This report was approved for publication by the Association’s Committee A on Academic Freedom and Tenure in June 2004 and adopted by the Association’s Council in November 2004.
Many employers in the United States have been initiating or expanding policies requiring background checks of prospective employees. The ability to perform such checks has been abetted by the growth of computerized databases and of commercial enterprises that facilitate access to personal information. Employers now have ready access to public information that had heretofore been difficult to collect without an expenditure of considerable effort and money—criminal records, litigation history, worker-compensation claims, marriage records, bankruptcy liens, court judgments, and more. They also have ready access to private information—creditcard history, airline use, certain telephone records, bankaccount histories, pharmacy records, and even records of medical visits. The ready availability of these data creates the serious possibility of promiscuous, unfair, and perhaps even abusive investigations, and yet access to these data is subject to few legal limits.1 Systematic inquiry into such personal information is nevertheless commonly understood to constitute a serious and harmful intrusion upon an individual’s privacy.
Higher education has not been immune to the siren call for background information. Legislation mandating background checks for all employees of certain public institutions (which may include all or some public institutions of higher education) has been adopted in at least one state. The purchasing consortium of the Committee on Institutional Cooperation (made up of the Big Ten universities and the University of Chicago) now makes available at a discount the services of a backgroundchecking company; it is for each participating institution to decide whether and how these services will be used. Some colleges and universities have initiated or expanded background investigations of candidates for faculty appointments. This interest in background checks has arisen despite the absence of any systematic study of the need for the information such checks might produce. The interest within higher education was especially stimulated by the extraordinary discovery in 2003 that a respected member of the faculty at Pennsylvania State University had for decades been on parole for murders he committed in another state when he was a teenager. The misrepresentation of faculty credentials or experience is not totally foreign to higher education.2 But such sensational incidents are fortunately few, and almost all can be avoided if faculty search committees exercise reasonable care.
Because colleges and universities are now considering extensive and intrusive background checks with an urgency that seems quite out of proportion to the actual problems facing the academy, the AAUP’s Committee A on Academic Freedom and Tenure appointed a subcommittee to consider the question of the standards that should guide academic institutions in the implementation of background checks.
Fortunately, we need not write on a clean slate. Almost three decades ago, the Privacy Protection Study Commission, created by the federal Privacy Act of 1974, addressed the tension between individual privacy and institutional needs for information in the context of employment.3 It focused on two central issues of relevance here: the scope of background investigations and the procedures employed when such investigations are conducted. It concluded, rightly in our estimation, that the former must be guided by a norm of proportionality and the latter by concerns for both accuracy and fairness. We believe these governing principles to be applicable to higher education, and we address them below.
Scope of Investigation
Because colleges and universities must repose a high degree of trust in their faculties, they are justified in attempting to ascertain whether candidates are worthy of that trust. For this reason, they ordinarily charge search committees with the responsibility of conducting thorough checks of a candidate’s references and of interviewing a candidate’s present and former colleagues. Search committees should take this responsibility very seriously. To ensure the accuracy of a candidate’s curriculum vitae, search committees also check educational credentials, prior employment, professional experience, and the like. No doubt such reference checks entail some compromise of the privacy of candidates, but it is justified in light of reasonable institutional needs.
We recommend that this principle of proportionality be preserved. The privacy of a candidate should be compromised only as necessary in order to secure information that may ensure that applicants are qualified to meet the particular obligations of specific positions. Thus, for example, appointments involving access to federally classified data may require government approval, which may justify background investigations of sufficient scope to ensure that a candidate will meet government standards. Similarly, appointments requiring the bonding of an appointee may require a background check of sufficient scope as to satisfy a bonding authority.
Our primary recommendation is that the principle of proportionality prohibits the adoption of a general policy of searching the criminal records, if any, of all applicants for all faculty positions. The mere fact of an applicant’s having been swept up into the criminal-justice system is not, by itself, relevant to his or her suitability for a faculty position. For example, many faculty members, as students, were convicted of civil disobedience during the civil-rights struggle, and others were later arrested in protest of the Vietnam War.
While it is possible that a search of criminal records might disclose information that could reasonably be thought to have a negative bearing on a particular candidate’s suitability for a faculty position, such a discovery must surely be rare. At the same time, the moral cost of adopting a general policy of requiring such searches in order to identify the rare special case is great. Undertaking such searches is highly invasive of an applicant’s privacy and potentially very damaging. The probative value of criminal records is often small, because such records are notoriously imprecise. They contain information that ranges from arrest to indictment, from conviction to deferred prosecution and deferred sentencing, from completion of probation to the expunging of conviction. Context is often all important to understand this information, and context is never supplied. Casting a wide net in order to acquire such information in the hope that it might turn up matters of relevance to an appointment that is under consideration would risk great damage for small and speculative gain, and it would cede unacceptable discretion to those entrusted with assessing the significance of this information in the absence of a proper context. It also raises possible legal concerns about the exercise of due care in the secure maintenance and deployment of this information. Such concerns led the Privacy Protection Study Commission to recommend that only criminal-justice information “directly relevant to a specific employment decision” be acquired by employers.4
We agree. We recognize that there may be instances in which the nature of a particular appointment might justify an investigation of the criminal record of an applicant. For example, some states require criminal checks of persons to be employed in specific capacities, most notably in child care. But we conclude that for an ordinary faculty appointment, the likely benefits of a background criminal investigation of an applicant are dwarfed by the grave invasions of privacy caused by such investigations, as well as by the great potential of such investigations to facilitate the misuse of sensitive information. Such investigations are accordingly inconsistent with the principle of proportionality.
To enforce the general principle that the scope of a background check be limited by the specific requirements of a particular position, we recommend that whenever a college or university contemplates pursuing background checks that are more extensive than those now customary within the academy, it should specify the information it seeks and explain the reasons it believes such information is necessary. We hope this requirement of transparency will provoke discussion that will help to ensure that academic institutions do not unduly and unnecessarily expand the scope of background checks.
Concerns for Fairness and Accuracy
We do not in this report assess the procedures that accompany the forms of reference checking that are now commonly applied to candidates for ordinary faculty positions. We instead address the very different concerns for fairness and accuracy that arise when institutions of higher education seek to conduct extensive background investigations of sensitive information, including criminal records. Positive law already offers useful guidance on the question of proper procedures. The federal Fair Credit Reporting Act governs the procedures that employers must use when they retain businesses to conduct background checks. In summary: (a) the candidate must be informed of the proposed background check and authorize it in writing; (b) the candidate must be given a copy of the final report; and (c) no adverse action may be taken on the basis of the report unless and until the prospective employee has had an opportunity to contest or clarify its accuracy. There is much technical complexity in this law that we find unnecessary to consider. Suffice it to say, we recommend that these basic elements be adopted as governing principles whether a background investigation is conducted by a third party or by an academic institution itself.
Following a search, records should be discarded except with regard to successful candidates. If the report is to remain in a candidate’s file, it should be corrected to remove all inaccuracies. All irrelevant personally identifiable information in a faculty member’s file should be destroyed; relevant but harmful information should be kept confidential, perhaps in files segregated on that basis (if state law allows), and kept for a period no longer than justified by institutional need.
1. These limits are set out in Matthew Finkin, Privacy in Employment Law, 2nd ed. (Washington, D.C.: BNA Books, 2003), chap. 4. Back to text.
2. This assertion is based on the subcommittee’s request for a review of the prevalence of cases of misrepresentation or wrongful nonrepresentation in the Association’s files. The staff reported a case of what today would be called “identity theft,” an episode occurring more than two decades ago, that would have been headed off by more careful screening by normal means. Occasionally, such cases have reached the courts. For example, Fuller v. DePaul University, 12 N.E.2d 213 (Ill. App. 1938) in which a professor at a Catholic university had concealed the fact that he had been a priest and was dismissed, and In Re HadziAntich, 497 A.2d 1062 (D.C. App. 1985) in which a former law professor who falsified his resume in applying for a teaching position was subject to sanction by the bar. Back to text
3. U.S. Privacy Protection Study Commission, Personal Privacy in an Information Society (Washington, D.C., 1977), chap. 6. Back to text
4. Personal Privacy, 246. Back to text