Here Are The Results of SurveyUSA Election Poll #10367

Geography Surveyed: Virginia
Data Collected: 09/26/2006 - 09/28/2006
Release Date: 09/29/2006 11:40 AM ET
Sponsors: WDBJ-TV Roanoke, WUSA-TV Washington, DC

VA Continues to Shrug-Off Comments About Allen's Past: In an election for the United States Senate in Virginia today, 9/29/06, incumbent Republican George Allen maintains an advantage over Democrat challenger James Webb, according to a SurveyUSA poll conducted exclusively for WUSA-TV Washington DC and WDBJ-TV Roanoke. In a 3-day rolling average of interviews conducted Tuesday 9/26/06 through Thursday 9/28/06, Allen leads Webb, 50% to 44%. On 9/27/06, SurveyUSA released poll results based on interviews conducted Sunday 9/24/06 through Tuesday 9/26/06, which showed Allen ahead 49% to 44%. Day-to-day volatility remains very high. The 6-point Allen lead in today's results may change - but it appears as likely to change in one direction as the other. Today, Allen gets 88% of Republican votes. Webb gets 86% of Democrat votes. Independents favor Allen by 7 points. Allen leads by 18 points among men, trails by 6 points among women, a 24-point Gender Gap. Allen leads 89% to 8% among conservatives. Webb leads 82% to 13% among liberals. Moderates favor Webb by 2:1. In five tracking polls conducted by SurveyUSA since June, Webb has not polled above 45%; Allen has not polled below 48%. The election is in 39 days, on 11/7/06. The Republicans have a 10-seat majority in the U.S. Senate. Allen's seat had been considered safely Republican, and Allen himself had been widely considered to be a Republican candidate for President of the United States in 2008. Allen, a former Virginia Governor, was first elected to the U.S. Senate in 2000.

Question 1, With Revised Language, Passes 3:2, But Half of Voters Not Certain Yet How They Will Vote: In an election in Virginia today, 9/29/06, 52% of Likely Voters say they are not yet certain how they will vote on Question 1, which redefines marriage. Of those who are certain how they will vote on Question 1, 29% are certain to vote Yes, 19% are certain to vote No. Among those who are certain, the measure passes 3:2. The exact language that SurveyUSA read to respondents for today's poll release appears in the table below. This is a deliberate change from language that SurveyUSA read to respondents in a poll released 9/13/06. In SurveyUSA's 9/13/06 release, Question 1 passed 7:3. A discussion of the pros and cons of different pollster question wording appears in SurveyUSA's 9/13/06 release. Both SurveyUSA and Mason Dixon Opinion Research have now modified the language they are using to ask about Question 1 in Virginia. Mason Dixon went from reading a summary of the ballot question in one release to reading the full text in a second release. SurveyUSA has gone from reading a summary of the ballot question to, today, reading no summary at all. In all 4 cases - Mason Dixon original; Mason Dixon revised; SurveyUSA original; SurveyUSA revised - the measure passes. Data from SurveyUSA's 9/13/06 poll is not intended to be directly comparable with today's 9/29/06 results. It is wrong to infer that the number of "undecided" Virginians has increased, and it is wrong to infer that support for the measure has either increased or decreased since SurveyUSA's 9/13/06 poll. Such conclusions cannot be drawn. The two different approaches SurveyUSA has used to ask about Question 1 are designed to measure different things. Today's release shows that 52% of Virginians do not yet have a clear understanding of what Question 1 will do to the Virginia Bill of Rights, and may be persuadable on this issue. Of those who have made up their mind for certain, Republicans support the measure, Democrats and Independents oppose it. The older the voter, the more likely he or she is to vote "Yes." As the forces who support and the forces who oppose Question 1 spend money to raise the awareness of Virginia voters about just what the implications of Question 1 may be, SurveyUSA expects the number of "certain" voters to increase, the number of "not yet certain" voters to decrease.

Filtering: 1,200 Virginia adults were interviewed 9/26/06 - 9/28/06. Of them, 1,003 were registered to vote. Of them, 612 were judged to be "likely" voters. Crosstabs reflect Likely Voters.