What do voters want?

What do voters want from healthcare reform?

I’ve answered variants of this question hundreds of times over the last two decades and my answer has been consistent:

• Lower costs

• Coverage for the uninsured

• Protection from what voters regard as rapacious insurance companies that evade their obligations to pay for needed care

While the third goal receives far less attention than it deserves, the first two are at the center of the healthcare debate consuming Washington.

Many people ask which — cost or coverage — is the higher priority. For 20 years my answer has been — holding down costs. This is rooted in our extensive data. In the last two years alone, we have asked voters in 26 different surveys how concerned they are personally about two healthcare issues — “Healthcare is becoming too expensive” and “Too many people do not have health insurance.”

In 25 of the 26 polls, more people evinced concern about the cost of healthcare than about the plight of the uninsured. The difference varies from place to place, but cost is the greater concern, as it has been for the 20 years we’ve been looking at these questions.

So imagine my surprise picking up a CBS/New York Times poll in June and reading that insuring the uninsured was judged a “more serious problem” than keeping costs down by a 39-point margin.

In the past three months, two other surveys from equally distinguished researchers identified coverage as the higher priority, albeit by lesser margins. But that is not the whole story: Six other public polls during the same period reached the opposite conclusion, putting the emphasis on cost.

Why the discrepancy?

When polls produce widely divergent results, the culprit is often non-attitudes. Sometimes the issues voters have never considered an issue until we ask. Pollsters push for opinions respondents don’t really have and which could therefore be expressed quite differently in different polls. While healthcare costs and coverage are central to voters, they may have never really prioritized them, though it is hard to imagine a 53-point swing based just on non-attitudes.

Another possible explanation lies in the questions themselves. Queries that identify cost as the greatest concern tend to focus on the personal, whereas those that put coverage in the lead focus on the national.

Recall our question asks how concerned voters are personally with each problem. Quinnipiac found a 16-point margin for cost asking “what do you think is the more important goal for healthcare legislation…” Gallup gave cost a 10-point lead when voters were asked which was more important “if you had to choose.”

Questions from which coverage emerges as the top priority tend to frame the issue in terms of significance for the country. When Pew found a 20-point margin for coverage, they asked — “What do you think is the more important goal for the nation…?” CBS/New York Times asked which is “the more serious problem…for Americans.”

Ancillary evidence suggests the primacy of cost — more voters report experience with cost problems than with being uninsured; voters are more likely to think holding down costs will help the economy than will covering the uninsured — but pollsters and pundits should exhibit some humility in interpreting all our results.

Even a 39-point margin can be less meaningful than it appears.