Richard H. Sander (UCLA) started a firestorm of controversy with his article, A Systemic Analysis of Affirmative Action in Law Schools, 57 STAN. L. REV. 367 (2004). Using data from the LSAC National Longitudinal Bar Passage Study, Sander argued that minorities who attended more elite law schools performed did not do as well as minorities who were admitted to an elite school, but chose to attend a non-elite school at which their LSAT scores and undergraduate GPAs more closely matched those of other students--the so-called "mismatch hypothesis".
To test the mismatch hypothesis, Sander and others proposed to study candidates for the California Bar exam. Apparently, California collects detailed information on candidates for its Bar exam. One of Sander's proposed co-investigator was Dr. Stephen P. Klein, a psychometrician previously consulted by the California Bar examiners, as well as those of most of the other states. The study would be conducted using privacy safeguards, such as those employed in the LSAC Bar Passage Study, as well as in analyses of the Texas and the New York Bar exams. See, Proposed California Bar Study.
When the State Bar of California refused to release the data, Sander sued to force the release. The trial court held that the data was not subject to public inspection, because the data were "adjudicatory court records". In Sander v. State Bar of California, A128647 (Cal. Ct. App. June 13, 2011), a panel of the Court of Appeals reversed. The Court of Appeals found that the data were "public documents," rather than "adjudicatory court records," and remanded to the trial court
... to determine whether the Bar must produce the requested information after balancing the applicants‟ interest in confidentiality and the burden this request imposes on the Bar against the strong public policy favoring disclosure. The trial court is best suited to craft any qualifications to an order for production that can accommodate these concerns if possible.
Slip Op., at 1-2 (emphasis added). No doubt, the case will soon be on its way to Sacramento San Francisco and the California Supreme Court.
My problem with the proposed California Bar study is that, like the Texas and New York studies, on the primary investigator will have access to the underlying dataset. After the LSAC Bar Passage Study, a bowdlerized version of its dataset, stripped of names, law school identities to reduced to 6 clusters (see Interpretation 301-6: Low LSATs and High "Cut Scores"), and law school locations reduced to geographical regions, was made publicly available (see Note after the Executive Summary of the Bar Passage Study).
Hat tip: Chronicle of Higher Education, California Court Revives Affirmative-Action Scholars' Quest for Data on Law-School Grads (subscribers only).
Update: As Vladimir points out in a comment, I don't know how things work in California. It never occurred to me that the California Supreme Court did not sit in Sacramento, but in San Francisco.