Reading Between the Lines on Poll Results A poll out this week found that 8 out of 10 Americans would be very comfortable with an African-American or woman president. But are people completely honest when answering pollsters' questions about race and gender?
NPR logo

Reading Between the Lines on Poll Results

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Reading Between the Lines on Poll Results

Reading Between the Lines on Poll Results

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


Senators Hillary Clinton and Barack Obama are flagrantly weighing whether to run for the Democratic nomination for president. Senator Clinton is a woman. Senator Obama is the son of a white woman from Kansas, but also a black man from Kenya. If any of that's a problem, even now, all these years after the civil rights marches and the advent of feminism, the electorate might just be white-lying about it. This week a Wall Street Journal/NBC News poll said that eight out of 10 Americans would be very comfortable with an African-American or a women in the White House. But historically Americans may say one thing to a pollster, then do something else in the voting booth.

Celinda Lake is a Democratic pollster who's plumbed public opinion for women and black candidates, and joins us in our studios. Ms. Lake, thanks very much for being with us.

Ms. CELINDA LAKE (President, Lake Research Partners): Thanks for having me.

SIMON: And what - we sort of had a play on words there. What is what they call the white lie effect?

Ms. LAKE: Oh, I was laughing at that play on words and you're right. Voters do lie and they often tell the interviewer what they think is the politically correct thing to say and what they think they may want to hear. We found a couple of things. We found, for example, that women who are talking to women and men who were talking to men are much less likely to lie about whether they would support a woman candidate. Whites talking to white - people they think are white interviewers, often were honest; and people of color who were talking to people of color often were honest.

And then the other thing we find is that it's very hard to poll now because people are reading in partisanship. So where Republicans and Democrats used to be about equally supportive of a woman for president, when you ask that now, Republicans are less supportive because they assume you mean Hillary Clinton. And so there are lots of factors that go into those answers.

SIMON: Are some of these views dated? I mean, for example, we're going to see a new Congress, 71 women, 43 African-American members. Is it as important as it used to be?

Ms. LAKE: It's not as important as it used to be. And there have been very, very dramatic changes. And 30 years ago you would have had only less than half the people ready to elect either an African-American or a woman president, but they still affect people's attitudes.

SIMON: Can you share a trade secret with us as to how you craft a question?

Ms. LAKE: Well, first of all, I'd say, particularly when you're trying to measure sexism or racism, it is very, very difficult, because people are leery of sharing those feelings. Secondly, I would say we ask - a trade secret is we never rely on one question. We ask more than one question. And we would ask questions that may be indirect measures. So for example, do you think the father should be master of the house? And people who strongly agree with that are probably a little less likely to vote for a woman candidate. And then we would ask how comfortable are you, and we would ask how comfortable do you think your friends and neighbors are.

People will attribute great comfort to themselves, but if you ask them, do you think people like you or do you think your friends and neighbors would be comfortable, those numbers that are in the 80 percent range will drop down to like the 60 percent range or lower.

SIMON: Do people think differently about the presidency than they do about a legislative office, or for that matter governor and mayor?

Ms. LAKE: People think very differently about executive office versus legislative office. Executive office - and obviously the president at the top of that executive office - has been the harder office to break into.

SIMON: I have to ask you about some other results in the poll. That Wall Street Journal/NBC poll says that 53 percent of the respondents would be uncomfortable with a Mormon as president.

Ms. LAKE: Yeah. People actually, traditionally, have been very uncomfortable with new religions.

SIMON: A new religion being a religion they haven't encountered before.

Ms. LAKE: That hasn't been a president before.

SIMON: Is it more socially acceptable for people to say that they have doubts about a Mormon in office than about an African-American or a woman?

Ms. LAKE: Yes, because people think that religion relates to your belief system, and that your beliefs are legitimate grounds for presidency, where your gender or your race is not.

SIMON: It has to be asked, since we obviously are specifically talking about Senators Clinton and Obama; they were both elected and then reelected by huge majorities in their states. Are they beyond these kinds of taboos?

Ms. LAKE: I think that they do have a certain celebrity status that takes them beyond the stereotypes.

SIMON: But all these competing pressures and just the natural complexity of human thought and reservations that people have about revealing themselves to strangers on the telephone - how confident are you in your polling data?

Ms. LAKE: Well, we put in a lot of controls. And the absolute answer is there is still about a five to ten percent lying factor.

SIMON: Celinda Lake, president of Lake Research Partners, a Democratic polling firm, thanks very much.

Ms. LAKE: Thank you for having me.

Copyright © 2006 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.