MCRI Poll

READ THIS FIRST: New MCRI polling analysis is available here.

According to yesterday's Detroit Free Press, there is only 41% support for the MCRI (43% opposed, 3.5% margin of error).

While I hope that the MCRI goes down in flames, this doesn't tell us much. There are a few reasons why these numbers are insignificant.

• The margin of error for the poll is greater than the difference between the two of them (there is no significant difference between the two numbers)
• Most voters generally don't read the actual language of any ballot proposal until they reach the voting booth, and they might change their mind after reading the real thing.
  
• The "visibility" of most ballot proposals is also lower than that of races involving candidates, so a lot of people won't even know about the issue or won't make up their mind until closer to the election.
• Most importantly, the MCRI is a racially charged issue and respondents behavior may differ based upon that. People might say they're opposed because they're afraid of being viewed as racist, but privately support it on election day.

The DFP ran a similar article to this back in May. Here's what voter list deity Mark Grebner had to say about polling for ballot proposals back then:

It's a mistake to rely on polling related to ballot questions. They are often wrong by 30% or more.

In 1994, Proposal A (raising sales taxes, to permit property tax cuts) was supposedly a dead heat, the weekend before the election. Actual result: 70% yes.

In 2002, in a FAMOUS screwup, EPIC-MRA polled that year's Proposal A, which was a referendum on the Republican-passed legislation to eliminate single-action straight-ticket voting in general elections. According to EPIC-MRA, the proposal was leading 77% to 21%, with only 2% undecided. Four days later, the voters rejected it, 40% to 60%. The margin of error, calculated at 5% according to the laws of statistics, turned out to be 37%.

In another post, Grebner explains the importance of context in terms of how questions about ballot proposals are asked:

As long as I stick to a reasonably neutral way of asking about DeVos and Granholm, I'll get almost the same answers from people. Consider two exceptions: if I mention that Granholm is the Democratic candidate, the responses I get will be slightly more highly correlated with partisan attitudes than if I leave it out. If I refer to "Governor Granholm", I may trigger greater feelings of deference to an office-holder, or anger about job conditions than if I refer just to their names.

Because these effects are small in the case of candidate, we learn to overlook them. Candidate A is ahead of Candidate B by roughly the same margin, no matter how we ask the question. But in the context of issues, especially ones without clear definitions, context effects can be very large.

While we're on the subject of the MCRI, there was a slanted and patronizing editorial about the "initiative" a couple days ago in the Jackson Citizen-Patriot. Its message was essentially that BAMN sucks and that the MCRI should be on the ballot, while making no mention of how supporters of the MCRI collected signatures for it (they lied about it, basically).

As a student at the University of Michigan, I am very familiar with BAMN and agree that they ultimately undermine the cause of affirmative action (Rob Goodspeed has a great overview of the group's flaws on his website). However, BAMN doesn't represent all supporters of affirmative action, and just because the Voting Rights Act wasn't violated doesn't mean that it is right to put it on the ballot. If voters were told to sign a petition by MCRI operatives in support of affirmative action when the MCRI really aims to abolish affirmative action, their actions are fundamentally wrong. Ballot initiatives require a certain number of signatures to give them legitimacy. If these signatures were obtained using deceit, it means that the ballot initiative is not legitimate.

Oddly enough, I just noticed that my good friend David today published a post about this issue entitled "Racism Will Trump Decency." I don't think that the MCRI will pass, because with both the MDP and the MI GOP opposing the initiative, its going to be much harder for them to find funding. David isn't as optimistic, largely because of the relationship between the city of Detroit and the rest of the state. Its tough to say either way -- here's what he had to say:

I do not see the optimism others have for this initiative. While I hope and pray affirmative action stays as it is such an important program to help ease the problems we have in this state when it comes to economic opportunity and the correlation of race (see: Detroit, Flint, Benton Harbor), I don't see the voters of western Michigan or Macomb County and Western Wayne county allowing this to stand. In state politics everything seems to come down to this "us vs. them" mentality of Michigan vs. Detroit. This is just a nice cover for "white vs. black" and this same attitude is prevalent in suburban Detroit. I do not think that it is as large a problem in Oakland County (though you will certainly find your pockets of it there) as it is in Macomb though where I think the vast majority of white voters may say they won't vote against it, but in the booth by themselves have no problem approving the MCRI. It will be a sad day. Hopefully we can cut our losses and reelect Governor Granholm who has found ways to fix Republican right-wing messes in the past. But mark my words people, this initiative will pass. Racism will trump common sense and decency this November 7th.