In this political season, state is awash in polls
Share with others:
Quinnipiac University will unveil a new poll on the Pennsylvania governor's race today, following yesterday's poll results on the state's Senate race.
That came four days after Rasmussen Research issued its latest survey of Pennsylvania, which was the same day Temple University and the Philadelphia Inquirer offered their snapshot of the two statewide races. That polling pair arrived two days after Franklin & Marshall College's Keystone Poll competed for the headlines with a Pew Foundation analysis conducted by Princeton Survey Research.
"The state's almost hyperpolling right now," said Chris Borick, the Muhlenberg College political scientist who directs the school's polling operation. "The number of polls and information that's out there is fairly staggering.."
"I've been at this for 20-some years and I've never seen anything like it in my life," said John Brabender, the veteran media strategist for Sen. Rick Santorum. "We've become, I guess, the most questioned state in America."
And the pace of polling will only increase in the six weeks between now and the election. Spurred by a governor's race and one of the most closely watched Senate races in the nation, it seems that hardly a day passes without a new number on the state's political pulse.
Yesterday's Quinnipiac survey found Democratic Treasurer Bob Casey leading Mr. Santorum 54 percent to 40 percent in a two-way race. The survey offered lots of other findings but that's the one that will be remembered, at least until the next poll comes along.
While such surveys can convey deep and sophisticated insights into the opinions and issue attitudes of the public, inevitably, in an election season, the headlines focus on who's ahead. That tendency is widely criticized for the distraction it presents to debate on issues. But it feeds the all-American hunger to know who's winning.
The longest continually running poll in Pennsylvania is the Keystone Poll, supervised by Dr. Terry Madonna, and headquartered first at Millersville University and for the last few years at Franklin & Marshall College.
Muhlenberg College published surveys in conjunction with the Allentown Morning Call. The Philadelphia Inquirer has recently published surveys by political scientists at Temple University. The Pew Foundation has sponsored recent polls conducted by Princeton Survey Research, and those are in addition to an extensive list of commercial firms taking regular readings of the state.
Speaking of the proliferation of polls in recent years, Michael Hagen, director of Temple's Institute for Public Affairs, said, "In some sense, I think it's a problem for the industry. Having so many polls at one time begs people to focus on the disparities and that, in turn, makes some people more suspicious. Maybe a hidden benefit of that is that voters may be less likely to be influenced by polls."
The discrepancy between two of the more recent polls sparked a rare public clash in the usually collegial world of the pollsters focusing on the state. The Keystone Poll, completed on Sept 18, showed Mr. Casey leading Santorum, 45 percent to 38 percent. A Princeton Survey Research poll, completed the same day, depicted the Democrats with a significantly larger lead of 52 percent to 31 percent.
Noting the discrepancy, the communications firm that distributed the Princeton pointed to differences in the demographics of the two polling samples. The firm suggested that the Keystone sample was light on Philadelphia respondents and heavy on those with college degrees.
Mr. Madonna shot off an angry response.
"Your IssuesPa release yesterday was beyond unprofessional," he protested in an e-mail to the Pew officials. "In 15 years being involved in surveys, I have never witnessed the release of a poll that trashes another poll in the release. This violates all standards of professional conduct."
John Miller, an executive with Braithwaite Communications, the firm that prepared the release, was anxious to heal the rift, insisting that it was written to anticipate reporters' questions, not deride the Madonna poll.
Polling is rooted in science but includes an element of art, and, as that exchange suggests, leaves room for plenty of disagreements. Most surveys are conducted by people making phone calls from random lists of phone numbers.
In recent years, some firms using automated calls have gained more prominence. Among the firms polling in Pennsylvania with that technology are Rasmussen and SurveyUSA.
Zogby International has pioneered the technique of Internet polling. Even though the universe of voters with access to the Internet does not match the population at large, Zogby maintains that it weights its findings to match the demographic characteristics yielded by random phone interviews.
Another key variable in political polling is how the surveys identify the so-called likely voters, those most certain to end up at the polls. And pollsters face increasing challenges from the rising level of people who refuse to be interviewed and by the fact that a growing share of the population is relying on cell phones rather than the land lines that random digit dialing systems depend on.
Several polling experts said that those wishing to indulge in the guilty pleasure of following the horse race rather than the issues should pay attention to the aggregation of polls collected at sites such as www.realclearpolitics.com, and www.pollster.com.
First Published September 27, 2006 12:00 am
