By David Hill - 03/23/05 12:00 AM EST
Public polling conducted during last year’s presidential campaigns was much more sophisticated than in past election cycles. The research designs and strategies of innovative polling organizations — particularly Pew and Annenberg — were often very similar to those used by private pollsters. Bigger budgets supporting longer, more in-depth interviews allowed more incisive analysis of public opinion than ever before.
Now, we need public pollsters to bring that same approach to their analyses of public opinion about the policymaking process — particularly about Social Security reform.
The need for something better was illustrated by a question asked by NBC’s David Gregory during a recent presidential news conference. The reporter prefaced his query by stating that President Bush’s proposal for personal accounts for Social Security “remains, according to every measure we have, poll after poll, unpopular with the majority of Americans.”
Citing the body of extant polling on Social Security as evidence of anything is dubious. I would agree that there are lots of little polls that pose one or two cursory questions about Social Security. And I would agree that the takeaway from most of those would be that Americans are apprehensive about reforms. But it’s a shaky platform for serious analysis.
Much of this research can be faulted for failure to separate partisanship and opinions of Bush from opinions of Social Security reforms. A common question asks whether voters approve or disapprove of Bush’s handling of Social Security. One poll waded fully into partisan waters by asking voters whom they trust more on Social Security: Bush or Democrats in Congress? Is that a Social Security question or a generic election ballot?
From one perspective, there is nothing wrong with questions such as these. Today, voters know so little about the details of Social Security and its reform that they will indeed respond to potential reforms in a superficial, partisan manner. Democrats who think Bush is an illegitimate president and opposed his Iraq war will reflexively oppose anything else Bush is behind, including changes in Social Security.
But, from another perspective, these sorts of questions fail miserably in getting past voters’ initial reactions to an issue. Eventually, as many voters learn more about the full menu of reform choices, their opinions will step away from narrow partisan positions to broader self-interested positions. Sooner or later, many younger liberal Democrats will start to respond to Social Security more as young workers than as young partisans.
Private pollsters often employ quasi-experimental research designs to model this sort of transformation in perceptions of issues. This special approach to polling is intended to tell us where opinion stands today and to see where it will go once voters see, hear and know more about a topic.
Here’s how it would work with Social Security: At the beginning of each interview, I’d ask respondents where they stand on personal accounts. Then I would tell them that there is going to be a great debate on personal accounts during the course of this year. I’d say that I want them to hear arguments that supporters and opponents will make about personal accounts. Then I would read equal numbers of pro and con statements, asking how convincing each statement seems to be.
In a period of 15 minutes, I’d take voters through most arguments they are likely to hear in the next year. It’s akin to attending a fairly run town-hall meeting. Then I would come back and ask finally: Sometimes during a survey like this voters change their minds, so I’d like to ask you again, do you favor or oppose allowing personal accounts for Social Security?
I do this sort of thing all the time, and it’s fascinating how many people switch sides. Today’s cursory polling on Social Security underestimates the shift that’s coming toward support for personal accounts.
As voters know more, there’s going to be a lot of surprising movement in the polls.
Hill is director of Hill Research Consultants, a Texas-based firm that has polled for GOP candidates and causes since 1988.