Press, Pollster Confusion Over MCRI

Yesterday I reported the good news that, according to a recent poll, the Michigan Civil Rights Initiative has strong support. Today, in a backhanded acknowledgment of what its reporters obviously view as that sad fact, the Detroit News reports that “Opposition to race ban wanes.”

Why do I think the reporters, Charlie Cain and Mark Hornbeck, are disappointed at this news? Look at their lede today:

Voter support for an affirmative action ban on the November ballot still hasn’t quite cracked the 50 percent mark, but opposition continues to drop, a Detroit News/WXYZ-TV survey shows.

Moreover, either purposefully or unwittingly, the two reporters misstate what MCRI would do, a misunderstanding apparently shared by the pollster, Ed Sarpolus. Indeed, a constant refrain of this article, and indeed of all opponents of MCRI, is that voters who support it are “confused.”

It would ban affirmative action in government hiring and contracting, and in college admissions.

“Voters are confused by the Civil Rights Initiative and many don’t know if it’s for or against affirmative action,” pollster Ed Sarpolus said Thursday. A “yes” vote would ban affirmative action; a “no” vote would preserve it.

On the contrary, as the plain language of MCRI that will appear on the ballot makes clear, MCRI would ban only

affirmative action programs that give preferential treatment to groups or individuals based on their race, gender, color, ethnicity or national origin….

It would not ban any affirmative action programs that do not employ preferential treatment based on race, sex, or ethnicity.

In fact, the article itself presents some evidence that supporters of MCRI are much less confused about it than are opponents. Consider, first:

A separate question on the survey asked voters whether government agencies and universities should be prohibited from granting preferential treatment to individuals or groups based on race, gender or national origin. On that question, 53 percent said yes, and 34 percent said no.

In other words, the more clearly and accurately the polling question expresses the effect of MCRI, the more people support it.

More evidence from the article of clear support, confused opposition:

Lori Flory, a 56-year-old retail business owner from Davison, who took part in the poll, said she’ll vote yes on Proposal 2 because “there should be an equal playing field for everybody.”

“I’m a woman, but I don’t think that should play a part in college admissions or hiring,” Flory said.

But poll participant Ron Bell, a 40-year-old accountant from Detroit, said he’ll vote against the initiative.

“I don’t want affirmative action used as a quota system; it should be merit-based. But I also don’t want us to go back to the days when all the good ol’ boy alumni are getting their children into the colleges,” Bell said.

Perhaps in a future article the Detroit News could provide some examples of “merit-based” affirmative action programs that give race, sex, or ethnic preferences of he sort that rings Bell’s bell. No affirmative action based on “merit,” to repeat, would be affected by MCRI. Nor, for that matter, would its defeat, i.e., preserving race preferences, do anything to prevent ’ol boys from continuing to exercise whatever influence they currently have. They would continue to give their kids whatever leg up they’re giving now; other kids would be given a similarly artificial boost because of their skin color or sex; and the only losers would be middle and lower class kids who did a bad job choosing their parents and their race or sex or ethnicity.

But voter Bell is no more confused than pollster Sarpolus.

Sarpolus said Ward Connerly, the California businessman behind the ballot issue in Michigan, “is relying on the same thing he saw in California and Washington state happening here, that in the end he’ll be victorious because of the confusion and people’s belief there shouldn’t be preferential treatment.”

If people oppose preferential treatment, where is the “confusion” in supporting measures that will end it? On the evidence of this article, opponents of MCRI, some pollsters, and some reporters are much more confused than MCRI’s supporters.

The people of California and Washington did not reject race preferences because of any “confusion,” and this poll suggests the people of Michigan will be similarly clear-headed.


Mark Hornbeck, one of the two authors of the above article, and I have exchanged a few friendly emails prompted by my post. Here’s the gist:

On Oct 6, 2006, at 10:44 AM, Hornbeck, Mark wrote:

Thanks for your insightful comments.

Are you suggesting there’s something inaccurate about the lead? According to our poll results, is it not true the proposal’s support hasn’t cracked 50 percent? Is it not true the opposition is dropping?

Mr. Hornbeck:

Thanks for your prompt reply.

No, I don’t think your lead is inaccurate. I’m not a journalist, but the way I see it you had a choice of three accurate leads: 1) MCRI has a substantial lead over its opponents; 2) MCRI’s lead over its opponents has increased; or 3) MCRI support still has not cracked 50%. All of these are accurate. I personally think the first two are more newsworthy than the third. For whatever reason, you chose the third, which in my view leads with the least interesting news in the new poll and presents MCRI’s current position in the least favorable light of any reasonable and accurate lead.

I’ve listened for more than a year to the two sides debate over whether this proposal would eliminate affirmative action programs. Bottom line is this: Upon the recommendation of the non-partisan staff at the state Bureau of Elections, the four-member, bipartisan election board decided to include ballot language that says Proposal 2 would eliminate affirmative action programs.

I don’t mean to be rude or disrespectful, but your “bottom line” sentence above simply is not true. The ballot language selected (for better or worse) by the election board does not say that “Proposal 2 would eliminate affirmative action programs.” As I quoted in my blog post from the official language, MCRI would ban only “affirmative action programs that give preferential treatment to groups or individuals based on their race, gender, color, ethnicity or national origin….”

MCRI supporters make a point of emphasizing that MCRI would not ban all affirmative action programs, only the ones that employ race, sex, or ethnic preferences. Do you disagree with their argument on this point, i.e., do you really think all affirmative action programs in Michigan employ preferential treatment based on race, etc.? Do you believe Presidents Kennedy and Johnson were being deceptive or just blowing smoke in their respective Executive Orders on affirmative action (10925 from Kennedy, March 6, 1961; 11246, from Johnson, Sept. 28, 1965), both of which used the same language to require government contractors to

take affirmative action to ensure that applicants are employed, and that employees are treated during employment, without regard to their race, creed, color, or national origin. [Emphasis added]

But whatever you believe of the MCRI supporters’ argument that all affirmative action would not be banned, or the obvious intent of Presidents Kennedy and Johnson to require affirmative action in government hiring or contracting that does not employ racial preference (in fact, that would ban all racial preference), your obligation, which you stated in your email to me, not to take sides but simply “to report the facts” would have prevented you from asserting as uncontroverted fact that MCRI would ban all affirmative action by government.

I apologize for sounding harsh or rude, but I’m afraid that insofar as there is “confusion” in Michigan about the meaning of MCRI, a great deal of the responsibility rests with the media and articles like this one.

The good news is that you still have time to take corrective action, and I hope you will!


On Oct 6, 2006, at 1:26 PM, Hornbeck, Mark wrote:

Your description of our lead appears to be intentionally incomplete. We said support hasn’t cracked 50 percent AND opposition has dropped… which is precisely what happened. The support side of the equation has increased by only a point in the last three months (47, 48, 48) while the opposition side has plummeted (47,38,33). That IS the story whether folks think it’s more interesting or not.

A fair point, although I’m not sure I agree that the decline of the opposition is more newsworthy than the sheer size of MCRI’s lead in the polls at the moment.

We’ve never said in any story that MCRI would ban ALL affirmative action programs. We’ve consistently said it would ban affirmative action in government hiring and contracting and college admissions. The proponents in Michigan have not taken issue with that description.

But it wouldn’t even ban all affirmative action in government hiring and contracting and college admissions!

I remind you that the two presidential executive orders implementing affirmative action in the federal government dealt specifically with government hiring and contracting, and those orders explicitly required affirmative action to see that no one was given preference by race. Your position here is those orders were non-sensical, that affirmative action in government hiring/contracting can only mean preferential treatment. I doubt that Ward Connerly or Jennifer Gratz would agree with that view. They would also insist, I’m confident, that there are affirmative action programs in college admissions that do not employ racial preference and so would not be banned by MCRI, such as increased recruitment in schools that have not sent many graduates to UM or other Michigan colleges.

Thus I continue to believe that your claim that MCRI would ban affirmative action in government hiring and contracting and college admissions is wrong or, at best, slanted (because it’s contested) and misleading and thus contributes to the confusion you discuss.

Say What?