On tackling the deficit, voters by a margin of 2-to-1 support raising taxes on incomes above $250,000, with 64 percent in favor and 33 percent opposed.
Independents supported higher taxes on the wealthy by 63-34 percent; Democrats by 83-15 percent; and Republicans opposed by 43-54 percent.
Support for higher taxes rose by 5 percentage points after Obama called for that as one element of his deficit-reduction strategy last week. Opposition dropped by 6 points. The poll was conducted before and after the speech.

Let us stipulate, for a moment, that such polls accurately reflect the opinion of the American voting public. This demonstrates a political advantage for Democrats, who advocate such tax increases, but even if raising taxes on the rich is popular, a poll cannot tell us whether such a policy will succeed in reducing the deficit. Nor can a poll tell us whether a soak-the-rich tax policy will have negative economic consequences (e.g., disinvestment, as global investors shift their resources to foreign countries with lower taxes) that voters surely do not desire.

And here we rescind the previous stipulation as to the accuracy of the poll results, in terms of their guidance as to how the deficit should be reduced. If you look at the full 23-page results of the poll (PDF), you find that the question about raising taxes on those earning over $250,000 is the ninth question on the survey. Here is the order of the questions:

1. Do you approve or disapprove of the job Barack Obama is doing as president?
2. Do you approve or disapprove of the job the Republicans in Congress are doing in office?
3. Do you approve or disapprove of the job the Democrats in Congress are doing in office?
4. Overall, do you have a favorable or unfavorable impression of Barack Obama?
5. In general, thinking about the way things are going in the country, do you feel things are going in the right direction or that things are going in the wrong direction?
6. If you had to choose, which one of the following should be the top priority for Congress: Reduce the deficit, cut taxes or maintain services and benefits?
7. Do you approve or disapprove of how President Barack Obama is handling the federal budget deficit?
8. Looking ahead to the next round of budget battles in Washington, would you support a shutdown of the federal government if: It stopped the Republicans from cutting spending for Medicare and Medicaid, it stopped the Democrats from increasing the budget deficit, or you do not support a government shutdown under any circumstances?

Only after answering all of those questions is the respondent asked:

9. Do you support or oppose doing each of the following to deal with the federal budget deficit: Increase taxes on income over $250,000? Cut Medicare and Medicaid? Reduce military spending? Raise the federal debt ceiling?

Note that the respondent is asked this only after being asked (in Question 6) to make a choice between the congressional priority of deficit reduction, tax cuts or maintaining benefits, and immediately after being asked in Question 8 about the prospect of a government shutdown as an alternative either to Republicans cutting Medicare/Medicaid or Democrats increasing the deficit.

You don’t have to be an expert in psychology or polling methodology to see that the respondent’s answer to Question 9 might be influenced by having first been asked Question 6 and Question 8. And if your interest is policy rather than politics, you see that in some sense all three of these questions involve false choices, asking respondents to choose between oversimplified alternatives. Democrats would certainly dispute the implication of Question 8 that they are in favor of “increasing the deficit” as much as Republicans would object to the “throw-Granny-to-the-wolves” suggestion implicit in the question’s description of their Medicare/Medicaid policies.

All such considerations of bias, however, must take a back seat to the question of whether this poll’s sample is a true reflection of the active electorate.

Among the registered voters in the sample, the partisan breakdown is 35% Democrat, 27% Republican, 36% independent and 2% “other.” Which is to say that, in this sample, self-identified partisan Democrats are 29.6% more numerous than Republicans (35 minus 27 equals 8, and 8 divided by 27 equals 0.296).

How does that compare with actual election results?

In 2008, with record turnout in a high-tide year for Democrats, Obama got 69.5 million votes (52.9%) to John McCain’s 59.9 million (45.7%), so that Democratic voters were 15.8 percent more numerous than Republicans (52.9 minus 45.7 equals 7.2, and 7.2 divided by 45.7 equals 0.1575). This means that the Democrat partisan ID advantage in the Marist/McClatchy poll is nearly twice the actual Democrat margin in the 2008 election.

Therefore, comparing the 2008 election results to the Marist/McClatchy survey sample, we must conclude either:

A. The McCain-Palin ticket overperformed among independent voters enough to cut in half the 29% Democrat advantage in partisan ID; or
B. There has been a significant uptick in Democrat advantage in partisan ID since 2008; or
C. The Marist/McClatchy poll oversampled Democrats.

We have strong reason to disbelieve either A or B and therefore, the explanation is obviously C. So we then conclude that this poll is misleading in that it both oversamples Democrats and would appear to bias the responses to Question 9 by preceding it with two tendentious questions. Nevertheless, when McClatchy Newspapers reported the result of this survey, the headline declared with certainty, “Poll: Best way to fight deficits: Raise taxes on the rich.”

What would be the response to such questions if the public were accurately informed of the policy implications?

What if you just handed the respondents a copy of the graphic at the top of this post, and then asked:

“Given that the top 1% of income earners pay 39% of all federal income taxes, while the bottom 50% of income earners pay just 3% of such taxes, would you agree or disagree that President Obama is a lying sack of crap?”

I’d love to see that headline — “Poll: President is a lying sack of crap” — but I’m not holding my breath.

Comments are closed.