Cell phone use skews surveys

Lenski, who studied mechanical engineering at the University, was the third speaker on a panel composed of social scientists and pollsters. The panelists agreed that state polls are inaccurate, gave reasons for this and suggested improvements.

“How many of you don’t have landline phones?” Lenski asked the audience of undergraduate and graduate students.

Many hands shot up, supporting the panelists’ point that polls today are inaccurate because their samples do not include enough young people, who are more reliant on cell phones.

Contacting people by cell phone, though, is twice as expensive as calling a landline, said Larry Hugick, chairman of Princeton Survey Research Associates International. He explained that “[contacting people via cell phones] will not happen until we prove that [only dialing landlines] is a major issue.”

The panel was composed of two social science professors and two pollsters. Christopher Achen, a politics professor, and Andrew Gelman, a statistics and political science professor at Columbia, addressed what drives voting.

Achen said that people tend to make irrational associations between past events and political candidates, adding that people generally vote against incumbents when times are bad.

He provided the example of shark attacks along the New Jersey shore in 1916, when Woodrow Wilson, Class of 1879, was president. “Wilson lost as many votes in Monmouth and Ocean [counties] as Republicans lost during the Great Depression,” in the presidental election that followed a few months later, he explained.

“When the economy is down, voters tend to vote against the economists,” he added, explaining that “the Republicans are taking the hit” in this election cycle for that reason.

Gelman, author of “Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do,” disagreed with Achen on this point.

“I’m not sure comparing shark attacks to the Republicans taking the hit for the economy is accurate,” he said. “The shark attack comparison is a little too charitable to the Republican Party. There have been problems in the economy for the past couple of years.”

Gelman also said that variation in voting trends lies with the rich, not the poor, and added that religion influences the voting of the rich more so than that of the poor, saying “religion is the opiate of the elite.”

He added that social issues carry more weight now than in the past, especially among wealthy populations because they do not focus only on economic concerns.

Get the best of the ‘Prince’ delivered straight to your inbox. Subscribe now »

Stephanie Alvarez ’12 said she came to the panel to learn more about polling. “Anyone’s general opinion is that it’s easy to call someone and get a percentage, but it’s so much more complicated,” she said. “I’ll probably go to Wikipedia and look up polling information when I get back.”

Dan Myers GS said he thought the panel was especially interesting because it mixed social scientists and pollsters. “There was a little more diversity of topics that people talked about than I expected,” he said. “It helped me understand why the polls might be wrong in the upcoming election.”