Students at Penn High School1 brought this suit against the Penn- Harris-Madison School Corporation ("PHM" or "the School") for violating their Fourth Amendment rights against unreasonable searches and seizures. This claim arises from PHM's policy that allows for random, suspicionless drug testing of students involved in extracurricular activities and of students driving to school. The district court granted summary judgment for the School on both issues based on this circuit's precedent of Todd v. Rush County Schools, 133 F.3d 984 (7th Cir.), cert. denied, 525 U.S. 824 (1998), in which the court allowed suspicionless drug testing of students participating in extracurricular activities. The plaintiffs appealed.

* BACKGROUND

In 1998, PHM instituted a drug testing policy for its students. The policy, School Board Policy 360: Student Testing for Drugs, Alcohol and Tobacco, explains its purpose as follows:

The use of tobacco, alcohol and illegal drugs presents a threat to the safety, health and welfare of both our employees and our students. Because of the risks associated with such abuse, the board is implementing a student testing program for drugs, alcohol and tobacco. R.21, Ex.1 at 1. The policy focuses on five groups of students for drug testing and defines the groups as follows:

1. All students that participate in extracurricular activities. Activities will include all athletic teams, music groups, academic competitions, clubs and organizations. A full listing of activities will be provided. These students will be part of a pool of students that will be randomly selected for testing.

2. All students who drive to school. These students will also be part of the random pool.

3. All students and staff who volunteer to be part of the random pool.

4. All students who are suspended from school for three consecutive days for student misconduct or substantial disobedience. These students must submit to a drug test before being allowed to return to school.

5. All students for which there is a reasonable suspicion of being under the influence of drugs or alcohol must submit to a mandatory test.

Id. Members of the first two groups are the Plaintiffs-Appellants in this case.2

In its policy, the School states that extracurricular activities are a privilege, not a right. Also, the School explains, students participating in those activities assume greater responsibility and make certain sacrifices. These students, the policy states, are required to submit to random testing for drugs, alcohol, and tobacco.

All students in extracurricular activities must attend at least one drug education session before beginning the activity, and all students in extracurricular activities will receive a copy of the policy. Also, each participant shall sign and return a consent form that allows the School to conduct the drug testing. The consent form must be signed by the student and by a parent or guardian and must be returned to the School prior to the student's participation in the extracurricular activity. Failure to return the consent form results in nonparticipation in the activity.

Next, PHM's policy discusses student drivers. To receive a permit to park on school grounds, a student must pay $15.00 and provide proof of a valid driver's license. The policy explains that these students and their passengers are at a substantial risk for injury when operating vehicles under the influence of intoxicants. Also, according to the policy, "Studies indicate that young drivers have a greater risk of being involved in vehicular accidents caused by consumption of intoxicants." R.21, Ex.1 at 2. PHM partly based its determination to test student drivers on a newspaper article detailing a serious car accident involving two Penn High School students who had been drinking. The other basis for its policy for student drivers was several articles about high school students under the influence of alcohol who were involved in car accidents. Based on this data and the important policy interest of protecting student drivers and their passengers, PHM requires that students driving to and from school and other activities sponsored by PHM "must submit to the same random urinalysis as participants in athletics or extracurricular activities." Id. at 2-3. That means that students driving to school are subject to random testing for the presence of drugs, alcohol, and tobacco. Student drivers also must sign the consent form before receiving a parking permit, and, presumably, a student who does not return the consent form will not receive a parking permit.3

Students who refuse to take the drug test are deemed to have admitted they are under the influence of drugs or alcohol, which is a violation of school rules. The student "will be dealt with according to the student discipline policy or student extra-curricular code of conduct." R.21, Ex.1 at 5. Similarly, a positive test result will validate usage, and the consequences of validated usage "will coincide with the consequences outlined in the student handbook and school policy." R.21, Ex.1 at 5. What consequence applies is ambiguous. The section in the student handbook entitled "Policy on Alcohol and Other Drugs" discusses protecting students and prohibiting drug use. A student under the influence of alcohol or other drugs on school grounds or at school sponsored activities will be disciplined, and the discipline could result in suspension or expulsion.4 Use or possession of tobacco products is not permitted on school grounds and discovery of these will result in suspension from school.5

Another section, the section pertaining to student athletics and activities, also discusses the use or possession of drugs and alcohol. For drug or alcohol use, the first in-season offense will result in expulsion from the team or activity. The first out-of-season offense will result in a meeting that reviews the penalty and the guideline for future participation in the activity. A second violation, in-season or out- of-season, will result in expulsion from participation in all athletics and activities for one school calender year. For the use of tobacco, in any form, the first in-season offense will result in probation for one school calender year for all athletics or activities in which the student might participate. Also, the student will be subject to any penalties given by the office of student affairs. The second in-season offense will result in expulsion from the team for the remainder of the season.

The section governing student drivers only states that "Abuse of a student's driving privileges will result in the forfeiture of the parking permit and may result in further school discipline." R.29, Ex.C at 3. The consequences for a student driver over 18 whose test results reveal the presence of nicotine are not mentioned.

David Wade Risner, Director of Pupil Personnel at PHM, developed and implemented the drug testing policy. In his deposition, he averred that students participating in extracurricular activities and student drivers, if receiving a positive test result, may be subject to exclusion from any extracurricular activities and/or to revocation of parking privileges. However, he claims that they will not be subject to suspension, expulsion, or any discipline in connection with the academic school day for receiving a positive test result. See R.37 at 2.

The policy also discusses the testing procedure. Students are selected on a random basis without advance notice. First, they fill out paperwork and list any over-the-counter or prescription medications that they are taking. Next, they remove all outer garments and leave all bags and purses outside the collection facility. Then they are asked to wash their hands with water only, and the collector checks the stall visually for anything unusual, flushes the toilet, and treats the water with dye. While the student is producing the sample, the collector remains outside the facility and notes any unusual circumstance, behavior, or appearance of the student or of the specimen. Also, the collector checks the specimen for signs of contamination and notes the temperature of the bottle. Both the student and the collector sign the chain-of- custody form.

The test checks for the presence of alcohol, nicotine, and any drug listed as a controlled substance. The results will be provided to the designated school official who always shares the results with the student's parents. For the first and second positive test, the result is shared with the parents and an attempt is made to provide evaluation and/or treatment. For the third positive test, the parents are required to pay for the test. Again, the results are shared with the parents and the school, and an attempt is made to provide evaluation and/or treatment.

As to confidentiality, the policy states that the results will always be shared with the student and a parent or guardian. Thereafter, information regarding the positive test result will be shared on a "need to know" basis with school staff. Under the program, any staff member, employee, coach, or sponsor of PHM with knowledge of a student's positive test result shall not reveal the information to anyone other than the student or the parents unless under order of a court.

As mentioned above, several PHM students filed suit against the School. In their suit, they alleged that the School's suspicionless drug testing of students involved in extracurricular activities and of student drivers violated their Fourth Amendment rights against unreasonable searches and seizures.

In response to the School's motion for summary judgment, the district court followed this circuit's opinion in Todd v. Rush County Schools, 133 F.3d 984 (7th Cir.), cert. denied, 525 U.S. 824 (1998), which allows random drug testing for students participating in extracurricular activities, and upheld the extracurricular activities prong of PHM's policy.

For the students who possessed parking passes and who were subject to the random drug testing, the district court stated that, except for those students close enough to school to walk and "a limited category of students otherwise," PHM provides public transportation. Therefore, students do not have a compulsion to drive themselves to school. The court explained that students sign a consent form in exchange for the privilege of parking on school premises and that the safety issues evolving from students driving to and from school while under the influence of illegal substances justifies the testing.6

The Fourth Amendment to the Constitution protects individuals from unreasonable searches and seizures by the government.7 The Supreme Court has held that the Fourth Amendment protects students from unreasonable searches and seizures by school officials. See New Jersey v. T.L.O., 469 U.S. 325, 341-42 (1985). The Court also has held that random drug testing through urinalysis constitutes a search and seizure within the meaning of the Fourth Amendment. See Skinner v. Railway Labor Executives' Ass'n, 489 U.S. 602, 617-18 (1989); see also Chandler v. Miller, 520 U.S. 305, 313 (1997); Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 652 (1995); National Treasury Employees Union v. Von Raab, 489 U.S. 656, 665 (1989); Willis v. Anderson Community Sch. Corp., 158 F.3d 415, 417 (7th Cir. 1998), cert. denied, 526 U.S. 1019 (1999).

Under the Fourth Amendment, a search usually is not reasonable unless the government obtains a warrant issued upon probable cause; there are, however, certain limited exceptions. See Skinner, 489 U.S. at 619. To be a reasonable search without a warrant and probable cause, the government must show a "special need," beyond the normal need for law enforcement, that makes the warrant and probable cause requirement impracticable. See id. The Supreme Court in Skinner explained that, when such a special need exists, courts should "balance the governmental and privacy interests to assess the practicality of the warrant and probable-cause requirements in the particular context." Id.

The Supreme Court has "found such 'special needs' to exist in the public school context," because "the warrant requirement 'would unduly interfere with the maintenance of the swift and informal disciplinary procedures [that are] needed,' and 'strict adherence to the requirement that searches be based on probable cause' would undercut 'the substantial need of teachers and administrators for freedom to maintain order in the schools.'" Vernonia, 515 U.S. at 653 (quoting T.L.O., 469 U.S. at 340, 341). This finding of a "special need" in T.L.O. meant that school officials could justify a search of a student upon reasonable and individualized suspicion "that the search [would] turn up evidence that the student has violated or is violating either the law or the rules of the school." T.L.O., 469 U.S. at 342. In Vernonia, the Court found a special need in preventing student athletes from using drugs and upheld the legitimacy of suspicionless drug testing of the athletes. The Court previously had upheld suspicionless searches and seizures, by drug testing, of railroad employees involved in train accidents, see Skinner, 489 U.S. at 634, and of federal customs officers who carry arms or who are involved in drug interdiction, see Von Raab, 489 U.S. at 677.

When the Court allows suspicionless drug testing based upon a special need, the Court engages in a balancing test between the "intrusion on the individual's Fourth Amendment interests" and the search's "promotion of legitimate governmental interests." Vernonia, 515 U.S. at 653 (quoting Skinner, 489 U.S. at 619). The factors to consider are: (1) the nature of the privacy interest upon which the search intrudes, see Vernonia, 515 U.S. at 654; (2) the character of the intrusion on the individual's privacy interest, see id. at 658; (3) the nature of the governmental concern at issue, see id. at 661; (4) the immediacy of the government's concern, see id. at 662; and (5) the efficacy of the particular means in addressing the problem, see id. at 663.

In Vernonia, the Court upheld random, suspicionless drug-testing of student athletes. The school had presented evidence of a sharp increase in drug use at the school, which resulted in an increase in disciplinary problems, and the district court had found that the athletes were the leaders of the drug culture. See id. at 648-49. The drug testing policy instituted by the school required all students wishing to participate in interscholastic sports to sign a consent form permitting random, suspicionless drug testing. Students' names were chosen at random to undergo the test, and, during the test, a monitor stayed in the restroom to listen for signs of tampering as the student produced the sample. See id. at 650. A student wishing to play on the football team refused to sign the consent form and filed suit against the school for equitable relief. See id. at 651.

The Court considered first the nature of the students' privacy interest. In doing so, it paid particular attention to the facts in the record relating to the specific students being tested. Minors, the Court explained, are subject to the control of their parents or guardians, and, when they attend school, they are temporarily in the custody of the school. See id. at 654-55. Furthermore, the Court stated, students routinely are required to undergo physical examinations and vaccinations. See id. at 656-57. The Court concluded that these facts show that students generally enjoy a lesser expectation of privacy than the public at-large. In the context of student athletes, the Court explained, the expectation of privacy is even less. Athletes undergo a state of communal undress when, daily, they change in a common locker room and shower in a community shower. See id. at 657. The athletes voluntarily subject themselves to regulation by signing up for the sport. They must submit to a preseason physical examination, obtain insurance coverage, sign an insurance waiver, maintain a minimum grade point average, and comply with rules on dress, training hours, and conduct. See id. Therefore, the Court held that "students who voluntarily participate in school athletics have reason to expect intrusions upon normal rights and privileges, including privacy." Id.

Next, the Court discussed the character of the intrusion. It focused on the manner in which the production of the urine sample is monitored. See id. at 658. Because having a monitor present to listen for sounds of tampering presented conditions nearly identical to conditions encountered in public restrooms, the Court found the intrusion to be negligible. The Court also noted that the information disclosed by the drug test about what the student had ingested or about the student's physical conditions was not disclosed to law enforcement personnel and was provided to only a limited number of school personnel. The limited dissemination, in turn, limited significantly the intrusion on the student's privacy interest. See id. Finally, the Court determined, the revelation of any medications the student was taking was not per se unreasonable nor was it a significant invasion of privacy. See id. at 659.

The Court then focused on "the nature and immediacy of the governmental concern at issue" and "the efficacy of this means for meeting it." Id. at 660. The Court determined first that the nature of the concern, deterring drug use by schoolchildren, was obviously important, especially given that "[s]chool years are the time when the physical, psychological, and addictive effects of drugs are most severe." Id. at 661. Also, the Court stated that drug use by students affects the whole student body because it disrupts the educational process. Finally, the Court explained that athletes are particularly subject to risk from drug use because of the physical harm to the individual or to his peers that may result from the use of drugs. See id. at 662. As the Court concluded: "Finally, it must not be lost sight of that this program is directed more narrowly to drug use by school athletes, where the risk of immediate physical harm to the drug user or those with whom he is playing his sport is particularly high." Id. (emphasis supplied). The Court detailed the psychological effects of drug use, including impairment of judgment, slower reaction time, and a lessening of the perception of pain, and the physical risks of drugs to athletes, such as increased heart rate, higher blood pressure, and a masking of normal fatigue. See id. These effects of drug use on student athletes created a significant governmental interest in deterring drug use by student athletes. Next, the Court determined that the immediacy of the school's concerns in that case was unquestionable given the finding by the district court that the student body was in a state of rebellion, fueled by alcohol and drug use, and that the school's athletes were the leaders of the drug culture. See id. at 662-63.

Finally, in discussing the efficacy of the means, the Court held that employing random drug testing to address the problem of rampant drug use by athletes ensured that the athletes did not use drugs. See id. Therefore, after weighing the individuals' Fourth Amendment interests against the governmental concern, the Court upheld the random, suspicionless drug testing of student athletes by the school. See id. at 665.

The other recent Supreme Court case to address suspicionless drug testing is Chandler v. Miller, 520 U.S. 305 (1997). In Chandler, the government attempted to require drug testing for all candidates for public office. The Court acknowledged that the character of the search was not intrusive, and, thus, if the government could show a special need for drug testing these individuals then the search was reasonable. See id. at 318. The Court examined first the immediacy of the governmental concern and noted that there was no demonstrated problem of drug abuse by the targeted group. See id. at 318-19. Although proof of a drug problem was not determinative, the Court explained that it would help clarify and substantiate the hazards of drug use in the particular context. See id. at 319. The Court distinguished the drug testing in Chandler from the allowance of suspicionless drug testing in Von Raab, which had allowed suspicionless drug testing of customs agents, even though there was no evidence of a demonstrated drug problem, because of the difficulty in subjecting the employees in Von Raab to day-to-day scrutiny. The Court stated that, in contrast to the customs agents, candidates for public office were subject to relentless scrutiny by their peers, the public, and the press. See id. at 321. Next, the Court discussed the efficacy of the means chosen and stated that requiring a candidate to schedule his own appointment for a drug screen did not identify well who was violating the anti-drug laws nor did it perform credibly as a deterrent. See id. at 319. Finally, the Court explained that candidates for public office were not required to engage in high-risk, safety-sensitive tasks that would justify suspicionless drug testing, i.e., no special need existed. See id. at 321-22.

[W]here the risk to public safety is substantial and real, blanket suspicionless searches calibrated to the risk may rank as "reasonable"-- for example, searches now routine at airports and at entrances to courts and other official buildings. But where, as in this case, public safety is not genuinely in jeopardy, the Fourth Amendment precludes the suspicionless search, no matter how conveniently arranged.

Id. at 323 (citation omitted).

The question whether random drug testing of high school students involved in extracurricular activities is an unreasonable search and seizure under the Fourth Amendment was presented to this court in Todd v. Rush County Schools, 133 F.3d 984 (7th Cir.), cert. denied, 525 U.S. 824 (1998).8 In Todd, the school presented evidence that cigarette and alcohol use was higher than the state average, although marijuana use was lower. Also, witnesses testified that drug use at the high school had been increasing and had caused a drowning of one student and a car crash involving other students. See id. at 985. The school's drug testing policy required all students desiring to participate in extracurricular activities or planning to obtain a parking permit to consent to random drug, alcohol, and tobacco testing. See id. at 984. Also, a student could be tested if school officials had a reasonable suspicion that the student was using drugs, alcohol, or tobacco. See id. at 985.

Without employing the methodology presented by the Supreme Court in Vernonia, the court in Todd stated that the reasons compelling random drug testing for student athletes applied to students participating in extracurricular activities because "[c]ertainly successful extracurricular activities require healthy students." Id. at 986. Also, the court stressed that, similar to athletics, extracurricular activities are a privilege and that students must voluntarily choose to participate. See id. Because students in extracurricular activities, like athletes, can assume leadership roles, the court determined that it was not unreasonable to subject those students to drug testing in exchange for enhanced prestige and status in the student community. See id. Finally, the court stated that the crux of the program was to protect the health of the students involved and to deter drug use. Therefore, the court concluded that the drug testing program was reasonable under the Fourth Amendment. See id. at 986-87.

This circuit, in Willis v. Anderson Community School Corp., 158 F.3d 415 (7th Cir. 1998), cert. denied, 526 U.S. 1019 (1999), clarified its holding in Todd. In Willis, the school implemented its policy because of growing disciplinary problems and because of the perception that drug and alcohol use had increased. See id. at 417. Before making its decision, the school district had reviewed the policies of other Indiana school districts, the results of tests administered pursuant to those policies, and literature discussing the connection between drug use and disruptive behavior. See id. Thereafter, the school instituted a drug testing policy that required drug testing for, among others, any student who was suspended from school for 3 or more days for fighting. See id. The student who brought the suit had been suspended for fighting but had refused to consent to the drug test. See id.

The court in Willis examined first whether the school had reasonable suspicion to believe that Willis was using drugs. The school's principal admitted that he had observed Willis immediately after the incident and that, at that time, he did not have reasonable suspicion to believe that Willis was under the influence of drugs or alcohol. The school claimed, however, that the fight itself constituted reasonable suspicion that Willis was using drugs. The court rejected this argument by stating that such a blanket rule would belie the concept of individualized suspicion, which requires a case-by-case determination. See id. at 418.

Next, the court asked whether the school had demonstrated a special need for suspicionless drug testing of students suspended for more than 3 days for fighting. See id. at 420. The court acknowledged the school's concerns justifying the search: deterring drug use, disciplining its students, and protecting the health of children. However, the court concentrated on whether a suspicion-based search was practical and stated that "it may be that when a suspicion-based search is workable, the needs of the government will never be strong enough to outweigh the privacy interests of the individual." Id. at 421.

The court found that the nature of the privacy interest of students suspended for fighting was similar to that of the student athletes in Vernonia because school children enjoy a lesser expectation of privacy than the general public. See id. However, the privacy interest differed in significant respects. First, there was no aspect of communal undress as in Vernonia, and, second, the students did not voluntarily choose to participate in the activity as in Vernonia and Todd. See id. at 422.

Next, the court considered the nature and immediacy of the drug problem. The court refused to allow mere deterrence to justify suspicionless drug testing because that would "sanction[ ] blanket testing of all children in public schools." Id. The court explained that the Supreme Court had had the opportunity to allow for suspicionless drug testing for all students in Vernonia but had refused to do so. See id. The court cautioned against dividing students into broad categories and drug testing on a category- by-category basis because then "all but the most withdrawn and uninvolved students [would] fall within a category that is subject to testing." Id. at 423. Although deterrence did not suffice, the court held that the nature and immediacy of the school's concern was sufficiently similar to the schools' concerns in Vernonia and Todd. The nature of the concern was narrowly targeted at the group of students the school perceived to be at risk for drug use, and the immediacy of the concern, although not as high as in Vernonia, was to reduce the perceived increase in drug use at the school. Therefore, the court found that the nature and immediacy of the school's concern was not meaningfully less than in Vernonia. See id.

Finally, the court turned to the efficacy of the policy in question. The court recognized that, in Vernonia, drug testing based on individualized suspicion presented substantial difficulties; however, no evidence of similar difficulties had been presented in Willis. Given that every student must meet with the principal prior to suspension, the court determined that requiring the school to find individualized suspicion of drug use before testing the student for drugs was both feasible and practical. See id. at 423-24. Therefore, based on the lack of efficacy of the means for addressing the problem and on the nature of the privacy interest involved, the court determined that the governmental concern did not outweigh the individuals' Fourth Amendment interests and, thus, that suspicionless drug testing of students suspended for more than 3 days for fighting was unreasonable under the Fourth Amendment. See id. at 424.

To justify its policy, PHM relies properly on this court's decision in Todd. We do not believe that the result in Todd is compelled by the Supreme Court's decision in Vernonia. Therefore, as we explain below, if we were reviewing this case based solely on Vernonia and Chandler, we would not sustain the random drug, alcohol, and nicotine testing of students seeking to participate in extracurricular activities.9 Nevertheless, we believe that the doctrines of stare decisis and precedent require our adherence to Todd, and we affirm the judgment under review on that basis. Also, on the basis of Vernonia and Chandler, we uphold the drug and alcohol testing- -but not the nicotine testing--of students desiring to drive to school.

Public high school students have a lesser expectation of privacy than the general public. However, students do not shed their constitutional rights at the schoolhouse door. See Tinker v. Des Moines Indep. Community Sch. Dist., 393 U.S. 503, 506 (1969). Similar to the students in Vernonia, PHM students are in the temporary custody of the School. Also, students in general are subject to routine physical examinations and vaccinations. However, unlike the athletes in Vernonia, PHM students who participate in extracurricular activities or who drive to school do not subject themselves to more explicit and routine loss of bodily privacy as a necessary component of their participating in the activities in question.

Indeed, unlike the student athletes in Vernonia, these students otherwise do not subject themselves, by virtue of their participation in these activities, to regulations that further reduce their expectation of privacy. For example, as the Supreme Court noted in Vernonia, the athletes in that case were required to submit to a physical examination before the beginning of the season. Also, the athletes needed to obtain insurance coverage and they agreed to abide by rules on conduct, dress, and training hours. Finally, they also expected a degree of "communal undress" not experienced by other public school students. See Vernonia, 515 U.S. at 657. Although PHM students in extracurricular activities, other than athletics, also volunteer to join a particular group and to subject themselves to the rules of that organization, those rules do not require the same surrender of physical privacy as required of the student athletes in Vernonia. In the case of students driving to school, the contrast is even more stark. Overall, the expectation of privacy for students in extracurricular activities or with parking permits, although less than the general public, is still greater than the expectation of privacy for athletes.

The Supreme Court in Vernonia held that the presence of a monitor in the bathroom who listened for signs of tampering was a comparable condition to the experience of a public restroom. Therefore, the Court held that the intrusion was negligible. See Vernonia, 515 U.S. at 658. As to the information disclosed by the test, the Court stated that the test only looked for drugs and not for physical conditions and that the test was standard: it checked for the same substances for all students. See id. Finally, the Court explained that the results were disclosed to only a limited number of school personnel who had a need to know and were not provided to law enforcement officials or used for any internal disciplinary function. See id. at 658-59. The combination of these factors, according to the Court, made the character of the intrusion minimal. See id. at 660.

The conditions at PHM parallel those in Vernonia; therefore, we conclude that the character of the intrusion is not overly invasive.

According to the Supreme Court's methodology in Vernonia, we should assess the government's interest from two perspectives--whether there is any correlation between the defined population and the abuse, and whether there is any correlation between the abuse and the government's interest in protecting life and property.

We turn first to whether there is any correlation between the defined student population and the abuse. Here, however, the School has not proven, or even attempted to prove, that a correlation exists between drug use and those who engage in extracurricular activities or drug use and those who drive to school. Indeed, at oral argument, counsel for PHM admitted that there is no correlation between students involved in extracurricular activities and drug abuse. Counsel also stated that student drivers do not differ from the general school population based on its statistical abstract. Finally, counsel conceded that the lack of such relationship distinguished this case from the facts in Vernonia in which the evidence demonstrated that the athletes were the leaders of the drug culture. Thus, counsel for PHM is admitting that, at least in this respect, the district is attempting to do what this court in Willis admonished against: dividing the students into broad categories and drug testing on a category-by-category basis, which allows for drug testing for all but the most uninvolved and isolated students. See Willis, 158 F.3d at 423. In fact, at oral argument, counsel announced that the goal is to test all students on a random, suspicionless basis.

We now turn to whether there is any correlation between the abuse and the government's interest in protecting life and property. We have no doubt that a legitimate and pressing need for drug and alcohol testing of students driving vehicles on school property stems from the ability of one student under the influence of drugs or alcohol to injure seriously another student. With the mass exit of students after classes into the relatively close confines of a student parking lot, one student under the influence of drugs or alcohol could cause serious injury or death.

On the other hand, the decision of PHM to test student drivers for the presence of nicotine is not so easily justified. Tobacco use is legal if a person is over 18 years of age. PHM's school policy validly prevents use of tobacco products on school grounds. However, if a student smokes at home, leaves the cigarettes at the house, drives to school, and is drug tested, the results would reveal the presence of nicotine. This student could be subject to sanctions under PHM's policy for a perfectly legal activity. In the absence of supporting data, this expansive view of the School's interest goes too far. Furthermore, PHM simply has not documented any serious risks associated with a student driving while using a tobacco product.

Finally, although PHM may have justified the risk of injury associated with student drivers under the influence of drugs or alcohol, PHM has not explained how drug use affects students in extracurricular activities differently than students in general.

4. The immediacy of the governmental concern

PHM presented evidence regarding its general student population that shows that, in some categories, especially for gateway drugs such as alcohol, nicotine, and marijuana, PHM's average use is greater than the national average. This situation would appear to justify PHM's taking action with respect to activities that pose a special risk of injury or death when drug or alcohol abuse is present. Just as such a case was established in Vernonia for athletic activities, it also can be established for driving in the relatively close confines of the school premises.

However, PHM has not shown the same possibility of immediate danger from students participating in extracurricular activities. PHM simply has not established that any immediate problem with drugs or alcohol exists for its students in extracurricular activities.

The Court in Chandler did not require proof of drug use by the candidates for public office but it stressed that such proof would help in finding a special need. In the circumstances here, we think that PHM was required to show a correlation between drug use and students in extracurricular activities, or other evidence of a particularized special need, before implementing its suspicionless drug testing policy for those particular student groups.

In Vernonia, the Court noted that it is difficult to use individualized suspicion to drug test a broad population of students, such as athletes. Read in isolation, this comment in Vernonia would permit a school district to implement a random program on a suspicionless basis as long as it would test a large subset of the entire school population. In Chandler, however, the Court stressed that suspicionless drug testing without evidence of a drug problem by the targeted group should not be used if suspicion-based drug testing is possible. See 520 U.S. at 321. We emphasized this restriction in Willis.

Here, there is no showing that the students subject to testing are the ones that must be tested to resolve the perceived problems. There simply is a lack of a correlation between drug use and either students in extracurricular activities or student drivers. Given the variety of circumstances under which students enter and exit school premises, it is reasonable to conclude that individualized suspicion of drug and alcohol use by student drivers is not feasible. It would be impossible for the school to determine whether each student driver was drug and alcohol free. However, PHM has made no showing that teachers, staff and sponsors of extracurricular activities would not be able to observe the students for suspicious behavior.

Accordingly, we conclude that PHM has demonstrated a sufficient government need to overcome the students' Fourth Amendment rights and to administer random drug testing to students who wish to drive on school property. The danger is well-defined, and the efficacy of testing on individualized suspicion is hardly an adequate preventive measure against the possibility of real and immediate injury. On the other hand, with respect to testing student drivers for nicotine, PHM has not demonstrated a sufficient government need to justify this intrusion. With respect to random testing of those who participate in extracurricular activities, we believe that, according to the methodology employed by the Supreme Court in Vernonia, there has been an inadequate showing that such an intrusion is justified. However, as we discuss in the paragraphs that follow, another consideration--stare decisis and precedent-- dictates a contrary result.

"Stare decisis is the preferred course because it promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process." Payne v. Tennessee, 501 U.S. 808, 827 (1991). However, the Supreme Court has stated that the doctrine of stare decisis has less force in the constitutional context because the interpretation may be altered only by a constitutional amendment or by overruling precedent. See Agostini v. Felton, 521 U.S. 203, 235 (1997). Therefore, the Court has overruled prior decisions concerning constitutional questions when there has been "a significant change in, or subsequent development of, our constitutional law." Id. at 236. In Agostini, the Establishment Clause jurisprudence underlying the case had changed significantly over the 12 years since the Court had first decided the constitutional question at issue. Therefore, the Court overruled its prior decision to the extent that it contradicted the intervening doctrine. See id.10

As the previous sections make clear, the judges of this panel believe that students involved in extracurricular activities should not be subject to random, suspicionless drug testing as a condition of participation in the activity. Nevertheless, we are bound by this court's recent precedent in Todd. Given that the opinion in Todd was issued only two years ago, that the facts of our case do not differ substantially from the facts in Todd, that the court in Willis reaffirmed the basic principles in Todd, and that the governing Supreme Court precedent has yet to address the matter, we believe that we must adhere to the holding in Todd and affirm the district court's grant of summary judgment for the School as it relates to testing students involved in extracurricular activities.

However, we caution against reading the opinion in Todd too broadly. At oral argument, counsel for PHM expressed the desire to use our holding in this case as a transition toward allowing suspicionless testing of all students. Counsel admitted that drug testing the entire student population on a suspicionless basis was the ultimate goal. After the School conceded that no correlation existed between drug or alcohol use and extracurricular activities, the panel expressed concern about this slippery slope. The relevant dialogue unfolded as follows:

THE COURT: So the slippery slope argument ought to be very much in our minds. I mean, you'll be back here in another year with another school district who wants to test everybody. And you will say there is no principled distinction between the holding you get today and the next case. It's just a matter of time till it gets here. Right?

COUNSEL: Absolutely, your honor.

Thereafter, counsel attempted to backtrack and to stress the importance of voluntariness in the discussion because students are consenting to drug testing in exchange for a privilege. However, later in oral argument, counsel again agreed that schools should be allowed to drug test everybody. If schools tested all students on a suspicionless basis, the element of voluntariness obviously would not be present.

The danger of the slippery slope continues to haunt our jurisprudence.11

The scope of Vernonia remains undecided today. Until we receive further guidance from the Supreme Court, we shall stand by our admonishment in Willis that the special needs exception must be justified according to the methodology set forth in Vernonia. Under that approach, the case has yet to be made that a urine sample can be the "tuition" at a public school.

On the basis of the doctrines of stare decisis and precedent, we are constrained to affirm the judgment of the district court insofar as it permits the use of random drug testing of students who desire to engage in extracurricular activities. We also affirm the judgment insofar as it permits the random testing of student drivers for drugs and alcohol. We reverse the judgment of the district court insofar as it sanctions the random testing of student drivers for nicotine. The Defendants-Appellees may recover their costs.