BEAM, Circuit Judge.

Appellants, members of the Linn State Technical College1 Board of Regents acting in their official capacities, and Donald M. Claycomb, President of Linn State Technical College ("Linn State" or "the College"), appeal the district court's grant of a permanent injunction and subsequent grant of attorneys' fees in favor of Appellees. We reverse.

I. BACKGROUND

This matter is before us a second time. In Barrett v. Claycomb, 705 F.3d 315 (8th Cir.2013) ("Barrett"") a panel of this court reviewed an interlocutory appeal, discussing, and ultimately reversing, the grant of a preliminary injunction in favor of Michael Barrett, IV, and other named individuals (collectively, Appellees) on their facial challenge to the drug-testing policy at issue. Id. at 325. Upon remand, Appellees clarified their claims to assert an "as-applied" challenge to the very same policy. Reviewing the as-applied challenge, the district court, in part, permanently enjoined Appellants from conducting any further collection, testing, or reporting the results of any testing of urine specimens from any plaintiffs who were not, are not, or will not be enrolled in five enumerated programs at Linn State, discussed in more detail herein. This resolution of the as-applied challenge is now at issue. Because the factual recitation from Barrett is brief, and because its reiteration is necessary to lay out the background and especially the drug-testing policy at issue once again, we liberally adopt the facts of this case as previously explicated. Id. at 318-20.

Linn State is a two-year, technical college located in Linn, Missouri. Linn State offers approximately thirty programs for a relatively small student body comprised of roughly 1150 to 1200 students. On average, 500 new students begin programs at Linn State each year seeking certificates, diplomas, and applied science associate degrees, or a combination thereof. Mo. Ann. Stat. § 178.636(2). Linn State does not offer associate of arts or baccalaureate or higher degrees. Id. Established by statute, Linn State is unique in that its purpose is to "make available to students from all areas of the state exceptional educational opportunities through highly specialized and advanced technical education and training at the certificate and associate degree level in both emerging and traditional technologies with particular emphasis on technical and vocational programs not commonly offered by community colleges or area vocational technical schools." Id. at § 178.636(1).

The programs Linn State offers can be divided into four primary categories: mechanical, electrical, civil, and computer. Each of these primary categories has further specialty areas. Most programs offered at Linn State involve manual exercises. The goal of the institution as stated in its admissions materials is to provide 75% of the class work in the field chosen by the student. For example, students in the Aviation Maintenance program spend roughly 62% of their time doing hands-on training, where students work in close proximity to active propeller blades. These students are also required to taxi airplanes. Students seeking accreditation in the Heavy Equipment Operations program spend between 51% and 72% of their time engaged in hands-on training, involving operation of Caterpillar D6R bulldozers and other heavy equipment weighing up to twenty-five tons. Students in the Industrial Electricity program spend about half their time engaged in hands-on functions, receiving training with live electricity and, at times, performing electrical services for members of the community.

On June 17, 2011, Linn State's Board of Regents adopted a mandatory drug-screening policy. The policy states:

The testing policy indicates that "[t]he purpose of the program is to provide a safe, healthy and productive environment for everyone who learns and works at Linn State Technical College by detecting, preventing and deterring drug use and abuse among students." The testing procedures provide that the test results do not serve law enforcement purposes and will not be revealed to law enforcement personnel.

As a condition of admission to Linn State in the fall 2011 semester, students were required to sign a form acknowledging the new drug-testing policy and also acknowledging that refusing to screen would result in administrative or student-initiated withdrawal. The condition of admission also explained to students that if a test returned positive, the student would have 45 days "to rescreen and test negative to remain enrolled." Approximately 550 students paid a $50 fee for the drug test that fall and were tested.

In conjunction with the new policy, on September 6, 2011, Linn State issued a series of procedures by which it would conduct the drug screening. The written procedures provided that students could "petition the Office of the President for a waiver of the general requirement to participate in the Drug Screening Program." According to the procedures, "[t]he student may advance any justification for the request." If a student filed a petition, President Claycomb testified that he would consider the student's reason and consult other personnel at the College, and possibly legal counsel, before he rendered a decision. There was also a full appeal and hearing process available for students wishing to challenge the initial determination. On September 7, 2011, Linn State began drug testing students.

On September 14, after providing urine samples in accordance with the drug-testing policy, Appellees commenced action on behalf of an enumerated class2 against members of the Board of Regents and President Claycomb. The complaint alleged that Linn State's drug-testing policy constituted a search that violated the Fourth Amendment.3 Appellees sought a declaration that the drug-testing policy was facially unconstitutional and further sought injunctive relief. The district court issued a preliminary injunction and we reversed, because we were unable to hold that the drug-testing policy is unconstitutional on its face in every conceivable circumstance. Barrett, 705 F.3d at 320-21, 321 n. 4, and 324-25 (reiterating that in order to receive injunctive relief, no matter whether the court applied a "likelihood of success on the merits" or a "fair chance of prevailing" standard, the appellees could not satisfy their ultimate burden in mounting a facial challenge under the Fourth Amendment that no set of circumstances existed under which the policy would be valid.).

Upon remand, Appellees clarified that they sought as-applied relief and the district court analyzed that claim. When analyzing Appellees' as-applied challenge, the district court conducted a program-by-program analysis to "ensure that the category of students subject to the drug-testing policy has not been defined more broadly than necessary to meet the policy's purposes." In doing so, the court stated it balanced the special need advanced by Linn State (which the court limited to safety concerns for others) with the privacy expectations intruded on, to discern whether Linn State's drug testing (i.e., the search) was reasonable in each instance. The district court adjusted the level of the students' expectation of privacy in each balancing analysis depending upon whether or not the student would be entering professions in heavily regulated industries or industries where drug testing was the norm in the future. If the students were, the court took into account a diminished privacy expectation. If they were not, the district court conducted its balancing analysis assuming the students had full privacy expectations common to all adults.

Conducting its analysis, the district court determined that Linn State could reasonably conduct drug testing in the following program areas: Aviation Maintenance, Industrial Electricity, Electrical Distribution Systems, Power Sports, and CAT Dealer Service Technician. However, the court held that it was unconstitutional for Linn State to drug test students participating in the following programs: Auto Body; Auto Mechanics; Heavy Equipment Technology; Medium/Heavy Truck Technology; Electronics Engineering Technology; Electrical Power Generation; Heating, Ventilation and Air Conditioning; Commercial Turf and Grounds Management, Machine Tool Technology; Computer Programming; Construction and Civil Technology; Networking Systems Technology; Design Drafting and the remainder of Linn State's approximate twenty-eight distinct academic programs. The drug testing in the Heavy Equipment Operations and Commercial Driver's License programs is not at issue in this case, as those students are subject to a separate drug-testing requirement, which the district court's ruling did not affect. Linn State appeals.

II. DISCUSSION

A. Standard of Review

We review the district court's issuance of a permanent injunction for an abuse of discretion, Capitol Records, Inc. v. Thomas-Rasset, 692 F.3d 899, 906 (8th Cir.2012), but where, as here, "the determinative question is purely legal, our review is more accurately characterized as de novo." Qwest Corp. v. Scott, 380 F.3d 367, 370 (8th Cir.2004). "Abuse of discretion occurs if the district court reaches its conclusion by applying erroneous legal principles or relying on clearly erroneous factual findings." Fogie v. THORN Americas, Inc., 95 F.3d 645, 649 (8th Cir.1996).

B. Legal Standard

In Barrett, we clearly delineated that the suspicionless drug testing at issue in this case constitutes a search subject to the demands of the Fourth Amendment.4 Barrett, 705 F.3d at 321-22. "As the text of the Fourth Amendment indicates, the ultimate measure of the constitutionality of a governmental search is `reasonableness.'" Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 652, 115 S.Ct. 2386, 132 L.Ed.2d 564 (1995). Ultimately, "whether a particular search meets the reasonableness standard is judged by balancing its intrusion on the individual's Fourth Amendment interests against its promotion of legitimate governmental interests." Id. 515 U.S. at 652-53, 115 S.Ct. 2386 (internal quotation omitted). Relevant here, a search unsupported by probable cause can be constitutional where the intrusion serves special governmental needs, beyond the normal need for law enforcement. Nat'l Treasury Emps. Union v. Von Raab, 489 U.S. 656, 665, 109 S.Ct. 1384, 103 L.Ed.2d 685 (1989). Once a special need is identified sufficient to depart from the usual warrant and probable-cause requirements that so often subsume the Fourth Amendment analyses, the requisite analysis involves the balancing of that need, with the students' privacy interest. Barrett, 705 F.3d at 321-22; see also Chandler v. Miller, 520 U.S. 305, 314, 117 S.Ct. 1295, 137 L.Ed.2d 513 (1997); Skinner v. Ry. Labor Execs.' Ass'n, 489 U.S. 602, 619, 109 S.Ct. 1402, 103 L.Ed.2d 639 (1989).

In general, to accomplish this balancing, courts weigh the interference with individual liberty that results from the particular search, against the special need or interest advanced by the government in support of such action. Barrett, 705 F.3d at 321-22. In all instances, three factors guide a court's analysis:

Id. at 322 (internal quotations omitted); see also Bd. of Educ. of Indep. Sch. Dist. No. 92 of Pottawatomie Cnty. v. Earls, 536 U.S. 822, 830, 832, 834, 122 S.Ct. 2559, 153 L.Ed.2d 735 (2002); Vernonia, 515 U.S. at 654-64, 115 S.Ct. 2386.

C. Analysis

Suspicionless drug testing falls into a "closely guarded category of constitutionally permissible suspicionless searches." Chandler, 520 U.S. at 309, 117 S.Ct. 1295. Broadly speaking there are two, relevant, interrelated lines of Supreme Court cases addressing suspicionless drug testing that inform our analysis: (1) those addressing suspicionless testing of adults participating in closely regulated industries or working in "safety-sensitive" positions, see, e.g., Skinner, 489 U.S. at 620, 109 S.Ct. 1402 (railroad employees); Von Raab, 489 U.S. at 668-70, 109 S.Ct. 1384 (customs officials);5 and (2) suspicionless drug testing of students in the educational setting, Vernonia, 515 U.S. at 648-56, 115 S.Ct. 2386 (determining that suspicionless testing of athletes was constitutional); Earls, 536 U.S. at 826-28, 122 S.Ct. 2559 (same for testing of students who participated in competitive extracurricular activities). The current matter is a hybrid of the two lines of cases, so while they are informative they are not wholly dispositive and ultimately do not alter the analysis that must be conducted-in each and every instance, these queries are conducted on a case-by-case, context-specific basis. Chandler, 520 U.S. at 314, 117 S.Ct. 1295; see also Earls, 536 U.S. at 830, 122 S.Ct. 2559 (reiterating that courts must conduct a fact-specific balancing on a case-by-case basis).

In Barrett, based on the evidence presented in support of Appellees' motion for a preliminary injunction on their facial challenge, we determined many points that remain relevant and buttress our instant analysis: (1) the public has a valid interest in deterring drug use among students engaged in programs posing significant safety risks to others, 705 F.3d at 322; (2) "some college students that attend Linn State have a diminished expectation of privacy because they are seeking accreditation in heavily regulated industries and industries where drug testing, in practice, is the norm," id. at 323; (3) Linn State's testing procedures significantly minimize the intrusiveness of Linn State's drug-screening program and are relatively non-invasive, thus the invasion of students' privacy is not significant, id.; and (4) the need to prevent and deter the substantial harm that can arise from a student under the influence of drugs while engaging in a safety-sensitive program provides the necessary immediacy for Linn State's testing policy, id.

In analyzing Appellees' as-applied challenge, the district court interpreted Barrett rigidly and conducted an exhaustive program-by-program analysis, permanently enjoining Linn State from administering its drug-testing program to students in specific, enumerated programs. In doing so, the district court erred on several fronts.

1. Special Need

The district court accurately held that Linn State bears the burden of demonstrating a special need sufficient to withstand the Fourth Amendment balancing test. It is well established that there are some searches, unsupported by probable cause that may be reasonable when special needs, beyond the need for law enforcement, make the warrant and probable cause requirement impracticable. Earls, 536 U.S. at 829, 122 S.Ct. 2559. Yet, the district court erred by taking its analysis further, embedding in Linn State's burden the requirement that it produce specific evidence, program by program, of a special need. In doing so, the court erroneously crafted a multi-part standard, or "parameters" as the district court stated, that the court gleaned from prior case law, addressing "indispensable" factors Linn State must identify and meet in order to advance a special need sufficient to conduct suspicionless drug testing of students in each program. We do not quibble with the district court's recitation of the various standards utilized by the Supreme Court in addressing the interests and needs advanced by the government in prior cases of this sort, but rather point out that the court unnecessarily pulled language from the prior analyses, applying each as cumulative and absolute standards that Linn State must meet. The court expended too much analytical energy pigeon-holing Linn State's interest into a nonexistent, one-size-fits-all legal rubric.

Citing this court's analysis in Barrett, the district court took pains to note that the only special need it would evaluate in its analysis was Linn State's interest in deterring drug use among students in programs posing significant safety risks to others. The court held that indulging any other need would promote the advancement of illusory safety concerns to mask unconstitutional purposes. We disagree.

Barrett did not so limit Linn State's special need as only an interest in deterring drug use among students in programs posing significant safety risks to others. This court discussed the risk to others as part of the special need analysis in the discussion of the public safety interest at play because harm to others necessarily is the most relevant factor in that particular consideration. Barrett, 705 F.3d at 322 (relying on Skinner's and Von Raab's consideration of the public's surpassing safety interest as a part of the special needs analysis). However, deterring harm to oneself is innate in all of the analyses of special governmental needs in similar circumstances when individuals are engaged in dangerous or safety-sensitive activities — and our discussion regarding harm to others did not limit that additional risk in any way. Id. at 323 (validating harm to one-self as part of Linn State's primary concern arising from students under the influence while engaging in a safety-sensitive program when addressing the nature and immediacy prong of the balancing analysis). For example, in Skinner, the Court acknowledged the axiomatic nature of the governmental interest in ensuring the safety of the public "and of the employees themselves." Skinner, 489 U.S. at 621, 109 S.Ct. 1402; see also Von Raab, 489 U.S. at 669-71, 109 S.Ct. 1384 (recognizing that not only is the physical safety of border employees themselves-those involved in drug interdiction and those who otherwise are required to carry firearms-threatened in that line of work but also the safety of others at large should those employees carry out their duties in an impaired state). Accordingly, the court erred in declining to include "harm to one-self" in its special need analysis.

As presented to the Board of Regents and as argued to the district court and on appeal, the purpose behind Linn State's drug-testing policy is to "provide a safe, healthy, and productive environment for everyone who learns and works at LSTC by detecting, preventing, and deterring drug use and abuse among students." This policy simultaneously advances the educational purpose served by fostering a drug-free environment at a technical school preparing students for the realities of the workplace environments they will soon be entering. These purposes-safety concerns juxtaposed with educational concerns-are not analytically distinct and establish a special need sufficient to support the balancing of interests necessary in these circumstances.

Using drugs while attending classes at a technical school uniquely limited to instruction and training in technical and vocational programs, where a large percentage of the students on campus are performing hands-on work in their respective, industrial programs on a daily basis, poses a unique safety risk that does not necessarily exist on other college campuses, or even at other, more similar, community colleges. The very nature of these programs and the unique vocational focus of the college itself involves dangerous aspects and creates safety risks for students under the influence of drugs or alcohol, as well as others.

The district court extensively evaluated many programs (although not all programs) to evaluate the level of safety risk. We see no justification, however, for the district court's decision to embark on such a tightly focused and burdensome trek in this case, even acknowledging the very critical constitutional right at play, given the very unique circumstances at this particular technical college. Even the differentiating nuances of risk between certain programs belabored by the district court were so slight. For example, on the one hand the district court found drug testing acceptable for students participating in the Power Sports program where students are exposed to on-road and off-road vehicles as well as hydraulic and air-type lifts, while it concurrently stated with certainty that the curriculum and risk exposure in the Commercial Turf and Grounds Management program was not enough to warrant testing even though those students used forklifts, large commercial mowers, flammable materials, propane torches and concrete saws. Unlike the district court, we do not find that expert testimony or a specific level of substantiating evidence is required to determine whether students under the influence operating, say, a forklift or a propane torch, pose a safety risk to themselves and others. They do. At bottom, on these facts and given the nature and inherent risks involved with the majority of Linn State's offerings, we need not go past the stated safety risks obviously at play in Linn State's programs and become curriculum and student-management experts.

Just as in the Title VII employment litigation context where we caution parties that this court does not sit as a super-personnel department in place of employers, Bone v. G4S Youth Servs., LLC, 686 F.3d 948, 955 (8th Cir.2012), we likewise cannot and do not operate as course-of-study-content experts discerning the relative safety issues in various programs offered at a technical school where significant safety risks abound. Courts are not, nor do we need to be, admissions and curriculum experts to conduct the requisite balancing of these matters in this case. What we do know, however, is that Linn State is a unique technical college, offering vocational programs to a relatively small student body, within a reasonably centralized location. By its very nature, this technical school offers a hands-on, in-the-field approach for a vast majority of its students seeking degrees in various traditional vocations. This is not to suggest that the evidentiary standard is immeasurable by any means. Certainly the government cannot rely upon a general, amorphous articulation of a special need to support a privacy invasion with immediate Fourth Amendment implications. Chandler, 520 U.S. at 320-21, 117 S.Ct. 1295. But the evidence does not support such shortcomings in this case. Linn State demonstrated that its students are primarily involved in vocational programs fraught with risks such that anyone participating in these programs who is under the influence, with impaired perception and judgment, creates an unnecessary and dangerous risk to themselves and others. Linn State admitted affidavits, witness testimony, early documents discussing the purpose behind the policy, and even onsite photographs of student work areas to better explain the nature of the course work in various programs as well as the numerous, potential risks created given the general tools and class work necessary to advance in each program. With this evidence, and because of the unique circumstance presented by this educational endeavor, we see no need to parse the information so finely.

In light of the particular programming taking place on Linn State's campus, Linn State has a justified interest in providing a safe, healthy, and productive environment for its relatively small student body and faculty population, while simultaneously preparing students for the realities of the workplace environments they are entering. This is a special need significant enough to conduct a balancing test of competing constitutional interests.

2. Balancing

a. Nature of Privacy Interest

As before, we must consider the nature of the privacy interest compromised by the drug-testing policy. Earls, 536 U.S. at 830, 122 S.Ct. 2559. In Barrett, addressing only whether Appellees were able to sustain a facial challenge to the policy at issue, we held that "some college students that attend Linn State have a diminished expectation of privacy because they are seeking accreditation in heavily regulated industries and industries where drug testing, in practice, is the norm," and thus held Appellees could not maintain a facial challenge. 705 F.3d at 323. Based on this statement, the district court held that only those students in programs for which specific evidence and expert testimony were presented at trial showing that industry testing was the norm or that the industry involved was heavily regulated, would be considered to have a diminished expectation of privacy. Outside of that limited context, the court found no diminished privacy interest whatsoever for the Linn State students, thus granting all remaining students with full, substantial privacy expectations common to all adults. The court erred in so limiting its analysis.

While the determination in Barrett remains relevant in the instant analysis because those students engaged in, or who will soon be working in, heavily regulated industries where drug testing is the norm, have diminished privacy expectations, our review today requires additional considerations. The Supreme Court has recognized circumstances in the educational system (outside the context of employees in especially hazardous occupations or safety-sensitive positions) that give rise to a substantial need that justifies suspicionless drug testing. See Vernonia, 515 U.S. 646, 115 S.Ct. 2386; Earls, 536 U.S. 822, 122 S.Ct. 2559. In the educational setting, specifically in the public school setting where the schools have a custodial and tutelary responsibility for the children they are educating, the Supreme Court has acknowledged an important interest in deterring drug use among student athletes and those students participating in competitive extracurricular activities. Earls, 536 U.S. at 831, 834, 122 S.Ct. 2559; Vernonia, 515 U.S. at 661-63, 115 S.Ct. 2386. This court has likewise sanctioned random drug testing of an entire high school student body, applying a similar analysis. Miller v. Wilkes, 172 F.3d 574, 581 (8th Cir.1999).

That this case involves students is a key component of the privacy interest at stake although not determinative on its own. We fully recognize that the privacy interests discussed in cases such as Earls, Vernonia, and Miller, rely heavily on the tutelary aspect of our nation's public schools, "permitting a degree of supervision and control that could not be exercised over free adults." Vernonia, 515 U.S. at 655, 115 S.Ct. 2386. We additionally recognize that the privacy interests of college students in a public technical school are more akin to those we bestow upon individual adults. But, the evidence establishes that Linn State certainly maintains a level of supervision appropriate for students in this particular college setting. Previously mentioned, the privacy interest here is a unique combination of that discussed for those adults subjected to suspicionless testing due to their participation in closely regulated industries or working in safety-sensitive positions, and students in more protected educational settings. Accordingly, the expectation of privacy for all Linn State students is somewhat diminished as they are either entering into areas of instruction and future fields of employment in highly regulated and safety-sensitive positions; or they are juxtaposed with students who are doing so; or they are attending classes in such areas on an intermittent basis due to the actuality of or potential of cross enrollment.

The district court found unpersuasive Linn State's argument that the possibility of cross enrollment renders this drug testing reasonable under the Fourth Amendment because students enrolled in non-dangerous programs could, according to Linn State, elect to take courses in programs that include tasks that pose a significant safety risk. The district court held this possibility was "abstract and unsubstantiated." We disagree, as the evidence supports a contrary conclusion. There is evidence that cross enrollment does, and can, occur. In any event, while there are certainly some programs that pose little safety risk relative to other programs where the risk is more prominent, the environment at Linn State on whole is the critical consideration because the student body does not participate in their respective disciplines in a vacuum. Indeed, by way of example and applying the district court's analysis, there is nothing prohibiting a student from enrolling in a program without testing, and then later taking classes in, or even transferring to, a program that requires testing-a scenario difficult to address in the current circumstances even from a logistical and bookkeeping standpoint.

The district court's refusal to acknowledge Linn State's unique role in this educational setting is error. In its exhaustive risk analysis of specific programs offered at Linn State, the court often mitigated the safety risks in programs based upon the supervision of Linn State staff in the classroom. Yet, by doing so, the court emphasized the important role of the instructors in the educational setting where students need supervision, and quite clearly acknowledged the somewhat diminished expectation of privacy of all Linn State students. Many of the Linn State students are performing dangerous work for the first time. If this technical college is to shoulder the obligation to educate its students in these vocational fields, that responsibility requires at least a concomitant obligation from its students to participate drug and alcohol free. Thus, this unique environment requires a heightened level of supervision and somewhat diminished expectation of privacy.

b. Character of the Intrusion

The analysis maintained in Barrett regarding the character of the privacy intrusion remains unchanged in Appellees' as-applied challenge. Barrett, 705 F.3d at 323. The procedures in place significantly minimize the intrusiveness of Linn State's drug-testing policy and the invasion of students' privacy is not significant. Id.; see also Chandler, 520 U.S. at 318, 117 S.Ct. 1295 (determining that the testing method employed by the state was relatively non-invasive and that the state could not be faulted for excessive intrusion).6

c. Nature and Immediacy of Concerns and Efficacy of the Policy

Here, we start with the premise established in Barrett: the need to prevent and deter the substantial harm that can arise from a student under the influence of drugs while engaging in a safety-sensitive program provides the necessary immediacy for Linn State's testing policy. 705 F.3d at 322. In support of the new policy, the proponents noted research that "[d]rug use has been found linked to . . . injuries and deaths," and emphasized that the incidence of drug abuse and addiction on college campuses is steadily rising, which is particularly acute in a vocational setting where the programs involved have dangerous and safety-sensitive components. This current state is a surprise to no one-the rampant problem with drugs and alcohol was readily apparent even in 1989. Even then the Court reiterated that there is no doubt "that drug abuse is one of the most serious problems confronting our society today," and that this social problem permeates indiscriminately. Von Raab, 489 U.S. at 674, 109 S.Ct. 1384; see also Earls, 536 U.S. at 834-35, 122 S.Ct. 2559 and Miller, 172 F.3d at 580-81.

Morse v. Frederick, 551 U.S. 393, 407, 127 S.Ct. 2618, 168 L.Ed.2d 290 (2007). While Morse surely limited its discussion to the pervasive drug use and abuse epidemic among our Nation's young people, we do not suppose that the problem abates the day after high school graduation and thus the data is certainly relevant to the instant discussion. In light of the massive problem in today's society, and given the previous discussion, articulating in detail the importance of Linn State's concern in preventing drug use by its enrolled students, Linn State's real and immediate interest in administering this policy is undeniable.

Additionally, as already noted in Barrett, "[w]hile it is true that random testing may be a more effective deterrent," Linn State's alleged failure to adopt the most effective drug-testing policy is not reason enough to override Linn State's substantial interest and pursuit in this matter. 705 F.3d at 323-24. In Von Raab, the Court addressed similar arguments from those challenging the efficacy of the chosen drug-testing policy, ultimately rejecting them because the Court held that focusing on ways to, say, manipulate a test too easily "overstates the case." 489 U.S. at 676, 109 S.Ct. 1384. At bottom, as in Von Raab, the policy here holds a "close and substantial relation" to Linn State's goal of advancing safety and educational interests on campus, given the unique vocational focus of this college. Id.

III. CONCLUSION

Linn State's student population comprised of roughly 1200 students are primarily engaged in safety-sensitive and potentially dangerous curriculum due to the unique nature of this particular vocational and technical college and its limited focus. On balance, testing the entire student population entering Linn State is reasonable and hence constitutional and an effective means of addressing Linn State's interest in providing "a safe, healthy, and productive environment for everyone who learns and works at LSTC by detecting, preventing, and deterring drug use and abuse among students."

For the reasons stated herein, we reverse the district court's permanent injunction and remand for dismissal of the case. Because we reverse this matter, we also necessarily reverse the district court's award of fees in favor of Appellees and, accordingly, remand the consolidated matter for dismissal as well.

BYE, Circuit Judge, dissenting.

I would affirm the district court's order granting the Appellees (hereinafter "students") a permanent injunction from Appellants' (hereinafter "Linn State") suspicionless drug-testing policy for all but five of Linn State's academic programs because the district court's program-by-program analysis is correct and Linn State failed to present sufficient evidence demonstrating a special need for drug testing. Based on the evidence presented, the district court did not abuse its discretion. I therefore respectfully dissent.

I

The district court did not abuse its discretion by conducting a program-by-program analysis because this approach is consistent with our instructions in Barrett v. Claycomb, 705 F.3d 315 (8th Cir.2013). In Barrett we held "some college students that attend Linn State have a diminished expectation of privacy." Id. at 323 (emphasis added). This Court's use of the words "some college students" can only be interpreted as meaning not all of Linn State's students have a diminished expectation of privacy. Therefore, this Court tasked the district court with sifting through Linn State's twenty-eight distinct academic programs to determine which programs "pos[ed] significant safety risks to others." Id. By closely analyzing Linn State's academic curriculum program by program, the district court followed our instructions in Barrett.

The district court's approach is also consistent with the nature of an as-applied constitutional challenge. In Barrett this Court held Linn State's suspicionless drug testing policy was facially constitutional. The question presented to us today is fundamentally different as we must determine the constitutionality of Linn State's suspicionless drug testing policy as-applied. This distinction is significant — we analyze facial challenges broadly to determine whether a statute or policy as written is constitutional under most circumstances whereas we evaluate as-applied challenges by applying the statute or policy to the particular person in their particular circumstance to determine constitutionality. See, e.g., Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 449, 128 S.Ct. 1184, 170 L.Ed.2d 151 (2008) ("[A] plaintiff can only succeed in a facial challenge by establish[ing] that no set of circumstances exists under which the Act would be valid.") (internal quotations omitted); United States v. Stephens, 594 F.3d 1033, 1039-40 (8th Cir.2010) (holding the record did not contain enough facts to determine whether the statute was constitutional as applied to the particular facts and circumstances of the case).

The facts of this case exemplify why the district court's program-by-program analysis was required for this as-applied constitutional challenge. This Court's facial analysis in Barrett makes clear that not all Linn State programs warrant drug testing. Subsequently, the purpose of the district court's analysis was to reveal which specific Linn State programs required drug testing. The district court correctly performed this exact analysis by applying Linn State's drug-testing policy to the individual settings of the twenty-eight distinct programs and ultimately determined only five programs justified drug testing. In contrast, the majority applies Linn State's drug testing policy to the entire 1100-1200 student body as one homogenous mass to assess overall constitutionality. In this all-or-nothing analysis, the majority concludes drug testing is warranted for all students. The majority's broad approach is misplaced: it is more akin to a facial analysis, and it ignores this Court's findings in Barrett.

For the reasons stated above, I agree with the district court's program-by-program approach.

II

I think it is inappropriate to analyze Linn State's policy at a campus-wide level, but because the majority does so, and because I disagree with their analysis, I will analyze the campus-wide policy under governing Fourth Amendment law. Accordingly, to determine whether Linn State's drug-testing policy is constitutional, it is undisputed the Fourth Amendment two-step balancing test applies. Linn State is first required to establish a special need for its policy. Barrett, 705 F.3d at 324. Once a special need is proffered it is balanced against three factors: (1) the nature of the privacy interest; (2) the character of intrusion; and (3) the nature and immediacy of the governmental concern and efficacy of the policy. Id. (citing Bd. of Educ. of Indep. Sch. Dist. No. 92 of Pottawatomie Cty. v. Earls, 536 U.S. 822, 830, 832, 834, 122 S.Ct. 2559, 153 L.Ed.2d 735 (2002)).

A. Special Need

Operating under the framework discussed in Part I, Linn State has only met its burden of demonstrating a drug-related public safety concern for five academic programs. The Supreme Court cautioned when "public safety is not genuinely in jeopardy, the Fourth Amendment precludes the suspicionless search, no matter how conveniently arranged." Chandler v. Miller, 520 U.S. 305, 323, 117 S.Ct. 1295, 137 L.Ed.2d 513 (1997). Accordingly, Linn State had the burden of establishing its academic programs posed a genuine public safety risk. Linn State met this burden for five academic programs because it demonstrated the particular program presented a concrete risk of injury to others in the vicinity or simply that drug testing was the industry norm. Linn State failed to prove how the remaining twenty-three programs placed public safety genuinely in jeopardy.

Further, we must be prudent in analyzing a generalized drug-related public safety justification for an exception to the Fourth Amendment. In reviewing the record, it is clear "public safety" is just one of several reasons the Board cited for enacting the drug-testing policy. For example, the six "Program Goals" adopted by the Board, the advisory committee minutes, and testimony from Board members all reveal non-safety motivations for Linn State implementing the drug-testing policy, including: to improve retention and graduation rates, to "up enrollment numbers," to appeal to students' parents because "parents want their kids to attend a school that enforces a drug-free environment," and to prepare students for employment in fields in which drug screening might be required. While these rationales may be valid from a business perspective, they do not provide permissible exceptions to the Fourth Amendment. In light of these additional motivations, I view Linn State's public safety proclamation with skepticism and ultimately do not believe it provided sufficient evidence to demonstrate public safety is genuinely in jeopardy to justify constitutional infringement.

For those reasons, I believe the evidence presented is insufficient to support Linn State's proffered generalized special need of public safety.

B. Balancing Test

Although I believe we need not reach the balancing test, the majority does. Even assuming Linn State met its burden of demonstrating a special need, the drug-testing program cannot withstand the three-prong balancing test. See Barrett, 705 F.3d at 324.

1. Nature of the Privacy Interest

We start from the position that adults have a strong Fourth Amendment privacy interest in being free from warrantless search and seizure. The majority discounts this position by attempting to analogize the privacy interests of high school students as being consistent with the privacy interests of adult college students.7 Second, the majority erodes Fourth Amendment protections even further by rendering the mere possibility of cross enrollment between programs as being enough evidence to justify drug testing for all students.

First, the majority seems to argue Earls, 536 U.S. 822, 122 S.Ct. 2559, Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 115 S.Ct. 2386, 132 L.Ed.2d 564 (1995), and Miller v. Wilkes, 172 F.3d 574 (8th Cir. 1999) can be interpreted as standing for the proposition that students, regardless of age and level of education, have a diminished expectation of privacy given the tutelary nature of being a student. I strongly disagree. Drug testing was found to be constitutional in Earls, Vernonia, and Miller because the individuals affected were children that happen to be high school students. See Earls, 536 U.S. at 824, 122 S.Ct. 2559 ("The need to prevent and deter the substantial harm of childhood drug use provides the necessary immediacy for a school [drug] testing policy." (emphasis added)); Vernonia, 515 U.S. at 649, 115 S.Ct. 2386 (holding drug testing high school student athletes was constitutional); Miller, 172 F.3d at 582 (holding random drug testing of students between grades seven and twelve was constitutional). Here, Linn State's suspicionless drug-testing policy affects adults who happen to be college students. This is a distinction with a difference because adults are entitled to greater protections of their privacy interests than are children. The majority is extending the scope of the holdings in Earls, Vernonia, and Miller to assert the dispositive factor in determining the constitutionality of drug testing is whether the individuals have student status as opposed to whether the student is an adult or child. I believe such an extension is erroneous and overreaching.

Second, Linn State did not meet its burden of demonstrating students' cross enrollment across Linn State's academic programs presents a safety concern justifying a Fourth Amendment exception. The majority accepts Linn State's assertion that because some programs require a diminished expectation of privacy and students may freely cross enroll into any Linn State program then all students must have a diminished expectation of privacy for the safety concerns associated with cross enrollment. It is undisputed Linn States carries the burden of demonstrating a recognized exception to the Fourth Amendment including that cross enrollment establishes enough of a safety concern to permit exception. See Der v. Connolly, 666 F.3d 1120, 1127-29 (8th Cir.2012); see also Chandler, 520 U.S. at 319, 117 S.Ct. 1295. Here, not only did Linn State fail to present any evidence of students engaged in cross enrollment but, more egregiously, Linn State submitted it did not even have to. Appellants' Br. at 32-33 ("[S]urely the school is in a position to know whether cross enrollment actually happens, without having to present student-specific occasions of it." (emphasis added)). Accordingly, it is clear Linn State did not meet its burden of proof and the mere possibility of cross enrollment is insufficient to justify a Fourth Amendment exception.

For those reasons, I find the majority is incorrect in determining adult college students have a diminished expectation of privacy just because they have student status and may cross enroll in programs requiring drug testing.

2. Character of Intrusion

In Barrett, this Court found relevant, "Linn State's written procedures inform the students that the testing will be conducted in accordance with federal drug-testing procedures outlined in 49 C.F.R. Part 40, which `significantly minimize the program's intrusion on privacy interests.'" Barrett, 705 F.3d at 323 (quoting Nat'l Treasury Emps. Union v. Von Raab, 489 U.S. 656, 672 n. 2, 109 S.Ct. 1384, 103 L.Ed.2d 685 (1989)). The majority relies on Barrett in concluding the analysis remains unchanged in this as-applied challenge because "[t]he procedures in place significantly minimize the intrusiveness of Linn State's drug-testing policy." However, Linn State's drug-testing policy does not comport with federal drug-testing procedures and therefore the majority errs in holding the federal procedures in place significantly minimize the intrusiveness of the drug-testing policy.

Linn State's testing procedures differ from federal testing procedures in a number of ways but most egregious is Linn State's parental notification clause which permits a student's drug-test results to be shared with a third party — namely parents. This is unlike federal drug-testing procedures which require strict confidentiality of test results. There is a greater invasion of privacy when drug test results are capable of being shared with a third party. See Ferguson v. City of Charleston, 532 U.S. 67, 68, 121 S.Ct. 1281, 149 L.Ed.2d 205 (2001); Lebron v. Wilkins, 820 F.Supp.2d 1273, 1283 (M.D.Fla.2011) (finding a "far more substantial" invasion of privacy when positive drug tests resulting from the collection of urine samples are shared with third parties). Although Linn State asserts its drug-testing policy remains non-invasive because there is no evidence to demonstrate it has notified a parent, this argument is without merit. Regardless of whether Linn State has previously notified a third party, the ability to do so remains. Therefore, Linn State's mere ability to notify a third party automatically results in a greater invasion of privacy than what was found acceptable by this Court in Barrett.

I therefore believe the majority erred in relying on our facial analysis in Barrett to conclude the character of intrusion remains unchanged in this as-applied challenge. In actuality, Linn State's drug-testing policy as-applied is more invasive than the federal procedures with which it claims to comply. Accordingly, Linn State's drug-testing policy is a substantial invasion of privacy.

3. Nature and Immediacy of the Governmental Concern and Efficacy of the Policy

The majority concludes, "Linn State's real and immediate interest in administering this [drug-testing] policy is undeniable" in light of the "massive [drug] problem in today's society." I find the majority's fear-ridden rationale to be troubling. Founded in 1961, Linn State successfully operated for fifty years before deciding in 2011 that a drug-testing policy was essential for safeguarding its students. Despite asserting the immediate necessity of the drug-testing policy, neither Linn State nor the majority cites any specific events or studies applicable to Linn State in particular to substantiate this assertion. Rather, the majority relies on a generalized concern over drug use based on the Supreme Court's comment in Morse v. Frederick, 551 U.S. 393, 407, 127 S.Ct. 2618, 168 L.Ed.2d 290 (2007), where it listed statistics regarding drug use amongst secondary school students and concluded drug abuse problem "remains serious today." I find the majority's reference to a case decided in 2007 providing drug-use statistics of secondary school students to be unpersuasive. While I certainly acknowledge drug and alcohol abuse are very serious problems, the majority's reference to Morse does not explain why there is a real and immediate risk present at Linn State today. In the absence of empirical evidence to bolster Linn State's special needs argument requiring suspicionless drug testing, I remain unconvinced Linn State's "real and immediate interest in administering the [drug-testing] policy" is even plausible, let alone "undeniable."

In light of the undiminished privacy interest and invasive drug-testing procedures, I find Linn State's campus-wide program fails the three-prong balancing test.

III

For the above stated reasons, the district court did not abuse its discretion. I would affirm the district court's order granting the students a permanent injunction from drug testing in all but the five academic programs identified.