LAME-ASS PUSH-POLLING: Got a call from a polling outfit. They asked me a lot of questions, but they mostly seemed interested in making sure that I knew a potential candidate for State Senate had just gotten divorced. What I mostly know now is that another potential candidate for State Senate is ready to play the sleaze card at the first opportunity.
Guess which discovery is more likely to affect my vote?
UPDATE: A reader emails:
As data specialist for a polling outfit myself, I think I may be able to shed a little light on the nature of “push-polling” (quite a sensitive subject among reputable pollsters). The questions the interviewer asked you about a candidate’s divorce sounds less like muckraking and more like what is generally known as “message testing,” and not only is it considered perfectly ethical, but in my years of working with polls I’ve seen very few questionnaires that don’t include message testing to some degree.
If the message-testing questions refer to information that is accurate, and are positioned later in the survey than the initial test ballot question, then it’s likely you’re talking to an interviewer from a legitimate polling outfit. Such questions are an ethically acceptable branch of an aspect of campaigning affectionately known as opposition research; in the case of polling, it’s simply an effort to sound out what could be one’s opponent’s greatest strengths and weaknesses (or, for that matter, one’s own).
Hmm. I’m not sure, but I think that the divorce question came first. But I definitely finished the poll with the impression that they wanted me to remember the divorce issue, and not that they were just asking about it.
ANOTHER UPDATE: Reader Andrew Boucher emails:
Your e-mailer is exactly correct. This was no push poll. At the outset of most campaigns, candidates will put a benchmark poll in the field to test messages and determine vulnerabilities. They’re trying to figure out what to use for television ads, press conferences and general campaign issues. They want to know if you’re more or less likely to vote for the candidate based on the divorce issue (or the fact that he voted for higher taxes, or supports gay marriage, or is pro-choice). It’s not even out of the realm of possibility that the poll was conducted by the divorced politician’s campaign. (We always test the negatives on our own candidates, especially the glaring ones.) They’ll need to know ahead of time if they’re going to be on damage control in October (or if he’s even going to stand for re-election).
The other thing that’s relevant is the length of the poll. Push polls are very quick (they’re going for the widest possible audience), and they’re designed to drop a single negative on the respondee immediately. “Hi, I’m calling from a research group with a quick poll. Would you be less likely to vote for Senator A if you knew he molests collies? Thank you.” Then it’s on to the next call. These are considered very unethical and are actually pretty rare.
The call you described was clearly an early benchmark poll, probably to about 400 likely voters, by a candidate who trying to figure out whether he or his opponent is vulnerable on a slew of issues. If the data comes back that people don’t care about his divorce, you won’t hear a thing about it in the campaign.
As a political consultant, I, like the e-mailer, am sensitive on this issue. Benchmark polls are purely ethical, very useful, and often wrongly portrayed in the press as push polls.
Hmm. Okay. . . . But I think that even considering the use of a divorce as a campaign issue is tacky.
MORE: Another reader emails:
I think you are too quick to concede your initial anger at what you thought was a “push-poll” to the objections raised by your correspondents who are, themselves, practioneres in the field. Forgive me if I think their perspective is not entirely without self-interest.
Your correspondents are attempting an ethical slight-of-hand by drawing a bright moral line where there is none, between a poll designed “only” to diagnose that the electorate can be manipulated by a issue-free ad-hominem sleaze attack, and a poll designed to actually carry one out.
Is an unarmed artillery spotter who calls in coordinates to the gunner less a part of the army than the guy who actually fires the gun? The opposing army will have no trouble answering that question. If the gun blows Christ the Savior to bits, would one condemn the gunner and excuse the spotter, on the grounds that the latter “merely” diagnosed the vulnerability of the manger to the actions actually carried out by the former?
Even from a perfectly utilitarian viewpoint, I think your initial uncompromising response is better. By reacting with visible anger to even the suggestion that this topic is appropriate for discussion in a campaign, you help raise the bar for sleaze campaigning. It would be quite desirable if the damnfool who commissioned this poll begins to wonder whether even asking in a theoretical blue-sky gee-what-if kind of way about this sort of issue is political Russian roulette with five cylinders loaded.
Well, I hope so.