By Cynthia L. Cooper
Thursday, July 12, 2001
An unidentified conservative client financed a poll on welfare and birth control that used false information and biased questions. Highly publicized, the poll claimed the public wants women on welfare to be required to use birth control.
(WOMENSENEWS)--A poll on birth control and welfare purports to show that the public wants women on welfare to take birth control pills--but it tells less about the public's thinking than about how polling is used to spin the agendas of interest groups and shape public debate. In this case, it's at the expense of women.
Results from a poll conducted for an unidentified private client were sent in an e-mail press release in early May from Zogby International, an established political polling firm with offices in Washington, D.C., and Utica, N.Y. The news release said a poll showed Americans support proposed federal legislation that would require all welfare recipients to use birth control as a condition of eligibility for benefits. The results can be purchased online.
Five days after the release, Fox Television featured a segment about welfare and birth control. The Kaiser Family Foundation Reproductive Health Reports then included a synopsis of the segment, mentioning the poll, in its daily online journal.
There is no proposed legislation as stated by the poll, according to Ron Haskins, a key drafter of welfare reform legislation and currently a senior fellow at the Brookings Institution in Washington. "I never heard of it," he said in an interview.
Further, Anna Greenberg, vice president of Greenberg Quinlan Rosner Research Inc. in Washington and until recently a professor on survey techniques at Harvard University, said that respected academic researchers now decline to ask questions about fictitious legislation simply to measure people's opinions.
More troubling, say critics, are indications that the poll was not unbiased or failed to meet standards for the release of public polling data established by professional associations of pollsters. For example, the question was asked in a way that "you would expect a skewed result," said Murray Edelman, president of the American Association for Public Opinion Research, an association of 1,500 organizations.
The exact Zogby question:
"Today, some people believe that mothers on welfare continue to have children that they are financially unable to support. There is proposed federal legislation that would require all welfare recipients to use birth control as a condition of their public benefits. Do you strongly support, somewhat support, somewhat oppose, or strongly oppose requiring welfare recipients to use birth control?"
"It doesn't sound very balanced to me," said Edelman. The opening statement, he explained, gave only one perspective, instead of offering the reasons why some people might oppose the same proposal. "Usually you try to give both sides," he explained.
"I take issue with the introduction," said Greenberg. "The lack of balance is troubling."
But John Zogby, chief executive officer of Zogby International, defended his poll in an interview, saying it is not biased because people could agree or disagree with the single statement presented.
"It passed a smell test," said Zogby. "I've got a whole lot of folks here who respond to client concerns and who develop questions, and there you go," he said. "I mean, do you call General Motors and ask them how did you design the automobile and why did you design it the way you did?"
As for the identity of the client who financed the poll, Zogby declined to name the organization or persons. He declined to disclose the other questions in the poll or why it was undertaken, except to note it was done for a "fairly conservative client."
Pollsters are obligated in any public report to state who sponsored the survey, according to the Code of Professional Ethics and Practices of the American Association of Public Opinion Research. Its Standards for Minimal Disclosure states: "Good professional practice imposes the obligation upon all public opinion researchers" to include in a public release "1. Who sponsored the survey." Edelman said, "It's one of our general principles."
The sponsor is important because subtle differences in a question can influence the results. A sponsor with a particular advocacy position is a warning sign, said Greenberg.
In an article for the National Council of Public Polls, authors Sheldon Gawiser and G. Evans Witt emphasize that the sponsor is a critical piece of information. Polls, they said, "are always done for a reason--to advance a cause or gain helpful information." Knowing the sponsor explains "who thought these topics are important enough to spend money finding out what people think."
The question: Who thinks that arguing for government control of the reproductive ability of poor women is important enough to spend money on?
"I think it taps into a very emotional issue for the right wing," said Dick Morris, a seasoned political consultant who has advised conservative, moderate and liberal campaigns. "They are trying to gin up a hot button issue and to make welfare alive as an issue in the aftermath of welfare reform, which took it off the table. It combines the Moral Majority position with the economic right wing position and is politically effective," he said in an e-mail interview.
The forces that produce a private poll for an unnamed organization and a public release of an unbalanced question by a polling organization might spell trouble for poor women.
The 1996 welfare law, for example, permitted states to exclude children born to mothers receiving assistance, and some states, such as New Jersey, now refuse aid even to infants born as the result of rape or incest. In fact, the notion of controlling poor women's reproductive lives remains a perennially popular topic, despite the fact that the average welfare family is composed of an adult and one or two children. The roots of this impulse go back at least as far as the early 20th century.
"There's a tremendous growth in opposition to welfare programs," said Edith Klein, author of "Gender Politics" and the head of a New York research firm on social issues. The American people are quite willing to pursue punitive policies toward women on welfare, she explained.
"We should be very disturbed by the notion that you can be denied basic necessities because you have exercised the right to have a child," said Lynn Paltrow, executive director of the National Advocates for Pregnant Women in New York.
Vigorous opposition should be mounted before the atmosphere is poisoned further, said Gwendolyn Mink, a professor of political science who will be joining the faculty of Smith College. She is the author of "Welfare's End."
Next year Congress is expected to reauthorize the 1996 welfare laws that made dramatic changes in the nature of assistance available to single mothers.
"We saw the way that the public was seduced into accepting and trading on and intensifying myths about poor people in the five to 10 years leading up to welfare reform in 1996," Mink said.
"I certainly like to think after due deliberation," said Mink, "the public would choose that this kind of intervention in the intimate lives of women just because they are poor is so profoundly anti-democratic that they have no place in the American policy spectrum."
The Zogby poll release said it surveyed 1,000 likely voters nationwide. It reported that 62 percent said they would support proposed legislation, 33 percent said they would oppose it and 5 percent said they were not sure.
More specifically, it said 46.5 percent would strongly support a requirement, 15.4 percent somewhat support it, 21.9 percent strongly oppose it, 11.4 percent somewhat oppose it and 4.9 percent said they were not sure.
Cynthia L. Cooper is a free-lance journalist in New York who writes frequently on reproductive rights.
American Association for Public Opinion Research:
National Council on Public Polls: