My sister-in-law visited yesterday, and at 2:53 p.m. CDT, our phones beeped. We both received the following text message from “Research-Polls”:
Research-Polls.com purports to be a market research company from Fort Lauderdale, Florida, based in the Casa Grande condos at 508 Hendricks Isle. It’s hard to tell who owns the company, but the Web domain was created on April 21, 2022.
So sure, let’s see what kind of research they’re doing in the Big House:
The survey opened by asking sex and whether we are 18 or older and registered to vote in South Dakota. It apparently sought to exclude people who work in the media or polling:
Then it got to political business. The survey only let me go through the questions once, so I cannot tell which questions “forked” the survey—i.e., I don’t know if I would have gotten different questions by telling the anonymous pollster that I was unlikely to vote or that I was leaning toward one party or candidate. But note that every question is marked with a red asterisk, indicating that every question was mandatory: I could not leave any question blank and advance to the next question.
Interesting: the question text leaves “headed in the right direction” unqualified, but prefaces “headed off on the wrong track” with “seriously”, suggesting that respondents should only mark the negative option if things are really, really bad.
Now the poll is testing the message, seeing if it can get people to respond differently to the same question by tagging the generic Democratic candidate with “Biden”, “Liberal”, and “Congress”. Note that at no point does the poll try pushing in the other direction by asking, for example, if we would be most likely to vote for “a Republican candidate who supports insurrectionist Donald Trump and the conservative Republicans in Congress.” (However, there was a question later asking whether I approve of Donald Trump—I failed to screencap that question.)
I marked “Undecided”,so I think the poll served me this fork question to see which way undecideds might break:
…as well as this probe to identify obstacles Noem might try to counter with some last-minute marketing (I would humbly suggest not flying to Florida for fundraisers for the next three weeks):
The poll looks for correlations with the votes on the ballot measures for Medicaid expansion and marijuana legalization. The poll does not appear to push for the No votes Noem would like on both IM 27 and Amendment D:
The poll then poorly words this question about job performance. Evidently, the pollsters can’t bring themselves to make explicit the idea that Noem has performed her job poorly; they thus awkwardly pair “performed her job as Governor well enough to deserve re-election” with “give a new person a chance to do better.” The two options don’t fit: an incumbent could do a good job but still not deserve to keep the job against an even better candidate, and an incumbent could do a bad job but still be better than the alternative.
The poll ignores Smith’s job performance and instead digs for the intensity of Biden sentiment:
The poll then piles a whole bunch of issues into a question and asks respondents to pick the single most important:
I checked Abortion and Reproductive Rights to see what would happen. Strangely, the poll paired abortion with gun violence and asked me to say whether those two issues are more important than the redundant “Inflation and rising prices”. I’d be curious to know whether any other issue people picked was paired with “gun violence” and then put head to head with “inflation”.
The poll then dug into negative statements about Noem and Smith. Note that the negatives on Noem all relate to specific actions Noem has taken has Governor, while the knocks on Smith are all generic labeling, not actual positions, votes, or actions:
The poll sneaks in one more political question, about support for the NRA, before ending with one question about my race:
The original text invitation claimed that Research Polls was interested in how things are going in our state, but not one question dealt with any issues specific to South Dakota. This poll, like Kristi Noem, seems uninterested in the details of events and policy in South Dakota and far more interested in identifying messages that will help Noem win one more election.