Aug 11, 2006  |  By: ELIZABETH A. PERRY  | COMMENTS     

Two recent polls seeking to gauge voter support for a proposed constitutional amendment banning gay marriage in Virginia produced different results, raising questions about whether the complexity and length of the ballot language will confuse voters in November.

A poll conducted by Mason-Dixon Polling & Research found that 56 percent of respondents support the marriage amendment with 38 percent opposed — an indication of strong statewide support for the ban.

The Commonwealth Coalition, a group opposed to the amendment effort, released its own poll that revealed that just 45 percent support the measure, with 40 percent opposed and 14 percent undecided. The group touted the results in a press release, declaring “Amendment ballot support under 50%.” 

The 11-point discrepancy in support of the measure between the two polls may have something to do with the text of the question asked of poll respondents, according to the Commonwealth Coalition.

The Mason-Dixon Poll was conducted using a sample of 625 registered voters and yielded a majority response in favor of the amendment. Respondents answered a question about the first sentence of the amendment only:

“Shall the Constitution of Virginia be amended to state that only a union between one man and one woman may be a marriage?”

Meanwhile, the Commonwealth Coalition poll showed that a 54 percent majority would vote “no” or were undecided. Those results were based on two telephone polls conducted by Fabrizio, McLaughlin & Associates, and Schapiro Research Group. Each poll surveyed 800 voters, culled from voter registration records statewide.

The Commonwealth Coalition poll tested the exact wording of the Virginia marriage amendment, as it will appear on the November ballot:

“Shall the Constitution of Virginia be amended to state ‘That only a union between one man and one woman may be a marriage valid in or recognized by this Commonwealth and its political subdivisions. This Commonwealth and its political subdivisions shall not create or recognize a legal status for relationships of unmarried individuals that intends to approximate the design, qualities, significance, or effects of marriage. Nor shall this Commonwealth or its political subdivisions create or recognize another union, partnership, or other legal status to which is assigned the rights, benefits, obligations, qualities, or effects of marriage’? If the election were held today, would you vote yes in favor of the amendment, no in opposition to the amendment, or are you undecided?”

Claire Guthrie Gastanaga, campaign manager of the Commonwealth Coalition, said the difference in survey results indicates that when voters are aware of the amendment text in its entirety, they are more likely to oppose it.

“The difference in these results comes simply from voters’ common sense reading of the fine print in this ill-considered proposal,” said Gastanaga. “Virginians who actually read the whole amendment see that it opens a Pandora’s box of unintended consequences, and they don’t want to do that.”

John Stahura, professor of sociology and director of the Perdue University Social Research Institute, reviews polls and surveys for a number of professional journals in the field of sociology. He examined the Commonwealth Coalition’s poll and critiqued its methodology and results. Many factors can impact the accuracy of a poll, including the language used in the ballot question and even the length of the question asked, he said.

Stahura said a number of questions about registration, voter participation in past presidential and local elections, and interest in upcoming elections are typically asked during phone surveys as part of the methodology. He said the length of the statement that was read to the respondents could have impacted their responses. He said that when people hear a statement read to them over the phone they could be easily confused because of the length and wording.