IA-03: Poll testing short and sweet messages about David Young

A survey is testing brief, positive messages about two-term Representative David Young in Iowa’s third Congressional district. A respondent took notes on the call on the evening of March 14.

My first thought was that the National Republican Congressional Committee or some outside group supporting GOP House candidates commissioned the poll. The phrases about Young didn’t have the level of detail I would expect from a survey designed by a campaign, and the question order was somewhat unusual.

On the other hand, Federal Election Commission filings don’t appear to show any expenditures by Young’s campaign on polling during the third or fourth quarters of 2017. Perhaps this survey is the incumbent’s early attempt to see where he stands.

The live interviewer (calling from 424-322-2159) asked for my source by name, indicating that the pollster is drawing from list of registered voters, rather than using a random digit dialing method. My tipster is a strong Democrat; I don’t know whether Republicans or no-party voters are also receiving the call. My source was not recording, so the questions below are paraphrased, not verbatim from the script.

Many political surveys start with questions designed to identify active or likely voters, such as did you vote in the last election, or how certain are you that you will vote this year. This poll started by asking, “Where do you get your information?” I assume they are collecting data to determine what kind of paid advertising (radio, television, Facebook, other online sources) will best reach targeted voters.

Strangely, the second question asked for the respondent’s year of birth. More often, political polls will ask about age near the end, along with other demographic questions (race, income, last level of education completed).

I would welcome hypotheses about why age might be used to screen respondents near the beginning of a poll about a member of Congress. My source speculated that the goal was to gauge Young’s popularity among voters above a certain age. Iowa’s electorate skews toward older groups. UPDATE: Bleeding Heartland user zeitgeist comments below that maybe they wanted to avoid oversampling any one age group, so could end the interview right away if they already had plenty of respondents of a certain age. That makes sense, though the same principle could apply to party affiliation or ethnicity or education, which were at the end of this survey (as is typical). Thinking further, I wonder if they are trying to verify that the respondent is the person they are trying to reach. Voter files include each individual’s age. Polling expert Charles Franklin suggested asking about age up front could be “to confirm this is the right respondent from the voter list.” Polling expert Mark Blumenthal agreed, adding the question order “could be about filling age quotas near [the] end of field period.”

SECOND UPDATE: Another reader, also a strong Democrat, got the same call on March 13 but was asked about age near the end of the survey and was not asked about the main source of news. That supports Blumenthal’s speculation that the interviewers needed more respondents from certain age groups near the end of the sampling.

Next, respondents were asked whether they had a favorable or unfavorable opinion of three politicians:

• Donald Trump
• Joni Ernst
• David Young

For Young, the caller asked whether your opinion was very or somewhat favorable/unfavorable.

(Note that this survey didn’t test name ID or favorability for any of Young’s six Democratic opponents.)

Thinking of all the work President Trump has done, do you approve or disapprove of the job he is doing? Somewhat or strongly?

Next, the caller asked about approval or disapproval for Republicans in Congress.

The rest of the survey focused on Young.

Do you remember how David Young voted on fixing health care? (My source was not sure whether the script included the name of the American Health Care Act, which House Republicans approved last May. Young had pledged to oppose that bill until shortly before the floor vote. UPDATE: Another respondent thinks they asked about repeal of the affordable health care law. LATER UPDATE: A third respondent thinks the question referenced the “health care bill” without naming the Affordable Care Act.)

Do you approve or disapprove of how he voted on health care?

For the next issue, the order of the questions was reversed. First, do you approve or disapprove of the tax bill Congress passed last year? (Again, my source was not sure whether the official name, Tax Cuts and Jobs Act, was part of the script. UPDATE: Another source thinks the question referred to “tax reform.” LATER UPDATE: A third respondent says the caller asked about the “tax bill.”)

Then: Do you remember how David Young voted on the tax bill?

Now I’m going to read some phrases: for each, tell me if you agree or disagree.

• David Young is on your side

• David Young has the same values as you

• David Young stands up for the middle class

(More than anything else, those questions made me suspect an outside group commissioned the poll, with a view to testing sound bites that would fit into 30-second television commercials or online ads for Young. I would expect a campaign to test more meaty messages, and probably also to include some negatives that Democrats might use against the candidate.)

In the presidential election last November, who did you vote for? My tipster confused the interviewer by pointing out that the presidential election was in November 2016; last November, we voted in city elections. Sloppy questionnaire wording is a pet peeve.

The last set of questions:

• party affiliation (they should know this already if they were working from a voter list, though what people consider themselves politically doesn’t always match their registration)

• Are you Hispanic?

• What ethnic description would fit you?

• last level of education completed

I will update this post as needed if other readers can provide additional notes, or better yet, a recording, of this poll.

  • Age question

    I’ve been asked the age question up front before and then told they had the sample size they needed of my demographic already; if they know the sample mix they want and you would create oversampling, it makes some sense to know that and move quickly on instead of wasting time on an interview that will be discarded.

    • that's a good hypothesis

      Though the same principle could apply to other considerations like party affiliation or education level, and they asked about those factors near the end, which is normal.

You need to signin or signup to post a comment.