Tweet
The Evidence of Things Not Seen
A study that supposedly validates "diversity" may do just the opposite.
BY CHETLY ZARKO
Friday, May 16, 2003 12:01 a.m. EDT
The New York Times' Jayson Blair, an individual reporter given to fabrication, was also, alas, a member of a group--a favored group, if we are to believe the Times' boasts about its efforts to achieve "diversity" in its newsroom. For better or worse, diversity policy has become part of the Blair controversy. It is also something on which the Supreme Court is about to rule.
The modern idea of diversity, of course, was given its first big push into the culture not in corporate hiring and promotion but in university admissions, where the 1978 Bakke decision enshrined both the word and the practice. Fittingly, the university campus is now the site of the biggest push against it--in a case before the high court questioning preferential admissions at the University of Michigan.
One of Michigan's major claims, in its legal arguments, is that student diversity enhances the environment for learning and improves the quality of education. Implied is the notion that when a greater number of blacks and other minorities are introduced into the classroom, a more diverse pool of ideas and "perspectives" is generated and everyone gains.
But is this true? The university has said yes, asserting that it has conducted studies showing just such an educational benefit. In its legal briefs and testimony, it has leaned heavily on something called the Michigan Student Study (1990-94). Patricia Gurin, a University of Michigan professor of psychology and the university's key statistical witness, has testified that the 1994 study, with two others, "consistently confirms that racial diversity and student involvement in activities related to diversity have a direct and strong effect on learning and the way students conduct themselves in later life, including disrupting prevailing patterns of racial separation."
But that wasn't the way the 1994 study was first understood. As it happens--through a Freedom of Information Act request--I was able to obtain a copy of the study's first "Executive Summary," submitted on May 24, 1994. It concluded that Michigan's racial preference programs actually "stigmatized" African-Americans and "increasingly polarized" the campus; that "self-segregation" was common; that "diversity of skin color" is not equivalent to "diversity of ideas" (financial disparities were more telling); and that diversity "quite simply . . . does not, in itself, lead to a more informed, educated population."
Supporting economically based race-neutral preferences, the report noted that "our results suggest that too often we tend to overemphasize differences in terms of the kinds of people we are, while underemphasizing real differences in resources (especially financial)." Ironically, the summary was co-written by Ms. Gurin's husband, Gerald, also a professor of psychology (now retired).
So what happened between then and now to lead Michigan to read the study another way? That is hard to say. Right now "executive summaries" are all we have--confusingly. In 1998, the first summary was consigned to the ashbin of history and another version was posted on the Internet--and it agrees with Patricia Gurin's testimony. Mr. Gurin is the co-author of both versions and is given credit by his wife for assisting in her expert testimony. (The Gurins never responded to my request for comment.)
A puzzle indeed. One would think, given the importance of the claims being made for the Michigan study and the differing accounts of what it says, that the court and the average citizen should be able to judge the numbers for themselves. But the university has locked away the raw data--for years, as it turns out. Thereby hangs a tale.
The Bentley Historical Library, Michigan's archival repository, used to allow researchers unfettered access to its collections, except for the records of the university's president and vice provost for academic affairs, which were sealed for 10 years. After 2001, however, a new policy sealed the records of all executive officers for 20 years. The logic, according to Fran Blouin, the director of the archives, was to protect historical accuracy by ensuring that individuals wouldn't be tempted to withhold or destroy records that might be politically damaging in the short run.
Unfortunately, the policy changed while the admissions lawsuits were going forward and after I (and other researchers) had made several requests for the admissions-related material. True, a successful appeal through the Freedom of Information Act could unloose some documents, but not easily. The university requires that its FOIA officer review each document, and it will charge up to $57 an hour for such a review. In these conditions, research becomes impossibly expensive.
Fortunately, the material I was looking for was carefully indexed, and when I sent a targeted request in February 2003 for a select few of the newer admissions-related folders from former President James Duderstadt's "restricted" collections, the university was obliged to produce something. And they did. On March 27, four days before oral arguments before the Supreme Court, I discovered the 1994 Executive Summary.
In April, alerted to the importance of the Michigan Student Study, I requested access to the raw data itself, hoping to be able to turn them over to statisticians so that they could perform their own tests and "peer review" the original data. On April 22, the university's FOIA office denied me access to the data, citing a Michigan law protecting data of "commercial value" to the university.
The current Internet version of the Executive Summary does discuss many of the issues raised by its predecessor, but it either focuses on the (hidden) data differently or suggests different data to support its more pleasant viewpoint. In one of the finer distortions of the first work, the Internet version uses a different data snapshot to conclude that "the findings call into question the common perception that students of color are self-segregating themselves on college campuses. The greatest self-segregation occurs among White students." This starkly contrasts with the first summary's conclusion that "students of color [are] not as concerned [as whites] with social 'integration.' "
Perhaps this is all explicable, but how are we to know when researchers and the public are denied access to raw data and when an internal, contradictory report is concealed from the public for years? In her testimony to the court, Patricia Gurin wrote: "Students learn better in a diverse educational environment, and they are better prepared to become active participants in our pluralistic, democratic society once they leave such a setting." What does a student learn about "our pluralistic, democratic society" from a university that behaves in such a way?
A study that supposedly validates "diversity" may do just the opposite.
BY CHETLY ZARKO
Friday, May 16, 2003 12:01 a.m. EDT
The New York Times' Jayson Blair, an individual reporter given to fabrication, was also, alas, a member of a group--a favored group, if we are to believe the Times' boasts about its efforts to achieve "diversity" in its newsroom. For better or worse, diversity policy has become part of the Blair controversy. It is also something on which the Supreme Court is about to rule.
The modern idea of diversity, of course, was given its first big push into the culture not in corporate hiring and promotion but in university admissions, where the 1978 Bakke decision enshrined both the word and the practice. Fittingly, the university campus is now the site of the biggest push against it--in a case before the high court questioning preferential admissions at the University of Michigan.
One of Michigan's major claims, in its legal arguments, is that student diversity enhances the environment for learning and improves the quality of education. Implied is the notion that when a greater number of blacks and other minorities are introduced into the classroom, a more diverse pool of ideas and "perspectives" is generated and everyone gains.
But is this true? The university has said yes, asserting that it has conducted studies showing just such an educational benefit. In its legal briefs and testimony, it has leaned heavily on something called the Michigan Student Study (1990-94). Patricia Gurin, a University of Michigan professor of psychology and the university's key statistical witness, has testified that the 1994 study, with two others, "consistently confirms that racial diversity and student involvement in activities related to diversity have a direct and strong effect on learning and the way students conduct themselves in later life, including disrupting prevailing patterns of racial separation."
But that wasn't the way the 1994 study was first understood. As it happens--through a Freedom of Information Act request--I was able to obtain a copy of the study's first "Executive Summary," submitted on May 24, 1994. It concluded that Michigan's racial preference programs actually "stigmatized" African-Americans and "increasingly polarized" the campus; that "self-segregation" was common; that "diversity of skin color" is not equivalent to "diversity of ideas" (financial disparities were more telling); and that diversity "quite simply . . . does not, in itself, lead to a more informed, educated population."
Supporting economically based race-neutral preferences, the report noted that "our results suggest that too often we tend to overemphasize differences in terms of the kinds of people we are, while underemphasizing real differences in resources (especially financial)." Ironically, the summary was co-written by Ms. Gurin's husband, Gerald, also a professor of psychology (now retired).
So what happened between then and now to lead Michigan to read the study another way? That is hard to say. Right now "executive summaries" are all we have--confusingly. In 1998, the first summary was consigned to the ashbin of history and another version was posted on the Internet--and it agrees with Patricia Gurin's testimony. Mr. Gurin is the co-author of both versions and is given credit by his wife for assisting in her expert testimony. (The Gurins never responded to my request for comment.)
A puzzle indeed. One would think, given the importance of the claims being made for the Michigan study and the differing accounts of what it says, that the court and the average citizen should be able to judge the numbers for themselves. But the university has locked away the raw data--for years, as it turns out. Thereby hangs a tale.
The Bentley Historical Library, Michigan's archival repository, used to allow researchers unfettered access to its collections, except for the records of the university's president and vice provost for academic affairs, which were sealed for 10 years. After 2001, however, a new policy sealed the records of all executive officers for 20 years. The logic, according to Fran Blouin, the director of the archives, was to protect historical accuracy by ensuring that individuals wouldn't be tempted to withhold or destroy records that might be politically damaging in the short run.
Unfortunately, the policy changed while the admissions lawsuits were going forward and after I (and other researchers) had made several requests for the admissions-related material. True, a successful appeal through the Freedom of Information Act could unloose some documents, but not easily. The university requires that its FOIA officer review each document, and it will charge up to $57 an hour for such a review. In these conditions, research becomes impossibly expensive.
Fortunately, the material I was looking for was carefully indexed, and when I sent a targeted request in February 2003 for a select few of the newer admissions-related folders from former President James Duderstadt's "restricted" collections, the university was obliged to produce something. And they did. On March 27, four days before oral arguments before the Supreme Court, I discovered the 1994 Executive Summary.
In April, alerted to the importance of the Michigan Student Study, I requested access to the raw data itself, hoping to be able to turn them over to statisticians so that they could perform their own tests and "peer review" the original data. On April 22, the university's FOIA office denied me access to the data, citing a Michigan law protecting data of "commercial value" to the university.
The current Internet version of the Executive Summary does discuss many of the issues raised by its predecessor, but it either focuses on the (hidden) data differently or suggests different data to support its more pleasant viewpoint. In one of the finer distortions of the first work, the Internet version uses a different data snapshot to conclude that "the findings call into question the common perception that students of color are self-segregating themselves on college campuses. The greatest self-segregation occurs among White students." This starkly contrasts with the first summary's conclusion that "students of color [are] not as concerned [as whites] with social 'integration.' "
Perhaps this is all explicable, but how are we to know when researchers and the public are denied access to raw data and when an internal, contradictory report is concealed from the public for years? In her testimony to the court, Patricia Gurin wrote: "Students learn better in a diverse educational environment, and they are better prepared to become active participants in our pluralistic, democratic society once they leave such a setting." What does a student learn about "our pluralistic, democratic society" from a university that behaves in such a way?
Comment