Legal Implications 1 LEGAL IMPLICATIONS OF PROFILING STUDENTS FOR VIOLENCE

Predicting violent activity with a youth violence profile in schools raises a host of legal concerns focusing on the validity and use of profiles as social science evidence: the impact of potential discrimination, search and seizure, and the implications for privacy. Schools differ from airports or other settings where profiles are used. Profiles are useful if they properly establish reasonable suspicion to stop an individual. They raise more problematic constitutional issues when they support referring a student to alternative educational services. Where a profile identifies students based on race, gender, or proxies for these characteristics, it is invalid. It can also be invalid if it is not a reasonable method of achieving the government’s interest in safe schools. In addition, profiles present significant problems for confidentiality. These issues rest on fundamental concerns about the general validity of profiles as scientific tools: their objectivity, accuracy, sensitivity, over-inclusiveness, and general scientific acceptance. Consequently, the use of profiles in a school setting is highly problematic and controversial. ∗ The author wishes to thank Laird Kirkpatrick, U.S. Department of Justice, and many members of the staff of the Hamilton Fish Institute, including Dr. Paul Kingery, Dr. James Derzon, Nancy Budd, Aaron Alford and Janet Humphrey; for their contributions and suggestions for this article. The strength of this article is the result of their clarifying arguments, concise editorial comments, and significant intellectual curiosity; while any deficiencies remain those of the author. The Hamilton Fish Institute is administered by The George Washington University Institute for Education Policy Studies, Graduate School of Education and Human Development. Prepared under a grant from the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, U.S. Department of Justice (97-MU-FX-K012), points of view or opinions in this document are those of the Institute and do not necessarily represent the official position or policies of the U.S. Department of Justice. Legal Implications 2 Introduction In April 2000, Secretary of the Department of Education, Richard W. Riley, announced his opposition to the use of behavioral profiling systems to identify potentially violent students by schools, in a speech to school counselors in Chicago (Cooper, 2000). Secretary Riley went on to state “[w]e simply cannot put student behaviors into a formula to come up with an appropriate response.” The Secretary’s comments are some of the latest in a long-standing debate over the use of violence prediction tools to identify potentially violent students before they unleash their carnage on an unsuspecting school or community. On the other side of the issue, parent groups, students and law enforcement who see the daily press reports of school violence, teen shootings and similar tragedies throughout America are calling for effective ways to prevent school violence. (A full discussion of school violence trends may be found in the article by Dr. Paul Kingery and Mark Coggeshall in this special issue). In light of the level of violence affecting America’s youth, it only seems natural that as each new school shooting grips the public’s attention, parents, teachers, school officials and policy-makers would demand a profile of school shooters in order to identify the source of potential violence before it erupts. Typically, a profile identifies a person likely to commit an act (usually criminal) of a certain type. These types of profiles are developed from extensive research and analysis of the crimes and behaviors of actual perpetrators: terrorists, drug couriers, and assassins, for example. The assembled information is compiled to identify the key characteristics of each type of crime and the corresponding characteristics of the type of person most likely to commit that type of crime. Accordingly, they are used to assess all persons in a given location or under a particular to set of circumstances to determine if they match the profile and represent a potential threat to the safety of others. A youth violence profile would most closely resemble the dangerous passenger profile used by the Federal Aviation Administration to identify possible drug couriers and terrorists in the airport setting. First, both are based on the methodology just described. Second, both are Legal Implications 3 utilized in specific, highly insular environments; namely an airport or a school. Third, they identify potentially violent individuals from a large population of highly similar individuals most of whom are engaged in innocent behavior. Finally, the assessment based on the profile is done under significant time constraints; time is of the essence in deciding on a profile match before a person boards a plane, or enters a school. Predicting violent activity in the school context raises a host of legal and ethical concerns. These concerns center on the validity (meaning the profile’s ability to measure what it purports to measure and conformance to accepted scientific standards), and use of profiles as social science evidence, the impact of potential discrimination, search and seizure; and implications for privacy and the use of student records. The use of profiles as social science evidence will be addressed in the initial section of this article regarding General Validity, while the search and seizure aspects of profile usage are addressed under the second section, Profiles as Grounds for Search and Seizure. The third section, Constitutional Issues: Access to Education Services based on Profiles, addresses constitutional issues involving referral of students to alternative education based on profile matches. The implications for student privacy are discussed in the fourth section on Privacy and Educational Records. This article attempts to outline the breadth of these issues without suggesting concrete answers. The development of the law and ethics in this area is too new to allow for statements of definitive guidelines, and true to form, a legal article is long on problems and short on solutions. In truth, this is less a testament to the limitations of legal analysis, than it is a recognition of the need to engage in the thoroughly democratic process of assembling potential rules of conduct and guidelines for profiles in a comprehensive and collaborative fashion involving all the disciplines of science, medicine, health and law. Hopefully, this special issue will contribute to the creation of just such a shared understanding. Legal Implications 4 General Validity: Objectivity, Over-Inclusiveness, Evidentiary Issues The general validity of a profile depends on the circumstances of its use. There is no governing legal rule or case involving a youth violence profile or the use of a profile in a school setting. As discussed in the introduction, however, profiles have been used in a number of other settings and their validity assessed by the Supreme Court and a number of lower federal courts. The courts have avoided ruling on broad cases involving profiling instead focusing on the specific facts of individual cases. Based on this body of law, it may be said that where a profile is used as an investigation tool to assist the determination of the need to stop and search a suspect, it is generally regarded as a valid practice (U.S. v. Lopez, 1971). In the case of U.S. v. Riggs, the court upheld stopping a suspect for identification and inquiry purposes because the Federal Aviation Administration’s “behavioral profile” gave security “reasonable and objectively articulable grounds to suspect Riggs of both trafficking in narcotics and also posing a threat to aircraft security” (U.S. v. Riggs, 1972). The contours of the use of profiles as an investigation approach are explored more deeply in the next section on search and seizure, but it is important to consider the court’s reference to objectivity before leaving the matter for later discussion.