How Has Mich.'s Ban on Affirmative Action Affected Minority Enrollment?

The U.S. Supreme Court on Tuesday affirmed Michigan’s constitutional amendment banning race-conscious admissions. Although the decision didn’t directly address the constitutionality of race-conscious admissions policies, the dissenting opinion, written by Justice Sonia M. Sotomayor, cited student-demographic data as proof that the ban, which went into effect in 2008, has adversely affected minority enrollment and diversity at the University of Michigan at Ann Arbor.

Justice Sotomayor may not be wrong, but the numbers are a bit more complicated than they may seem.

According to the data from the U.S. Department of Education cited by Justice Sotomayor, the proportion of African-American freshmen at the University of Michigan at Ann Arbor dropped by about one-third after the ban on race-conscious admissions went into effect.

In the four years before the ban, black students made up 6.1 percent of first-year students, on average. In the four years after the ban, they composed 4.5 percent of first-time, full-time students.

The percentage of Hispanic freshmen has also dropped, from an average of 4.8 percent in the four years before the ban went into effect to an average of 4.0 percent in the four years after.

Although both the number and the percentage of black and Hispanic freshmen at the University of Michigan have dropped since the ban went into effect, it’s not clear that the campus has become any less diverse as a result.

We can look to the University of Michigan’s diversity index, which measures the likelihood that two students picked at random will be from different racial or ethnic groups, on a scale of 1 to 100, with 100 being the most diverse.

We see that, although Michigan’s diversity dropped precipitously in the first year after the ban went into effect, it has since rebounded, and one of its most diverse freshman classes in the past decade matriculated after the ban took effect.

In the four years since the ban went into effect, the average diversity index of the freshman class was 48.03. The average diversity index in the four years before the ban was 48.99.

But here’s a caveat: The diversity index can be affected by the number of students who don’t report their race or ethnicity to an institution and who are therefore not included in the calculation. Often, particularly in states with race-conscious admissions policies, many of those students tend to be white.

That pattern seems to have played out at Michigan, where the percentage of students who didn’t report their race dropped from 6 percent to 2.5 percent in 2009, the year after race-conscious admissions were banned, while the percentage of white students rose from 71 percent to 73 percent. The unusually small percentage of students whose race was unknown may also have caused the seemingly large drop in Michigan’s diversity index that year.

However, in the two most recent years with available data, the percentage of incoming students whose race is unknown leveled out, as did the diversity index. This may be a result of the way race and ethnicity were reported to the Department of Education. Starting in 2010, students could be categorized as “two or more races,” a group which the diversity index treats as “diverse by default"--in other words, picking two people who are both two or more races is the same as picking two people who are two different races. From 2010 through 2012, about 3 percent of first-time freshmen at Michigan were reported as two or more races, which could have slightly increased the campus’ diversity index.

So while Justice Sotomayor is right that the ban on race-conscious admissions might have had an adverse effect on the number of minority students who enrolled, it has not necessarily made the campus any less diverse, at least statistically speaking.

Clarification (4/23/2014, 12:30 p.m.): The graphs in this post have been updated so that the y-axis starts at zero.