Wednesday, January 26, 2011

Adventures in Polling

David Lublin provides us some content from the once-formidable Maryland Politics Watch in the fact that Vinny DeMarco's folks over at the Maryland's Citizens Health Care Initiative is shopping around results of a poll that indicates two-thirds of Marylanders would conditionally support hike in the alcohol tax. The executive summary of the poll is here.

Now I once you see all of this stuff, you will know why it took almost a month for the poll to become public knowledge; the poll is complete garbage, for a number of reasons:

  1. The Pollster: DeMarco and Co. used OpinionWorks, a notoriously biased polling firm run by a former liberal operative and is tied in with the CenterMaryland crew of O'Malley Administration refugees.

  2. The Methodology: Check out the methodology on this poll:
    OpinionWorks interviewed 663 randomly-selected registered voters across Maryland who said they had voted in the 2010 Maryland general election. Interviews were conducted by telephone December 20–28, 2010. The poll has a potential sampling error of no more than ± 3.8% at a 95% confidence level, meaning the survey results would differ by no more than that margin at least95% of the time if every likely voter in Maryland had been interviewed.

    Interviewees were drawn randomly from the adult population of Maryland, balanced regionally, and screened for voting behavior to ensure that only active voters were interviewed. Weights were applied to bring the voter sample into strict compliance with the partisan and demographic breakdown of the voters who turned out in the 2010 General Election.
    It's bad enough that the data was culled from registered voters and was not appropriately determined to be active voters (if it was, the phrase "screened for voting behavior to ensure that only active voters were interviewed" would be unnecessary because instead of point out that "registered voters" were being polled, it would have noted "likely voters.")

    The timing is another issue. No credible poll canvasses voters for nine calendar days in an effort to get their result. But check out the dates as to when this poll was conducted. I thought this had to be a joke at first. I mean really, what credible pollster who actually accepts money conducts a poll around Christmas? You have a situation where a huge number of both registered and likely voters are just simply out of town and unavailable to be polled. That in an of itself is enough to throw out the data for me.

  3. The Raw Data: Where is it? It isn't on OpinionWorks website. It isn't on the Maryland Citizens Health Care Initiative Website. So where is the Raw data? Are the questions listed in the executive summary the only questions that OpinionWorks asked? Or were there others. And speaking of the questions....

  4. The Questions: From the questions that we know were asked (and again, we don't have the raw data so we can't be certain what was and what was not asked), we know that the people supported the alcohol tax increase only under certain limited situations. The questions were specifically phrased in two ways:

    1. “Would you favor or oppose an increased tax of 10 cents per alcoholic drink if the money was dedicated specifically to alcohol and drug treatment and prevention, health care for the uninsured, training for health care workers, and programs for people with developmental disabilities and mental health needs?”

    2. “Would you favor or oppose an increased tax of 10 cents per alcoholic drink to help address the state's budget deficit?”

    Those are the two questions addressed in the Executive Summary. Of course, the questions themselves assume that either of those things are actually going to happen. Yes, the current version of the bill (SB168) divvies up the moneys to certain services. But the bill also directs the remaining money once the new services and the alcohol beverage tax maintenance account have been funded to the General Fund. So at no point will the entire tax increase go to fund new programs or to address the deficit, particularly when one takes into account the new spending that is surely to be proposed.

    And of course, the poll also goes with this "dime a drink" fallacy. The questions really don't address the true and real cost of this proposed tax increase:
    • A 668% increase in the tax on distilled spirits;
    • An additional 668% increase in the tax on distilled spirits higher than 100 proof on top of the regular 668% increase;
    • A 740% increase in the tax on wine;
    • A 1289% increase in the tax on beer

The left-wing should be ashamed of trying to sell the public on a poll so full of garbage. Of course, it isn't the only thing that the Maryland Citizens Health Care Initiative should be ashamed of considering that we've already exposed the hypocrisy of their founder's work for Big Tobacco and we've thoroughly discredited of an alcohol tax that is one of many regressive taxes Maryland's liberal elite want to impose in their continuing quest to get their pound of flesh from Maryland's middle and working class families.

The OpinionWorks poll should be considered the same as Vinny DeMarco and the Maryland Citizens Health Care Initiative: disreputable.

1 comment:

John J. Walters said...

I think the main problem is that people are being asked if they would support an increased tax on alcohol -- not whether or not they would support an increase of this magnitude.

It's like saying, "yeah, I would be willing to pay more for a cheeseburger at McDonald's," unequivocally -- before you know whether the price increase will be 10 cents or a dollar.

ShareThis