Poll analyst explains tactics
By Cara Passaro
Flat Hat Staff Writer
Claudia Deane, assistant director of polling and a staff writer at The Washington Post visited the College yesterday to discuss her job and the relationship of polling to reporting the repercussions of the Sept. 11 attacks.
Deane, who holds a master's degree in international relations from the University of Sussex and a bachelor's degree from Vanderbilt University, said her polling unit is now in crisis mode as opposed to typical election and standard modes. The job of pollsters is to gauge public opinion. They collect ideas from news reporters and current events and write up questionnaires to elicit responses that will demonstrate a trend.
"What you want to do is compare to some point in the past," Deane said. "In that case you really just spend a lot of time looking at old questionnaires ... trying to see what you can ask again so you can do something more comparative."
A normal poll is conducted over three to four nights on a random sample. Pollsters examine the results, choose the big news and turn it into a story or hand the information off to a reporter already working on a project. Deane's basic information is also passed along to ABC, which shares the polling unit with The Washington Post.
Deane also talked about another problem in polling: actually talking to the public.
"It's very hard to get people to cooperate these days, it's hard to get them at home and it's hard to get them to cooperate," Deane said. "Sometimes we use the word 'Peter Jennings,' because people like Peter Jennings."
Deane does not complete initial survey phone calls, but she calls people back for more information and interviews for articles.
"It's amazing, because you need the opinions of average Americans," Deane said. "How many people will say, 'Oh you don't want to talk to me, I don't really know anything about it. You want to talk to my husband'? I get that all the time from women ... There's a number of people you have to talk into thinking they're smart enough. They get intimidated."
According to Deane, pollsters almost always address the public's opinion of the President, which determines his approval rating. For example, in June 2001, President George W. Bush's approval dropped from 63 percent to 55 percent. Deane's unit had information from a variety of demographics, spanning minorities, genders and political parties. Statistics regarding opinions on specific issues within the administration is also collected. Bush's approval ratings rose overnight following the Sept. 11 attacks. Sept. 9, 55 percent of Americans were satisfied with the administration, while Sept. 13, the number had risen to 86 percent.
Deane also spoke extensively about public support in times of national conflict.
"Public support is important," she said. "You can prosecute a war without public support, and it has been done. It's just very difficult. It's difficult to maintain over any length of time. The administration is following these polls with eagle-eyes, just hoping that [approval ratings] stay as high as they are right now. The president can leave public opinion...[It's not] like 'here's what the public thinks, this is what the president has to do.' That's not the way it works. The president can convince you. The president can lead you, especially on topics that you don't know very much about ... When it comes to foreign affairs, there is a big learning curve that has to happen in this country."
Deane said a poll was taken on the evening of Sept. 11, and its results showed that 94 percent of the U.S. public was in favor of military action, while three percent opposed and three percent had no opinion. Seven polls on issues surrounding the attacks have been conducted since. According to Deane, her team has talked to 4,729 people over the past nine weeks. The results of the Nov. 6 poll showed 90 percent in favor of military action, nine percent opposed and one percent with no opinion.
Deane's pollsters posed the question, "Do you think the attacks will make you personally more suspicious of people you think to be of Arab decent or not?" The results were 43 percent yes, 56 percent no and one percent no opinion.
Also being gauged is the extent to which the U.S. population is willing to support war. For example, the U.S. public may strongly support a general war against terrorism, but not a war in Afghanistan. Americans may also support the war, but only up to a certain amount of casualties.
Pollsters are also exploring methods of running polls on the Internet.
|
