Polling on Terri Schiavo
The Terri Schiavo situation continues moving toward its apparent end. Doctors indicate the Florida woman could die before the week is out, and conservative activists have taken their fight to Washington, seeking help to have Schiavo's feeding tube reinserted.
Poll data appear to be fairly consistent on this issue. Every poll I've seen on the topic shows Americans believe it was appropriate for Schiavo's feeding tube to be removed.
But these poll results have not gone unchallenged. Republicans and conservatives have attacked polling on the Schiavo situation, alleging that pollsters' questions are biased or misleading.
This type of poll-bashing is not a new phenomenon, of course. Pollsters got used to having their methods criticized last fall when Democrats took pollsters to task for alleged methodological flaws in surveys showing George W. Bush ahead of John Kerry in the presidential race.
Here are some recent examples of people criticizing polls about Schiavo:
- The Rev. Jerry Falwell on CNN, referring to a TIME magazine poll showing majority support for allowing Terri Schiavo to die: "… the results of the poll were derived as TIME magazine wanted them to be by misinforming the people when they asked the question."
- Rick Warren, author of The Purpose-Driven Life, on Larry King Live: "Yes, you know, Larry, the first place -- any poll, you can make a poll say anything you want by the way you frame the question, you know that. I think if people know the real story about Terri -- they think that she's on life support. She's not. They think she's brain dead, she's not."
- Terri Schiavo's brother, Bobby Schindler, on CNN: "Well, from the polls that I saw, the questions are horrible. I saw one poll that said, you know, would you want to live in this condition? And even I would answer no to that. Nobody would choose to live as a disabled person. But the answer isn't to starve them to death. And Terri is simply disabled. She has a family standing by, waiting to take care of her. And for the life of me, I don't understand why the courts are so adamant on having her killed."
- Commentator Fred Barnes on Fox News: "... I want to raise another issue. … there was this ABC News poll that was worded in a particular way … I quote it … 'Terri Schiavo suffered brain damage and has been on life support for 15 years. Doctors say she has no consciousness and her condition is irreversible. Do you support or oppose the decision to remove Schiavo's feeding tube?' unquote. Well, when phrased that way, 63 percent supported the decision to remove the tube. But that's not an accurate poll … that's a dishonest poll."
How do pollsters respond to these criticisms? Here are some key points I would make.
First, I strongly believe that the people who run these various polls are not in any way attempting to bias the results or achieve some preordained result. The people in charge of the ABC, CBS, TIME, and Fox polls are all professionals doing their best to accurately measure public opinion. (And of course that's true for us at Gallup.)
At the same time, it is absolutely true that question wording can affect responses to poll questions. Each of the critics cited above is correct to call attention to the importance of question wording when interpreting poll results. I have a long chapter on the issue in my book, Polling Matters, and have commented on the impact of question wording in this column numerous times.
All of the questions I've seen about Schiavo have described the situation to respondents in some detail. In addition to the ABC News poll described above, Gallup, CBS News, TIME, and Fox News/Opinion Dynamics have asked questions involving such phrases as "persistent vegetative state" and "coma."
In the broadest sense, the way these questions are worded could have affected the responses. For example, someone who disagrees that Schiavo is in a persistent vegetative state could certainly argue that it is not appropriate to describe her to respondents in that manner.
But I believe that most of these questions, including our own here at Gallup, reflect the most commonly agreed-upon understanding of Schiavo's situation at the time the poll questions were asked.
Also, given the case's high degree of visibility, I think question wording is less likely to affect responses than in other situations in which respondents don't know much about the subject of the question. Gallup research shows most people have strong feelings about the Schiavo case, and it is probable that just hearing the words "Terri Schiavo" is enough to provoke a response in most cases.
Finally, a good deal of data that predates the Schiavo situation show Americans favor the general idea of patients being allowed to die in certain end-of-life circumstances, and Americans are personally worried about having to live in a vegetative state. A 1997 Gallup Poll showed concern over being in such a condition was the No. 1 thing people worried about when given a long list of issues relating to death. Polling in the current situation has also shown that the majority of Americans believe they would want to die if they were in a situation similar to Schiavo's, and they would argue for the feeding tube to be removed for a spouse or child.
Schiavo and Politics
The long-term political implications of the involvement of Congress, the president, and Florida Gov. Jeb Bush in the Schiavo case won't be known for some time now. The next congressional and senatorial elections are a year and a half away, and neither President Bush nor his brother Jeb can legally run for re-election to their offices.
Still, American public opinion at the moment on this issue is clear: They don't want the government to be involved in the Schiavo case or others like it.
A March 22-24 TIME poll is representative of the findings of every poll I have seen on this issue. Seventy-five percent of Americans say that it was not right for Congress to intervene in this matter, and 70% say it was not right for President Bush to intervene. Perhaps just as importantly, both the TIME poll and a CBS news poll released last week indicate a large majority of Americans believe politics, not values or principles, was the motivation for Congress' and the president's intervention in the Schiavo case.
Bush's Job Approval
The Schiavo situation is coincident with a drop in the public's approval of the way George W. Bush is handling his job as president, down from 52% to 45% in the CNN/USA Today/Gallup poll, down from 53% to 48% in the aforementioned TIME poll, and down from 49% to 43% in the CBS poll.
The last two Gallup Polls were conducted before and after the point at which Bush and Congress intervened in the Schiavo situation. Gallup's Friday through Sunday poll (conducted March 18-20) showed Bush approval at 52%, while Gallup's Monday through Wednesday poll (March 21-23) showed a drop to 45%. (Bush signed the Schiavo bill into law early in the morning of March 21.) Therefore I think it is reasonable to assume the Schiavo case had something to do with the drop in Bush approval. (TIME's polls showing a five-point drop were conducted during a similar time frame, March 15-17 and March 22-24.)
Still, I can't rule out the hypothesis that other factors were involved in bringing down the president's job rating. One of the most important factors is the economy. Gallup has measured a continuing drift downward in Americans' willingness to say the economy is getting better, and this past week's poll shows the most pessimistic rating on this dimension in two years: 33% say the economy is getting better while 59% say it is getting worse.
Ratings of current economic conditions have also become more negative, as have views on whether now is a good time to be looking for a quality job. Investors' attitudes, as measured in the March UBS/Gallup Index of Investor Optimism poll, dropped from an index score of 82 in February to 74 in March, the lowest reading of the year so far.
The price of gasoline is apparently one of the major causes of this drop in consumer confidence. There has been a big jump in the percentage of Americans who say energy is the top economic problem facing the country, from 5% in mid-February to 17% in mid-March.
Guess Who?
The top-grossing movie of the weekend was Guess Who, a highly updated version of the 1967 classic, Guess Who's Coming to Dinner? The old movie revolved around the reactions of two white parents (played memorably by Spencer Tracy and Katherine Hepburn) whose daughter brought home a black fiancé to meet them. The current version reverses the situation and portrays a black woman who brings home her white fiancé to meet her black parents (played by Judith Scott and Bernie Mac).
The 38-year span between the two movies draws attention to the fact that American attitudes about interracial marriage have changed significantly over the years. Gallup polling in 1968 showed only 20% of Americans approved of "marriage between blacks and whites." By 2004, 76% approved. When Gallup first asked the question in 1958, 4% approved. It was not until Gallup's 1997 survey that a majority of Americans approved of interracial marriage.
According to the June 2004 Gallup Poll, 87% of blacks approve of interracial marriage, compared with 72% of whites. There is a significant age difference, with older Americans significantly less likely than younger Americans to approve.
College Admissions
This week, hundreds of thousands of high school seniors will find out whether they have been accepted at the college of their choice.
If a student could attend any college in America, what should it be? There are many different criteria that could be taken into account in answering this question, needless to say, but looking at sheer prestige alone, the choice is clear: Harvard.
Twice over the last six years, Gallup has asked Americans to answer this question: "All in all, what would you say is the best college or university in the United States?" Harvard came out on top both times, significantly ahead of other school. In 1999, 16% of respondents mentioned Harvard, way ahead of the 4% who mentioned Stanford (M.I.T. and Princeton each received 3% of the vote). More recently, in 2003, 15% mentioned Harvard, while the other top three schools were significantly lower: Stanford 6%, Yale 4%, and M.I.T. 4%.