This morning Montana State University-Billings will release the second half of its annual poll, and it’s expected to include a snapshot of the race between Jon Tester, Dan Cox, and Congressman Dennis Rehberg.

The poll is conducted every year by MSU-B political science professor Craig Wilson, the father of Evan Wilson, who is currently the political director of Rehberg’s Senate campaign (according to this wedding announcement).

Evan Wilson was also campaign manager of Rehberg’s U.S. House campaign in 2010.

Remember that year, MSU-B pollsters did not ask any questions about Rehberg because, as the poll itself noted, “Craig Wilson’s son is working in Dennis Rehberg’s campaign.”

So the question is, what changed this year?  What did MSU-B do to ensure accuracy and impartiality in this year’s poll if it was a concern two years ago?



61 Comments on "Impartiality?"

  1. One thing that has always bothered me is that lazy reporters continue to interview this guy for political stories–even though there are a dozen other poli-sci profs in Bozeman and Missoula etc they could interview whose kids don’t work for candidates.

  2. Im going to play Devils advocate here, so here it goes, Montana is a small state population wise. So for Craig Wilson’s son to be working for Denny Rehberg’s campaign I dont think it should be a big deal. Im just saying.

    • Yes, Montana is still a small state population wise – as compared to other states. There are still over 1 million people in this state. I find it hard to believe that they couldn’t find a poli-sci professor that didn’t have a son working for a candidate.

      Moreover, I don’t have an issue with this guy’s son working for Rehberg, but it does bring into question his bias and motivation. Farmboy, you may be right and there is no connection or bias. That said, I would take anything this person said with a large grain of salt.

  3. For me, the issue is not population size but rather that its hard to be unbiased when it comes to our own kids. The school clearly recognized this at one point- so why didn’t they this year?

  4. Larry Kralj, Environmental Rangers | October 11, 2012 9:04 AM at 9:04 AM |

    WHAT the..??? It’s easy, just SO easy. Simple really, as simple as this Moore’on. So, here’s aNOTHER fine Pubbie for craigy to defend!

    There are ONLY two types of Pubbies, the very rich and the very stoopid!

    Big Bird or a Big Kockh!

    I report, you decide!

  5. Expecting the worse, partisans dump the trash.

    About the poll:

    The “MSU Billings Poll” began in 1989. This is the twenty-fourth year the
    survey has been conducted and during this time twenty-six surveys have been
    completed. The Co-Directors of the Poll are Dr. Craig Wilson (Political Science),
    Dr. Scott Rickard (Director of the Center for Applied Economic Research) and
    Ms. Jennifer Pope (Sociology). We would like to thank Ms. Brenda Dockery for
    serving as the supervisor for this research project, Ms. Jessica Ridgway for also
    assisting us and Ms. Sandra Haley for coming out of retirement to again word process
    the survey results.
    This report summarizes the results of a statewide random sample telephone
    survey of adult Montanans who said they were both registered and likely to vote.
    The poll was conducted September 27-30. Students enrolled in the course
    Political Science 342, Media, Public Opinion, Polling completed 477 interviews
    for this survey. The survey was completed using random digit dial methodology,
    which selects both the phone numbers and members of a household to interview.
    Land line and cell phone numbers were included in the sample. 43.3% of the
    sample was reached through cell phone. The poll was conducted using MSU
    Billings’ CATI (Computer Assisted Telephone Interviewing) software. Multiple
    attempts were made to reach selected households.
    The survey results have a +/- 4.6% margin of error at a confidence level of
    95% (95% of the time the results will fall within the +/- 4.6% margin of error).
    Factors such as question order and wording can affect survey results. Statistical
    tests were employed to determine whether statistically significant relationships
    (p=.05) existed between key variables.

    It’s more than just prof. Wilson who are maligned and slimed by the nasty comments here. I’m sure the students appreciate it.

  6. The last two sentences were mine on my previous comment. Format error.

  7. I don’t know why MSU Billings lets Wilson embarrass the school like this, just to put out a fake poll of zero value. There is no upside, only negatives.

    That MSU Billings chooses to hurt the school for no value suggests a decision making process that Rolf Groseth needs to fix.

    • I wonder why for the benefit of students and citizens alike Wilson doesn’t simply recuse himself! It’s not that hard to do…but it makes you wonder.

  8. Craig Wilson is one of three principal investigators administering the poll, which is conducted by students. Even if Wilson wanted to cook the numbers, which I cannot imagine, his colleagues and students wouldn’t let him. His son’s choices are irrelevant to the poll and its results.

    The MSU’s poll reported 38 percent undecided in Daines v. Gillan. That three times the 12 percent undecided PPP just reported for that race. MSU’s result is at best flukey and may indicated a problem with the poll’s design and/or administration.

    The sample was 477, which is small.

    • “MSU’s result is at best flukey and may indicated a problem with the poll’s design and/or administration.”

      That’s typical for them. And it’s not just that MSU Billings misleads the state electorate with their fake poll, but it’s even sadder that they are teaching their students junk. Make up your own ballot language to test instead of surveying what is on the ballot? It’s just junk. I wouldn’t want to hire anyone who went through this process as I wouldn’t want to have to re-teach everything they were misinformed about.

      • The poll seems to have some problems. That can happen to any poll and is not evidence of incompetence. Craig Wilson is a consummate professional. It is one thing to raise an eyebrow at some of the poll’s findings. That’s legitimate. But it’s another thing, an illegitimate thing, to attack Wilson’s professionalism because the poll might be an outlier, and it is downright lowdown to suggest that he’s cooking a poll because his son works for Rehberg.

        • No one is saying he isn’t a professional, but its impossible to avoid unintentional biases completely, even if we are aware of them.

        • The poll ALWAYS has problems. It’s a joke.

          There is no professionalism. That’s the problem. As an actual political professional (not a professor) I look at this and wonder if MSU Billings is as bad at teaching in other professions as it is at teaching politics. Because when it comes to the profession I know, they are utterly incompetent. Students would be better of NOT paying MSU Billings to teach misinform them.

          The conflict of interest scandal is just drawing attention to something that was rotten long before Wilson’s son went to work for Rehberg.

        • Thank you, James Connor, for being, as ever, the voice of reason on this entertaining but sometimes wildly unfair blog. You are absolutely right in pointing out how low it is to accuse Wilson of bias without even a scrap of evidence.

          • Come on Ed, that’s why there is such a thing that is called a conflict of interest, for exactly such instances.

            The numbers are unprofessional junk, as usual. That’s a fact. That there is a clear conflict of interest just makes the process look as bad as the results.

          • This may be a “wildly unfair blog” but it often digs at the truth a bit better that the crap put out by the biased newspaper that employs you. If it were someone employed by the other side would you not be foaming at the mouth to “uncover” the story behind the story?

            • This blog may be liberal but it has stories and info that the Lee papers don’t or won’t cover. It wouldn’t hurt the papers to be entertaining once in a while either–good news coverage is supposed to be interesting. And since the IR no longer has an editorial page it fills a real gap in political analysis.

            • Richard: Short answer: no.

          • This post does not accuse Wilson of being biased. It simply asks a question, which is why Wilson has changed his policy this year. You guys don’t read very carefully.

  9. A true political professional would never utter such anti-intellectual garbage, not even under a nom de plume.

  10. Not long ago, when Romney was trailing Obama, some Republicans began trying to discredit public opinion polls, arguing that they were biased. I hope Montana’s Democrats are not trying to discredit the MSU poll because it reports that Democrats are trailing.

    • This poll has ZERO credibility.

      If instead of misleading students through this exercise, there was instead a class on how to critically read poll results, the students would be instructed to entirely ignore this poll. Not take it with a grain of salt, or realize a house-effect, no — they’d be instructed to entirely disregard this poll.

      This should be of no surprise to anyone, this poll has always been a joke. The only difference is that now there are more polls, real polls, coming out at the same time to embarrass MSU Billings. And the internet allows more people to be part of the discussion of what a joke this is. The conflict-of-interest begs people to talk more about it this year.

      But nobody, left of right, is going to say this is a good poll. It never has been. They should stop doing it, they are misinforming their students and misleading the electorate. They should both stop doing and refund the tuition money to their student victims. Unfortunately there’s no way they can compensate the electorate for their pollution.

  11. This poll was conducted over two weeks ago. How is this even still relevant information, and what kind of reputable political poll doesn’t release results for over two weeks? This is misleading at best, detrimental to the process and these students at worst.

    • The same can be said of any poll, regarding it’s relevance two weeks later. Still, I don’t think that is the point of the MSUB poll. I think it is supposed to be instructive concerning the process far more than predictive concerning the outcome. If one remembers that, then it shouldn’t be all that misleading, nor detrimental to the process of education which is its very purpose.

      • What’s misleading is that it is presented as current information, and implies that this is where the candidates stand right now– for instance where the reporter expresses surprise that so many voters are still undecided (not as surprising when you take into account the passage of 2-3 weeks). The paper is reporting it as though it is tracking public opinion. Headline should read “September MSU Billings poll numbers released today.” I’m sure the students learned a lot about statistics in this exercise, but it is hardly newsworthy 2 and a half weeks later.

  12. This poll is consistent with other polls that we’ve seen in the past- it has the senate and govs races within the margin of error and it has comparable favor ability ratings for Obama, Schweitzer, and Baucus.

    • I don’t mean these men’s ratings are comparable to each other, rather that they are comparable to the ratings they have received in the past.

  13. Look, folks, I’m not silly enough to suggest that polls don’t influence public opinion, but I am smart enough to realize that they shouldn’t. More to the point, polls this late in the game have very little influence at all, either as prediction or influencing factors. Bob Brigham is correct that all polling data, including any cross tabs or potential bias, should be released to indicate potential conflict of interest.

    But let’s be real. This poll is not a professional effort meant to predict outcome or sway voter opinion. It’s a school project. That more ios made of it, from the right or left, is absolutely ridiculous.

    • It is sold like it is the university releasing a study, not like a student project. If it were just a student project, the students should have all flunked.

      But I am utterly mystified how such junk meets the schools requirements to go out under they letterhead.

  14. Regardless, a two week old poll is a snapshot of how people felt two weeks ago, not today. It may have been relevant two weeks ago, but today it doesn’t mean much.

  15. Tester looked good in the PPP poll yesterday, and a new PPP Poll was just released an hour ago. Good news: Lindeen leads big over Skees! I-166 looks like it will pass, and Schweitzer’s numbers remain high.

    Bad news: The poll also says Juneau leads Welch by only one point, same for McCulloch over Johnson. Baucus numbers have dropped:

    “36% who disapprove. After seeing some improvement over the course of the year, Max Baucus’ numbers have headed back in the wrong direction this month with 35% of voters approving of him to 54% who disapprove.”

    Full results here:

  16. Bullock, Lindeen, McCulloch, and Schweitzer have higher favorability ratings with moderates and liberals than do Bucy, Juneau, and Baucus. Interestingly, voters of all age groups have a higher opinion of the first group than the latter.

    Young people sure don’t think much of Hill, which isn’t surprising.

  17. Youse guys, as they say in Butte, are giving Craig Wilson far too much credit.

    I have insider knowledge of the entire 22 year history of the MSU-Billings Poll. Craig Wilson cant even run the statistical software that is used in the Poll. He has never polled a single person. Wilson is rarely in the computer lab when the poll is being taken. Dr Joe Floyd (retired) and now Dr. Scott Rickard and his staff crunch the numbers. Wilson just shows up for the press conference and blabs the results.

    Wilson isnt the fastest boat on Flathead Lake and nobody at MSU-Billings will be sad to see him retire.

  18. This is odd, I just went back to the link I just posted for the PPP Poll memo and they have removed a sentence

    The memo had said:

    “36% who disapprove. After seeing some improvement over the course of the year, Max Baucus’ numbers have headed back in the wrong direction this month with 35% of voters approving of him to 54% who disapprove.”

    Now, that last sentence is missing.

    “Brian Schweitzer has his usual solid approval numbers this month with 54% of voters approving of him to 36% who disapprove.”


  19. The average of five polls taken since Labor Day puts Jon Tester and Denny Rehberg in a tie. Bullock is slight ahead of Hill. Gillan is way behind Daines. More at

  20. I don’t believe Craig Wilson cooks the books. He has made some bizarre comments over the years, like this one in the 2008 election:

    “Within the state, this may be one of the less interesting election years,” Wilson told Lee Newspapers. (Obama v. McCain, Schweitzer v. Roy Brown, a Democratic sweep of the tier Bs … I thought it was interesting.)

    In another interview, he said Missoula’s reputation as a haven for “liberal dope-smoking hippies” makes it hard to get candidates from the Garden City elected to statewide office. I suppose that’s true, although Linda McCulluch is from Missoula and Sen. Max Baucus claims Missoula (and Helena) as hometowns. It’s the language he used that bothers me. Let’s try this one on, “Billings reputation as a haven for tea party meth-snorting rednecks” makes it hard to get candidates from the Magic City elected to statewide office. I know, I know, Conrad Burns and Denny Rehberg are both from Billings. I rest my case.

    • I hit send before proofing. That’s McCulloch, not McCulluch, as in Montana Secretary of State Linda McCulloch.

    • I agree with Pete – its hard to take someone seriously who would malign an entire city like that. Has he even been to Missoula?

    • The only problem with your counter, Pete, is that Billings doesn’t have the reputation you would claim (except maybe inside Missoula). Now if you’d have applied that to Laurel …

      If memory serves, Wilson brought that up specifically regarding Van Valkenburg’s early aborted run for Governor, something I think we can all agree that we’re happy was so unsuccessful. Though indelicately put, the perception that Wilson is talking about is fairly accurate, not as a description of the city but how it is perceived by the majority of right-leaning Montanans. And if anything, statements of that nature show a particularly right-skewing perspective, that the perceptions of the right in Montana are somehow supreme in the professor’s mind.

      • Point taken. However, if I were to repeat my “Billings redneck meth culture” quote, over-and-over, the way the right does the “hippie dope smoking Missoula” tag, maybe it will stick.

        • I think you hit the point of bias regarding Wilson well. If you or I (you and I) were to repeat the Billings Redneck myth over and over, it would likely have little effect even across time. But for the Professor to repeat right wing myths over and over in interviews and TV appearances unfairly validates right wing perceptions as Montana perceptions.

  21. Craig Wilson’s ex-Mother-in-law, Esther Bengston (she looks like a Butte Madam) served a term or two in the legislature from Shepherd…until she was arrested for shoplifting in Billings.

    • This disgusting post should be classified under canine fecal matter. So should the author.

      • Agreed. Personally attacking Wilson, the supervision by the other faculty, and the students work with such nonsense to deflect from the results showing Tester in trouble is beyond weak. No one person steered this poll.

      • While most would agree that this isn’t one of Montana Cowgirl’s best posts — it lacks objectivity and authenticity — it does call into question the media’s reliance on Wilson as a source. Some also wonder about the methodology of the various MSU-Billings polls (not being a statistician, I’m certainly going to comment on that). I don’t believe that most readers think that Wilson has skewed the poll numbers, though.

        However, James, to call the post, and its author, canine fecal matter, aka dog shit, seems to be a bit harsh. One should keep in mind that this is a partisan blog, not the New York Times.

        • Um, I think James was talking about Firedog’s post, hence the canine references.

          • That’s right, I was responding to Fire Dog’s post. I think that should be clear from my comment’s position in the thread, but I probably should have written “Fire Dog’s disgusting post…”

            Pete, I incorporated the MSU poll’s results into my latest update on Tester v. Rehberg, Gillan v. Daines, and Bullock v. Hill:

            Readers should be mindful that campaigns are not beyond attempting to discredit polls reporting their candidate is behind in a effort to keep supporters and donors from becoming discouraged. I think that’s what’s going on with the criticism of Craig Wilson.

            • My error, James. I thought that might be the case but I’ve always considered that a post is the subject written by the blogger at the top of the page and everything that follows is a comment. A agree that Fire Dog’s comment was merde de chien.

  22. Speaking of Republican criminals, what is Pat Davison doing these days?Other than time that is.

  23. Tester wins, Rehberg kneels before Cox!

Comments are closed.