08/27/2013 06:22 pm ET | Updated Aug 28, 2013

HUFFPOLLSTER: Why An Anti-Spitzer Poll In NYC Isn't A 'Push Poll'


Someone is testing very negative (and often misleading) messages about Eliot Spitzer, but don't call it a "push poll." Most Americans oppose mandatory minimum sentences for drug offenses. And Obama's approval seems to change more in some small subgroups than others. This is HuffPollster for Tuesday, August 27, 2013.

'ANTI-SPITZER POLL' CALLS RECEIVED IN NYC - Erik Engquist of Crain's New York Business reports what appears to be a telephone poll an 8-minute telephone poll being taken in New York that tests "highly negative statements about comptroller candidate Eliot Spitzer....The poll, administered from a Virginia poll number by a live caller who said she was reading from a teleprompter, raises four issues apparently to gauge their effectiveness for use in attack ads against Mr. Spitzer, the frontrunner in the comptroller's race. All four of the issues seem designed to appeal specifically to black voters, who have been found by recent public polls to support Mr. Spitzer over his primary opponent by the astounding margin of nearly 50 percentage points.." Engquist notes that the the survey included other more innocuous questions, including standard demographic items at the end of the interview including "age, party affiliation, race, and religious background." [Crain's New York]

Not a 'push-poll' - To his credit, Engquist went beyond the typical reporting of this sort of survey and identified what was either untrue or misleading about the negative arguments tested in the poll. He could not, for example, "locate any evidence" to support the poll's assertion that Spitzer opposed reforms to racial profiling and, "to the contrary," noted a report released by Spitzer as Attorney general in 1999 that raised "'troubling questions' about the high number of police stops of minorities." Yet more notably, Enquist did not label the survey as a "push poll." Why not? Because whatever its problems, the survey did not appear to meet the definition of that higher order of campaign malfeasance, which as explained by the American Association for Public Opinion Research (AAPOR) should apply to "telephone calls disguised as research that aim to persuade large numbers of voters and affect election outcomes, rather than measure opinions...The fact that a poll contains negative information about one or more candidates does NOT in and of itself make it a 'push poll.’ Political campaigns routinely sponsor legitimate “message-testing” surveys that are used by campaign consultants to test out the effectiveness of various possible campaign messages or campaign ad content, often including negative messages." [AAPOR; see also Stu Rothenberg, Pollster.com]

Guardian's Harry Enten: "That anti-Spitzer poll is not a push poll. You don't keep voters on the phone for 8 minutes asking questions & then drop the negative stuff." [@ForecasterEnten]

Pollster Steve Koczela: "Also typically way too expensive to do live interviewer for push polls given goal of reaching large #s of voters." [@Skoczela]

This case is relatively straightforward, but others may not be - The distinction between push "polls" and ugly message testing is relatively easy to make when in involves a live interviewer. As noted by Enten and Koczela above, the combination of costly live interviewers and a lengthy questionnaire implies message testing. An actual push poll, something that intends to spread negative information under the false guise of a poll, would have simply asked the offending "questions" and hung up. Unfortunately, such distinctions are getting more difficult as the real push pollsters now often rely on automated, recorded voice methods, which can blur the accepted definitions (it costs the perpetrators nothing to ask more questions, and some may even track and analyze the "results" of massive automated calling).

But what to call ugly message testing? - To paraphrase political scientist Jonathan Bernstein, such a poll may be loaded, biased or stacked (or just plain untrue), but that's not the same offense as a “push poll." The distinction may be lost on many bloggers and reporters, who often use the term "push poll" to describe virtually any poll question deemed in any way unworthy. Perhaps at some point pollsters will give up on the distinction -- much the way Google now defines "literally" to include strong feelings that are not "literally true" -- but your HuffPollster editors have not yet reached that point. [@jbplainblog, WaPost on "literally"]

MOST OPPOSE MANDATORY MINIMUMS IN DRUG SENTENCING - Emily Swanson: “Half of Americans in a new HuffPost/YouGov poll oppose mandatory minimum sentences, the tough sentencing laws that have come under fire again recently for imposing long sentences on low-level criminals and contributing to overcrowding in prisons. According to the new poll, only 32 percent of Americans think that the government should require certain minimum sentences for anyone convicted of a given crime, while 50 percent said that judges should have more leeway in determining sentences….The poll also found that 38 percent of Americans think the sentences usually given for drug crimes such as possession or sale of illegal drugs are too harsh -- slightly less than the combined 43 percent who said that sentences for those crimes were too lenient (23 percent) or about right (20 percent).” [HuffPost]

OBAMA’S APPROVAL FLUID AMONG HISPANICS, OTHER GROUPS - Frank Newport and Jan Sonnenschein: “President Barack Obama's job approval is more fluid among certain groups of Americans than among others. Of the major subgroups, U.S. Hispanics' approval of Obama has shown the greatest variation during his more than five years in office, with quarterly approval averages ranging from a low of 49% to a high of 80%....This suggests that when Obama's overall job approval ratings change, the groups most likely to be leading that change are Hispanics and liberal/moderate Republicans, and, to a lesser extent, whites and those with a high school education or less. Groups that would be among the last to shift their opinions when his overall approval rating changes are blacks, liberal Democrats, and postgraduates.” [Gallup]

HUFFPOLLSTER VIA EMAIL! - You can receive this daily update every weekday via email! Just enter your email address in the box on the upper right corner of this page, and click "sign up." That's all there is to it (and you can unsubscribe anytime).

TUESDAY'S 'OUTLIERS' - Links to more news at the intersection of polling, politics and political data:

-Democrats are 8 percentage points more likely than Republicans to prefer dealing with all their email when they're back from vacation. [Fox News]

-PPP finds Mike Michaud taking a lead in the Maine gubernatorial race. [PPP]

-Nate Cohn questions the idea that Congress is more popular than intervention in Syria. [TNR]

-Awareness of the coming health care exchanges is low among young adults. [Health Care Polls]

-Republicans think the media has been too easy on Hillary Clinton, Democrats think they've treated her about right. [YouGov]

-WBUR and MassINC re-up their partnership through June 2014. [MassINC]

CORRECTION:This story has been updated to clarify that Erik Engquist was unable to locate evidence to support a poll's assertion that Eliot Spitzer opposed reforms to racial profiling. An earlier version omitted the word "reforms."