Let’s give the dark money Massachusetts Fiscal Alliance It’s due. The first poll it released in 2015 was such total garbage that it won the coveted Award for Fraud in Massachusetts Politics. But its latest poll released yesterday is only overwhelmingly garbage. MFA is on the upswing!
Mass Fiscal, which masquerades as a public interest group but is actually a front for moneyed interests, sent out a new “poll” yesterday purporting to show overwhelming public disagreement for the legislature’s recent advancement of the Fair Share Amendment. The Fair Share question will likely go on the ballot in 2018 and is a constitutional amendment that would impose an additional 4% tax on incomes over $1 million per year in Massachusetts. The additional revenue, according to its backers, would be dedicated to education and infrastructure.
Unlike its Fraud Poll, this one had some elements that make it seem like a real poll. MFA identified the pollster and a bit of its methodology, though if MFA really wants to earn our trust it could comply with best practices in the industry I outlined in A Cure for the Mass Fiscal Fraud Poll Blues.
Anyway, you can find the latest edition here. It starts with Part 1, some questions about politicians’ approval ratings, which according to the very good pollster Steve Koczela of MassInc Polling Group, are in line with the findings of recent reputable polls. Mr. Koczela went to on to suggest that Parts 2 and 3 are not an accurate evidence of current public opinion, but might serve as effective tools to test anti-Fair Share Amendment messaging. That’s a very fair assessment.
Then there is Part 2, some questions about the legislature’s pay raise vote, with questions that were highly misleading. To Mass Fiscal, highly misleading is almost the same as the Ten Commandments.
MFA turns to Part 3, where having prepped the respondents with questions in Part 2 that would make any Zen Master want to go out and physically attack his or her state legislator, MFA lowers the boom on the Fair Share Amendment. These corrupt State House bastards are jacking up our taxes 80% so they can give themselves a pay raise!
It’s a classic case of response bias, in which the data is useless because the information provided leading up to the question and the question itself is so biased. In fact after I tweeted about it a colleague at another college got in touch to thank me for providing him a “Textbook example of how to create false poll results through leading questions. I will try to remember to use this in class next semester.”
Rather than reprint the questions, let’s give an example of Mass Fiscal’s technique.
Q1. Are you aware that Professor Cunningham is a drug abuser?
Q2. If you knew that Professor Cunningham is a serial killer, would you still cut him some slack knowing of his drug problems?
Q 3. Now that you know that Professor Cunningham is a drug abusing serial killer, would you say you are more or less likely to approve his dating your daughter?
But as I said, give Mass Fiscal some credit. This poll was better than the last one.