |
2 |
2 |
|
Dislike opponent's platform/views/stand on issues |
1 |
1 |
1 |
|
His Family/background |
2 |
1 |
4 |
|
Vote to support President Bush |
1 |
* |
1 |
|
Education issue |
1 |
1 |
-- |
|
Running a good campaign |
2 |
2 |
1 |
|
Prefer the more liberal candidate |
* |
* |
* |
|
Iraq war issue |
1 |
1 |
* |
|
Intelligent |
1 |
1 |
* |
|
The economy |
* |
* |
-- |
|
Good leader/Leadership qualities |
* |
1 |
-- |
|
Government corruption issue |
* |
* |
-- |
|
Tax issue |
* |
-- |
* |
|
|
|
|
||
Other |
2 |
2 |
2 |
|
No reason in particular |
2 |
2 |
3 |
|
No opinion |
2 |
1 |
3 |
Results for Virginia |
All |
Webb |
Allen |
|
RV |
RV |
RV |
||
% |
% |
% |
||
Usually support that party/Voting along party lines |
25 |
28 |
22 |
|
Allen doing a good job/satisfied with performance |
18 |
6 |
29 |
|
Allen doing a poor job/needs to be replaced |
9 |
20 |
* |
|
Like his platform/views/stand on issues |
13 |
9 |
16 |
|
Like him |
8 |
6 |
11 |
|
Voting for change |
7 |
12 |
1 |
|
Don't like opponent |
5 |
4 |
6 |
|
Prefer the more conservative candidate |
4 |
2 |
6 |
|
Honest/Moral/Has Integrity |
3 |
1 |
5 |
|
Iraq war issue |
3 |
4 |
2 |
|
Opponent running bad campaign |
2 |
2 |
2 |
|
Better qualified/Past experience/Good track record |
2 |
1 |
2 |
|
Opponent dishonest/Immoral/Lacks integrity |
2 |
2 |
1 |
|
Moral values issues |
1 |
1 |
1 |
|
Dislike opponent's platform/views/stand on issues |
1 |
1 |
* |
|
Running a good campaign |
2 |
2 |
2 |
|
Abortion issue |
1 |
* |
2 |
|
Vote to oppose President Bush |
1 |
2 |
-- |
|
Vote to support President Bush |
1 |
2 |
* |
|
Lesser of two evils |
1 |
1 |
* |
|
Prefer the more liberal candidate |
1 |
3 |
-- |
|
Terrorism/National security issue |
1 |
-- |
1 |
|
Education issue |
* |
* |
* |
|
Better for/Sympathetic to the common man/average American |
* |
* |
-- |
|
Tax issue |
* |
* |
* |
|
Government corruption issue |
* |
-- |
* |
|
|
|
|
||
Other |
3 |
3 |
2 |
|
No reason in particular |
3 |
3 |
3 |
|
No opinion |
3 |
2 |
2 |
Survey Methods
Results are based on telephone interviews with 905 registered voters in Maryland, aged 18 and older, conducted Sept. 27-Oct. 1, 2006. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.
Results are based on telephone interviews with 911 registered voters in Missouri, aged 18 and older, conducted Sept. 27-Oct. 1, 2006. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.
Results are based on telephone interviews with 890 registered voters in New Jersey, aged 18 and older, conducted Sept. 27-Oct. 1, 2006. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.
Results are based on telephone interviews with 928 registered voters in Rhode Island, aged 18 and older, conducted Sept. 27-Oct. 1, 2006. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.
Results are based on telephone interviews with 891 registered voters in Tennessee, aged 18 and older, conducted Sept. 27-Oct. 1, 2006. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.
Results are based on telephone interviews with 890 registered voters in Virginia, aged 18 and older, conducted Sept. 27-Oct. 1, 2006. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.
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.