Rogers, Circuit Judge:

Confronted with evidence of increasing juvenile violence and victimization in the District of Columbia, and informed about the success of other cities in reducing such problems through the enforcement of juvenile curfews, the Council of the District of Columbia enacted the Juvenile Curfew Act of 1995. The Council modeled the Act on a Dallas, Texas, ordinance that the United States Court of Appeals for the Fifth Circuit had held was constitutional. See Qutb v. Strauss, 11 F.3d 488, 496 (5th Cir. 1993). The main provision of the D.C. Act bars unmarried and unemancipated persons under seventeen years old from being in public unaccompanied by a parent or equivalent adult supervisor from 11:00 p.m. to 6:00 a.m. on Sunday through Thursday nights or from 12:01 a.m. to 6:00 a.m. on Friday and Saturday nights, with certain enumerated "defenses." See D.C. Code §§ 6-05 2182(1), -2183(a)(1), (b)(1) (Supp. 1997). Thirteen months after the Act took effect, the district court enjoined its enforcement, ruling in light of evidentiary deficiencies that the Act violated the minor appellees' equal protection and due process rights and violated the appellee parents' right to due process. See Hutchins v. District of Columbia, 942 F. Supp. 665, 668 (D.D.C. 1996). The District of Columbia, joined by the United States as amicus, appeals the grant of summary judgment to appellees. We affirm, albeit with different analyses. While the court is unanimous that the case is not moot, see infra Part II, we apply different tests to evaluate the constitutionality of the Act. I apply an intermediate scrutiny test in light of competing individual and governmental interests, while Judge Tatel applies strict scrutiny and Judge Silberman applies a rational basis test. Judge Tatel and I agree that the Act fails to survive under intermediate or strict scrutiny review; Judge Silberman dissents, concluding that the Act survives rational basis review.

I.

Appellees, nine persons under the age of seventeen at the time and four parents, all residents of the District of Columbia ("the District"), and a movie theater corporation sued the persons under seventeen years old; the term "juveniles" refers to persons under eighteen years old; the phrase "juvenile curfews" refers to curfew laws affecting any class of young people.

District to enjoin enforcement of the Juvenile Curfew Act of 1995 ("the Act"). They sought a declaration that the Act violates rights guaranteed by the First, Fourth, and Fifth Amendments to the United States Constitution and exceeds the police powers of the District of Columbia. Their principal allegations were that the Act violates the minors' Fifth Amendment equal protection and due process rights by impinging upon their fundamental right to free movement; the Act violates their First Amendment rights to free speech and association and is both overbroad and unconstitutionally vague; and the Act violates their Fourth Amendment rights to be free of unreasonable searches and seizures in that it allows the police to stop minors and take them into custody based only on a reasonable belief that the Act has been violated. In addition, appellees alleged that the Act violates the parents' Fifth Amendment due process rights because, by removing parents' discretion to allow children to be in public places during curfew hours unaccompanied by a person at least twenty-one years old, the Act impinges upon parents' fundamental right to autonomy in raising children. Finally, appellees alleged that the Act exceeds the District's police powers by criminalizing minors' participation in legitimate educational, cultural, vocational, athletic, social, and family-related activities during curfew hours. Appellees argued that, because the Act infringes on both the minors' and parents' fundamental rights, it is subject to strict scrutiny review, which, they asserted, it fails to satisfy.

Upon considering the parties' cross motions for summary judgment, the district court granted judgment for appellees and enjoined enforcement of the Act. See Hutchins, 942 F. Supp. at 684. The court agreed that minors have a fundamental right to free movement, reasoning from the Supreme Court's acknowledgments that minors have constitutional rights and that adults have a fundamental right to free movement to the conclusion that, in the context of a curfew law, there is no reason to treat minors' right to free movement differently from that of adults. See id. at 670-74. Then, concluding that the Act infringes minors' fundamental right to free movement as well as parents' fundamental right to direct their children's upbringing, the court applied a strict scrutiny test and found that while the District had demonstrated a compelling need for the curfew, it had failed to demonstrate that the Act is narrowly tailored to serve that need. See id. at 674-80. Based on deficiencies in the District's evidentiary justification for a nexus between the curfew and a future reduction in juvenile victimization and crime, the court concluded that the Act affects too many minors engaged in legitimate activities. See id. at 680. The court further ruled that four of the Act's curfew "defenses" are unconstitutionally vague, see id. at 679, but did not reach the minor appellees' First and Fourth Amendment challenges, see id. at 680 n. 19. Because the Act also affects parents exercising appropriate supervision of their children, the court ruled that the Act infringes upon parents' fundamental rights in violation of the Fifth Amendment. See id. at 680.

On appeal, the District of Columbia contends that the Act is constitutional, curtailing only limited late night activities of unsupervised minors and thereby interposing only a minor interference with parental autonomy. The District maintains, first, that the district court erred in applying a strict scrutiny standard in the face of authority rejecting any fundamental right of minors to wander unsupervised at night, Supreme Court precedent regarding more general limitations on minors' rights and parental autonomy, and the Supreme Court's stringent guidelines for the identification of new fundamental rights. Alternatively, the District maintains that even if strict scrutiny is the proper standard, the Act still should be upheld: the district court's finding that the District has a compelling interest in preventing juvenile crime and protecting juveniles against victimization is supported by abundant evidence and the Act is narrowly tailored to that interest, as demonstrated by evidence that the district court rejected. Finally, the District maintains that the Act does not violate appellees' First or Fourth Amendment rights, and that the district court erred in ruling that four of the curfew exceptions in the Act are unconstitutionally vague without offering either an explanation or a saving construction as required by Supreme Court precedent.

The constitutionality of a juvenile curfew statute is a question of first impression in this court, and our review is de novo. See Wilson v. Pena, 79 F.3d 154, 160 n. 1 (D.C. Cir. 1996); Propert v. District of Columbia, 948 F.2d 1327, 1331 (D.C. Cir. 1991).

The Act establishes a curfew from 11:00 p.m. until 6:00 a.m. the next day for Sunday night through Thursday night, and between 12:01 a.m. and 6:00 a.m. on Friday and Saturday nights and daily throughout July and August. See D.C. Code § 6-2182(1) (reprinted in the Appendix to this opinion). The curfew applies only to "minors," who are defined as persons under age seventeen who are neither emancipated nor married. See id. § 6-2182(5). During curfew hours, a minor may not be "in any public place or on the premises of any establishment within the District of Columbia" without appropriate adult supervision. Id. § 6-2183(a)(1), (b)(1)(A). An "adult" is defined as a parent or any person twenty-one years or older whom the minor's parent has authorized to be a caretaker. See id. §§ 6-2182(8). The Act also provides that a person under age eighteen shall not operate a motor vehicle in the District after midnight, except when authorized under the Act. See id. § 40-301(g) (Supp. 1997). A police officer who "reasonably believes" that a curfew violation has occurred may inquire of the minor about the minor's age and reasons for being in public and, upon determining that the minor is violating the curfew, may hold the minor in custody until the minor's parent arrives or 6:00 a.m. the following morning, whichever occurs first. Id. § 6-2183(c).

A minor who violates the Act can be required by a court to perform up to twenty-five hours of community service for each violation. See id. § 6-2183(d)(4). Violation of the driving restriction can result in suspension of one's driver's license for up to one year. See id. § 40-301(g). Both a parent or guardian of a minor who either knowingly permits or, because of insufficient control, allows the minor to violate the Act and the owner, operator, or any employee of an establishment who violates the Act by knowingly allowing a minor to remain on its premises during curfew hours may be fined up to $500 or ordered to perform community service. See id. § 6-2183(a)(2)-(3), (d)(1). A parent (and persons in loco parentis) may also be required to attend parenting classes. See id. § 6-2183(d)(2).

The Act provides eight "defenses" to a curfew violation. Thus, a minor does not violate the Act if: (1) accompanied by a parent, guardian, or any other person twenty-one years or older authorized by a parent to care for the minor; (2) on an errand for the parent, guardian, or anyone twenty-one years or older authorized by a parent to care for the minor; (3) in a vehicle traveling interstate; (4) engaged in employment or commuting to or from employment; (5) involved in an emergency situation; (6) on the sidewalk abutting the minor's or a next-door neighbor's residence, if the neighbor has not complained to police; (7) attending an official school, religious, or other activity sponsored by the District, a civic organization, or a similar entity that takes responsibility for the minor; or (8) exercising First Amendment rights, including the freedoms of religion and speech and the right of assembly. See id. §§ 6-2182(8), -2183(b)(1)(A)-(H).

The Act originally also included a sunset clause whereby the curfew regime would expire after two years. See D.C. Act 11-90, § 6(b) (July 6, 1995). At least ninety days prior to the date of expiration, the Mayor was required to report to the Council of the District of Columbia ("D.C. Council") on the effectiveness of the curfew restrictions and to recommend whether the Act should be extended. See D.C. Code § 6-2183(e)(1). Prior to expiration of the two-year period, and without a report from the Mayor, however, the D.C. Council extended the Act indefinitely.

In view of the recent seventeenth birthdays of all the minor appellees, see supra note 2, we first address, in Part II, whether their challenges to the Act are moot and not otherwise before the court. Although the District is appealing the district court's judgment and order, upon review of the grant of summary judgment the case appears in this court in the same posture as it did in the district court and, therefore, this court must determine whether the minors' challenges to the Act remain "alive." See DeFunis v. Odegaard, 416 U.S. 312, 316 (1974). Concluding in Part II that the minor appellees' claims are presented to this court in a representative capacity by the appellee parent of a child under the age of seventeen, we address in our separate opinions the minor appellees' challenges to the Act. As Judge Tatel and I conclude that the Act violates the equal protection and due process rights of the minor appellees, we do not reach appellees' other challenges.

II.

While this appeal was pending, the last of the named minor appellees became age seventeen, and hence they all are no longer subject to the Act, except the motor vehicle restriction, which applies to those not yet age eighteen. Thus, their principal challenges to the Act are moot because the minor appellees were not certified as representatives of a class pursuant to Federal Rule of Civil Procedure 23. See United States Parole Comm'n v. Geraghty, 445 U.S. 388, 398, 400 n. 7, 404 (1980); Board of Sch. Comm'rs v. Jacobs, 420 U.S. 128, 129 (1975). Indeed, they expressly declined to pursue class certification in the district court. Of the remaining named appellees, however, at least one is a parent with a child under age seventeen. The question, therefore, is whether this parent can raise the claims of his minor child, claims that are the same as those of the appellees who were subject to the curfew when the complaint was filed, or whether instead those claims are moot except to the extent they relate to the motor vehicle restriction.

This is not a case in which the familiar exception to the mootness doctrine for issues "capable of repetition yet evading review" is applicable. That exception is confined to situations in which there is a reasonable expectation that the same complaining party would be subjected to the same action again. See Honig v. Doe, 484 U.S. 305, 317-20 (1988); Weinstein v. Bradford, 423 U.S. 147, 148-49 (1975); Sosna v. Iowa, 419 U.S. 393, 399-400 (1975); Burlington N. R.R. v. Surface Transp. Bd., 75 F.3d 685, 689 (D.C. Cir. 1996); Doe v. Sullivan, 938 F.2d 1370, 1378 n. 13 (D.C. Cir. 1991); Bois v. Marsh, 801 F.2d 462, 466 (D.C. Cir. 1986). Because the minor appellees have reached age seventeen, they are no longer subject to the curfew restrictions (except the motor vehicle restriction) and will never again be subject to the curfew restrictions for persons under age seventeen.

Instead, the third party (or jus tertii) standing doctrine of Craig v. Boren, 429 U.S. 190 (1976), and its progeny applies and provides the basis for our conclusion that the minor appellees' challenges to the Act remain before this court. "Ordinarily, one may not claim standing . . . to vindicate the constitutional rights of some third party." Barrows v. Jackson, 346 U.S. 249, 255 (1953); see United States v. Raines, 362 U.S. 17, 21 (1960). The Supreme Court observed in Craig, however, that its limitations on the right to raise the interests of third parties, where the interests of the litigant and the proposed third party are "in no way mutually interdependent," Craig, 429 U.S. at 195 n. 4, are "not constitutionally mandated, but rather stem from a salutary `rule of self-restraint' designed to minimize unwarranted intervention into controversies where the applicable constitutional questions are ill-defined and speculative," id. at 193. Acknowledging that the reasons for such limitations are not furthered where the lower court has already addressed the relevant constitutional challenge and the parties have never resisted an authoritative constitutional determination, the Court concluded that forgoing consideration of the merits in order to wait for a new challenge by injured third parties "would be impermissibly to foster repetitive and time-consuming litigation under the guise of caution and prudence." Id. at 193-94.

This is just such a case. Under the Act, parents can be sanctioned for allowing their minor children to violate the curfew. See D.C. Code § 6-2183(a)(2), (d)(1)-(2). In comparable circumstances in Craig, the Supreme Court held that, because a bartender could be sanctioned under a statute barring sale of beer to males under age twenty-one and females under age eighteen, the bartender had standing to raise the equal protection claims of a male plaintiff who had reached age twenty-one during the appeal and whose claims had thus become moot. See Craig, 429 U.S. at 195. Furthermore, the other relevant considerations underlying the third party standing doctrine, articulated in cases following Craig, all point toward the conclusion that the court should proceed to address the minor appellees' claims.

As further elaborated by the courts, the third party standing doctrine involves the consideration of four factors, one constitutional and three prudential. First, the litigant asserting the third party's claims must himself or herself suffer an Article III injury-in-fact. See Caplin Drysdale, Chartered v. United States, 491 U.S. 617, 623 n. 3 (1989). The alleged infringement of the parent's own rights, coupled with the parental sanctions under the Act, are sufficient to establish that the appellee parent meets this requirement. See Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992). Second, the court will look for the presence of three prudential considerations: (1) a close relationship between the litigant and the third party whose rights are being asserted; (2) a barrier keeping this third party from asserting such rights himself or herself; and (3) an impact of the litigation on the rights of the third party. See Caplin Drysdale, 491 U.S. at 623 n. 3; see also Powers v. Ohio, 499 U.S. 400, 410-11 (1991).

The first of these prudential considerations is clearly satisfied in the instant case. At least two circuits, which we join, have held that the parent-child relationship is sufficiently close to meet prudential standing requirements. In a case directly on point, the Fifth Circuit held in Johnson v. City of Opelousas, 658 F.2d 1065 (5th Cir. Unit A Oct. 1981), that the mother of children still subject to a juvenile curfew could raise her son's claims despite their becoming moot when her son became seventeen. See id. at 1069. In Lindley ex rel. Lindley v. Sullivan, 889 F.2d 124 (7th Cir. 1989), the Seventh Circuit held that an adopted child receiving disability benefits had standing to assert his parents' equal protection claims regarding denial of child insurance benefits, on the observation that the parent-child relationship was "much closer than [the relationship] in leading cases where standing has been found to exist, as between a physician and a patient or between a beer vendor and a class of potential purchasers of the product." Id. at 129 (citations omitted).

The third prudential consideration also militates in favor of allowing third party standing. The parent has explicitly referred to the direct impact of the curfew on the rights asserted by the minor appellee. A decision based on the parent's third party standing would definitely have a significant impact upon those rights.

The only prudential factor not clearly evident in the instant case is the second, since a very young plaintiff undoubtedly could be found to bring a lawsuit challenging the curfew (although even a plaintiff who is quite young at the beginning of the litigation might turn seventeen by the end). Yet in Caplin Drysdale, the failure to satisfy the second prudential factor did not defeat third party standing. See Caplin Drysdale, 491 U.S. at 624 n. 3; cf. Department of Labor v. Triplett, 494 U.S. 715, 720-21 (1990). Indeed, in Craig itself, the Supreme Court allowed third party standing even though there was no barrier to would-be beer-drinkers' pressing their own claims. See Craig, 429 U.S. at 192-97. Similarly, this circuit has repeatedly found that the absence of the second prudential factor may not outweigh the other factors in evaluating whether a litigant has third party standing. See FAIC Secs., Inc. v. United States, 768 F.2d 352, 360-61 (D.C. Cir. 1985); United Steelworkers of Am. v. Marshall, 647 F.2d 1189, 1241 (D.C. Cir. 1980). In the instant case, we conclude that the closeness of the relationship between parents and children and the magnitude of the potential impact of our decision on children's rights outweigh the absence of the second prudential factor. Cf. Caplin Drysdale, 491 U.S. at 624 n. 3.

Indeed, the proposition that parents who satisfy Article III standing requirements to raise their own claims may have standing to raise their children's claims as well, even without an actual barrier preventing children from doing so themselves, also follows from Smith v. Organization of Foster Families for Equality Reform, 431 U.S. 816 (1977), and like cases. In that case, the Supreme Court held that foster parents had standing to challenge the deprivation of foster children's right not to be removed from their foster homes without due process. See id. at 841 n. 44; cf. Bender v. Williamsport Area Sch. Dist., 475 U.S. 534, 546-49 (1986). Prudential standing principles have not barred suits by parents raising equal protection and First Amendment claims on behalf of themselves and their children in school desegregation and school prayer cases, although the cases have not explicitly addressed standing. See, e.g., Dayton Bd. of Educ. v. Brinkman, 433 U.S. 406, 408 n. 1 (1977); Engel v. Vitale, 370 U.S. 421, 423 (1962); Zorach v. Clauson, 343 U.S. 306, 309 n. 4 (1952). This court has similarly allowed parental challenges to violations of their children's equal protection rights in school. See, e.g., Bulluck v. Washington, 468 F.2d 1096, 1109 n. 14 (D.C. Cir. 1972) (Robinson, J., dissenting).

Furthermore, this result accords with the rationale for third party standing as articulated by the Supreme Court in Singleton v. Wulff, 428 U.S. 106 (1976), even if there is no actual barrier preventing children from raising their own claims. In Singleton, the Court instructed that third party standing is appropriate where "the enjoyment of the right [of the third party] is inextricably bound up with the activity the litigant wishes to pursue," and "the litigant . . . is fully, or very nearly, as effective a proponent of the right [as the party whose right is being asserted]." Id. at 114-15. The appellee parent with the minor child under age seventeen asserts that the curfew interferes not only with his parental right to allow his child to stay out, without adult supervision, after curfew hours, but also with his minor son's rights to engage in legitimate social activities and interests during times when the curfew is in effect, the very claims raised by the minor appellees. The instant case falls neatly within the rationale for third party standing.

Therefore, in light of Craig and its progeny, we hold that the appellee parent with a child under age seventeen who remains subject to the curfew has standing to raise the challenges to the curfew presented by the minor appellees whose claims have become moot as a result of the passage of time. The parent has suffered an injury-in-fact sufficient to confer Article III standing. Regarding the rights of his minor child, the appellee parent voices the same objections reflected in the minor appellees' challenges to the Act. The nature of the parent-child relationship suffices to ensure that the parent is an effective advocate for the minor appellees' interests, and disposition of the parent's claims will have a direct impact on the rights asserted by the minor appellees. Under the circumstances, it would be a waste of judicial resources at this late stage of the proceedings to abandon consideration of minor appellees' claims when the Act applies to other District of Columbia minors.

III.

The minor appellees contend that the Act restricts their fundamental right to free movement in violation of their due process and equal protection rights. In addition, they maintain that the Act violates their First Amendment rights of speech and association and their Fourth Amendment right to be free from unreasonable searches and seizures. First, I address in section III.A, the standard of review appropriate for analysis of minor appellees' due process and equal protection contentions; in section III.B, the District's purpose in enacting the Act; and in section III.C, the data offered to show the required connection between the problem and the solution.

A.

The District contends, and the United States agrees, that the district court erred in ruling that minors enjoy a fundamental constitutional right of free movement, that hence the Act must be reviewed under a rational basis test, and that the Act easily meets this standard. Appellees contend that the Act fails both a rational basis test as well as a strict scrutiny test, and that, in any event, because minors have a fundamental right to free movement upon which the Act impinges, the appropriate standard is strict scrutiny.

To date, the Supreme Court has not spoken on the precise issue and the lower federal courts have identified three standards of review with regard to juvenile curfews: rational basis, strict scrutiny, and intermediate scrutiny. Under the rational basis standard, no fundamental right is at issue and the District would only need to show a rational relationship between a juvenile curfew and any legitimate governmental interest: for instance, the need to stem juvenile violence and victimization in the District. See City of Dallas v. Stanglin, 490 U.S. 19, 25-28 n. 4 (1989). The rational basis standard is "true to the principle that the Fourteenth Amendment gives the federal courts no power to impose upon the States their views of what constitutes wise economic or social policy." Id. at 27 (quoting Dandridge v. Williams, 397 U.S. 471, 486 (1970)) (internal quotation marks omitted). By contrast, under a strict scrutiny standard, a fundamental right is implicated and the District would have to show that the Act is narrowly tailored to promote a compelling governmental interest. See Plyler v. Doe, 457 U.S. 202, 216-17 (1982). To be narrowly tailored, there must be a sufficient nexus between the compelling governmental interest and the provisions of the Act, see City of Richmond v. J.A. Croson Co., 488 U.S. 469, 493 (1989), and the Act must use the least restrictive reasonable means to achieve its goals, see Dunn v. Blumstein, 405 U.S. 330, 343 (1972). A third standard, intermediate scrutiny, also acknowledges the existence of a fundamental right but gives recognition as well to the existence of important governmental interests where minors are involved; it requires a showing that the Act serves "important governmental objectives" and that the means employed are "substantially related to the achievement of those objectives." Mississippi Univ. for Women v. Hogan, 458 U.S. 718, 724 (1982) (quoting Wengler v. Druggists Mut. Ins. Co., 446 U.S. 142, 150 (1980)) (internal quotation marks omitted) (gender); see Plyler, 457 U.S. at 225-30 (illegal immigrant minors); Lalli v. Lalli, 439 U.S. 259, 265 (1978) (plurality opinion of Powell, J.) (illegitimate children); Carey v. Population Servs. Int'l, 431 U.S. 678, 693 n. 15 (1977) (plurality opinion of Brennan, J.) (minor's right to obtain contraceptives). Described by one federal district court as a way to acknowledge both minors' claim to a fundamental right of free movement and the heightened interest of the government in protecting and fostering the development of its youth, see Schleifer v. City of Charlottesville, 963 F. Supp. 534, 540-42 (W.D. Va. 1997) (denying preliminary injunction), intermediate scrutiny requires a showing of a "substantial" or "important" rather than a "compelling" governmental interest, see Hogan, 458 U.S. at 724; Plyler, 457 U.S. at 217-18, 224, 230; and of a substantial fit between means and ends rather than narrow tailoring, see Hogan, 458 U.S. at 724; Lalli, 439 U.S. at 265, 268. The first question, then, is which standard is appropriate for the evaluation of the Act, and as our separate opinions indicate, there is more than one reasoned answer to this question.

Generally, legislation that treats one class of persons differently from others who are similarly situated is presumed to meet the equal protection requirements of the Fifth Amendment if the classification drawn by the legislation is "rationally related to a legitimate state interest." City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439-40 (1985). If the classification disadvantages a "suspect class" or burdens one group's exercise of a "fundamental right," the legislation is subject to strict scrutiny review. Plyler, 457 U.S. at 21617. Likewise, if a statute impinges upon a fundamental right, the substantive due process component of the Fifth Amendment requires that it satisfy strict scrutiny review. See Washington v. Glucksberg, 117 S.Ct. 2258, 2267-68 (1997). Because age does not determine a suspect class, see Gregory v. Ashcroft, 501 U.S. 452, 470 (1991), the court must examine whether the Act threatens the minors' exercise of a fundamental right, thus demanding strict scrutiny review.

The Supreme Court has held that adults have a fundamental right to free movement, and that minors have some fundamental rights entitled to constitutional protection. At the same time, the Supreme Court recognizes the state's heightened interest in the protection of children, see Prince v. Massachusetts, 321 U.S. 158, 168-69 (1944); Bellotti v. Baird, 443 U.S. 622, 635 (1979) (plurality opinion of Powell, J.), and acknowledges that "if parental control falters, the State must play its part as parens patriae," Schall v. Martin, 467 U.S. 253, 265 (1984). There is obvious tension between the propositions that there is a fundamental right to free movement and that minors possess some fundamental rights, and the proposition that the state has a greater interest in protecting minors than adults. So far, the Supreme Court has not explained how the tension is to be resolved with regard to juvenile curfews. Indeed, the Court has acknowledged that "[t]he question of the extent of state power to regulate conduct of minors not constitutionally regulable when committed by adults is a vexing one, perhaps not susceptible of precise answer." Carey, 431 U.S. at 692 (plurality opinion of Brennan, J.). Further, the Supreme Court has emphasized that courts must be "reluctant to expand the concept of substantive due process because guideposts for responsible decision-making in this unchartered area are scarce and open-ended." Collins v. City of Harker Heights, 503 U.S. 115, 125 (1992).

It is instructive that in a variety of contexts, the Supreme Court has distinguished between minors' and adults' constitutional rights. For instance, in Prince, while rejecting a challenge under the First and Fourteenth Amendments to a state statute prohibiting minors from selling merchandise on public streets, the Supreme Court explained:

The state's authority over children's activities is broader than over like actions of adults. This is peculiarly true of public activities and in matters of employment. A democratic society rests, for its continuance, upon the healthy, well-rounded growth of young people into full maturity as citizens, with all that implies. It may secure this against impeding restraints and dangers within a broad range of selection. Among evils most appropriate for such action are the . . . possible harms arising from . . . activities subject to all the diverse influences of the street. It is too late now to doubt that legislation appropriately designed to reach such evils is within the state's police power, whether against the parent's claim to control of the child or one that religious scruples dictate contrary action.

It is true children have rights, in common with older people, in the primary use of highways. But even in such use streets afford dangers for them not affecting adults. And in other uses . . . this difference may be magnified. This is so not only when children are unaccompanied but certainly to some extent when they are with their parents. What may be wholly permissible for adults therefore may not be so for children, either with or without their parents' presence.

Prince, 321 U.S. at 168-69 (footnotes omitted).

The state's greater authority over minors' conduct is similarly reflected in Ginsberg v. New York, 390 U.S. 629 (1968), where the Court, applying a rational basis standard to uphold a state statute banning the sale to minors of obscene materials, allowed New York to adjust the definition of obscenity for minor readers. See id. at 637-38. The same approach is reflected in the lines that the Court has drawn between adults' and minors' due process rights: while the government's burden of proof remains the same for both adult prosecutions and juvenile delinquency proceedings, see In re Winship, 397 U.S. 358, 368 (1970), and a minor has a right to counsel, a right to cross-examine witnesses, and a privilege against self-incrimination, see In re Gault, 387 U.S. 1, 36-37, 55-57 (1967), a minor does not have a right to a jury trial in juvenile delinquency proceedings analogous to an adult's right in a criminal prosecution, see Kent v. United States, 383 U.S. 541, 555 (1966). Similarly, although a state may not impose a blanket parental consent requirement for a minor to obtain an abortion, the Court has emphasized that this conclusion does "not suggest that every minor, regardless of age or maturity, may give effective consent for termination of her pregnancy," despite adults' possession of such a right. Planned Parenthood of Central Mo., 428 U.S. at 74-75. Generally, the linedrawing reflects the analysis of the plurality in Bellotti; in striking down a parental-consent statute as unduly burdensome on a minor's constitutional right to have an abortion, the plurality identified three factors, any one of which would suffice, justifying state action treating minors differently from adults in regard to constitutional protections: (1) "the peculiar vulnerability of children"; (2) children's "inability to make critical decisions in an informed, mature manner"; and (3) "the importance of the parental role in child rearing." Bellotti, 443 U.S. at 634 (plurality opinion of Powell, J.).

Further, the Supreme Court has concluded that the distinction between adults' and minors' constitutional rights applies with regard to certain rights to free movement. In Vernonia School District 47J v. Acton, 515 U.S. 646 (1995), the Court upheld a random urinalysis requirement for high school athletes against a Fourth Amendment challenge because the requirement was reasonable in light of the minimal legitimate expectations of privacy of students committed to the temporary custody of a schoolmaster, in particular those who join a school sports team and agree to change and shower in public locker rooms, to submit to preseason physical exams, to sign insurance waivers, to maintain minimum grades, and to comply with rules of conduct, dress, and training hours. See id. at 654-57, 664-65; cf. Tinker v. Des Moines Indep. Community Sch. Dist., 393 U.S. 503, 507 (1969). In so doing, the Court observed:

Traditionally at common law, and still today, unemancipated minors lack some of the most fundamental rights of self-determination — including even the right of liberty in its narrow sense, i.e., the right to come and go at will. They are subject, even as to their physical freedom, to the control of their parents or guardians. When parents place minor children in private schools for their education, the teachers and administrators of those schools stand in loco parentis over the children entrusted to them.

Vernonia Sch. Dist. 47J, 515 U.S. at 654 (citation omitted) (emphasis added). Likewise, in Reno v. Flores, 507 U.S. 292 (1993), the Court upheld an Immigration and Naturalization Service regulation allowing juvenile aliens detained pending deportation hearings to be released only to the custody of their parents, close relatives, or legal guardians, with the observation that "`juveniles, unlike adults, are always in some form of custody,' and where the custody of the parent or legal guardian fails, the government may (indeed, we have said must) either exercise custody itself or appoint someone else to do so." Id. at 302 (quoting Schall, 467 U.S. at 265) (citation omitted). Under these precedents, were the Supreme Court to hold that minors have the same fundamental right of movement as adults, it would have to jettison long-settled views on the distinct and yet at times concurrent roles of parents and the state, views that render some types of interference with free movement constitutionally permissible for minors even if not for adults.

Faced with the dilemma of what standard of scrutiny to employ in reviewing statutes that affect minors' privacy rights, the Supreme Court recognized the fundamental right of privacy that minors have in decisions affecting procreation while nonetheless applying a less rigorous test than strict scrutiny in examining whether a "significant state interest" justified the parental consent provision at issue. Planned Parenthood of Central Mo., 428 U.S. at 75. Soon thereafter a Supreme Court plurality explained that intermediate scrutiny "is appropriate both because of the States' greater latitude to regulate the conduct of children and because the right of privacy implicated here is `the interest in independence in making certain kinds of important decisions,' and the law has generally regarded minors as having a lesser capability for making important decisions." Carey, 431 U.S. at 693 n. 15 (plurality opinion of Brennan, J.) (quoting Whalen v. Roe, 429 U.S. 589, 599-600 (1977)) (citations omitted). The approach reflected in the intermediate scrutiny test fits comfortably in examining the rights affected by juvenile curfews. That minors have a fundamental right of movement in some regard, the nature of which admittedly is not precisely defined, is necessarily implied in the Court's decisions explaining circumstances in which that right is properly restricted. An intermediate scrutiny standard thus recognizes that in some circumstances, courts are compelled to equate the constitutional rights of minors and adults, and in other instances they are not. Compare Gault, 387 U.S. at 36-37, 55-57, with Kent, 383 U.S. at 555. Viewed in light of factors deemed to be significant in both Carey and Bellotti, juvenile curfews arise in a context in which children are more vulnerable than adults, see Prince, 321 U.S. at 168-69, and in which children's lesser ability to make important decisions wisely could cause them harm. As the district court in Schleifer explained, "[o]f course, on an isolated night, a decision to go out after curfew hours may not be a critical decision, but rather one of minimal importance; but that decision, made night after night, might have an adverse effect on a child's life." Schleifer, 963 F. Supp. at 542.

Consequently, because minors have certain constitutional rights that may include a right of movement under some circumstances, but the government also has interests that may override or infringe upon those rights, and because minors are generally more vulnerable on the street at night than adults, but might not be as able to make intelligent decisions about their outdoor late-night activities, an intermediate scrutiny standard should apply in examining the minor appellees' challenges to the Act. While Judge Tatel expresses concern that such a standard may impinge unnecessarily on minors' right of movement, see concurring opinion, infra at 1-2, he agrees that the government's special interest in and authority over children cannot be ignored, see id. at 1, 3. Consequently, his strict scrutiny test must give way in some respect in order to give effect and meaning to the governmental interest. In the end, he would redefine the strict scrutiny test to give special emphasis to the importance of minors' right of movement, and this is already accomplished by the intermediate scrutiny standard. Certainly in the instant case, where all agree that the District has demonstrated a compelling interest in reducing juvenile crime and victimization, the only question is whether the means are sufficiently tailored to respect minor's rights and to remedy the problems. Yet were the bar placed so high that virtual scientific certainty would be required to demonstrate that all other alternatives have proved insufficient, as is implied in Judge Tatel's analysis, see infra concurring opinion at 4 (calling for curfew laws that are "more effective" than other alternatives), the government would be stymied in its efforts to protect juveniles from serious or even deadly harm notwithstanding evidence that other communities have successfully implemented juvenile curfews. Such an approach appears inconsistent with Supreme Court teaching that when parental control fails to provide adequate protection or is absent, the government may intervene, as for example, in Prince, where even though a parent had given permission to a child to act in a certain manner, the government could preclude such action, and in Flores, where the government could restrict minors' release to certain persons, and in Vernonia, where parents were deemed to have designated school officials to act in loco parentis where their children's Fourth Amendment rights were at stake.

The intermediate scrutiny standard, properly understood, does not diminish the importance of the rights at issue, but does acknowledge that the government may have important interests as well, and thus the analysis under intermediate scrutiny will be demanding in its requirement that the means are "substantially related" to achievement of the identified objectives. The instant case illustrates the point. Under the intermediate scrutiny standard, a statute "must serve important governmental objectives and must be substantially related to the achievement of those objectives." Craig, 429 U.S. at 197. As described in sections III.B C, although the Act undoubtedly serves an important governmental goal, its "relation to the state interests it is intended to promote is so tenuous that it lacks the rationality contemplated by the [requirement of equal protection]," Lalli, 439 U.S. at 273, causing it to fail the intermediate scrutiny test.

B.

The first prong of the intermediate scrutiny test is not at issue because the District has doubtless shown an important interest in reducing juvenile crime and victimization sufficient to satisfy an intermediate scrutiny review. The District's interest in enacting a juvenile curfew, as stated in the Act, is to reduce juvenile crime and victimization and to aid parents or guardians "in carrying out their responsibility to exercise reasonable supervision of minors." D.C. Code § 6-2181(e)(1)-(3) (Supp. 1997). The Supreme Court has recognized that the state has a "legitimate and compelling" interest in protecting the entire community, including juveniles, from crime. Schall, 467 U.S. at 264 (quoting DeVeau v. Braisted, 363 U.S. 144, 155 (1960)) (internal quotation marks omitted). It has further recognized that the state "has a strong and legitimate interest in the welfare of its young citizens, whose immaturity, inexperience, and lack of judgment may sometimes impair their ability to exercise their rights wisely." Hodgson v. Minnesota, 497 U.S. 417, 444 (1990). Every federal court to reach this issue in the context of a juvenile curfew statute has found or assumed there to be a compelling state interest in protecting the safety and wellbeing of children and in reducing juvenile crime.

If ever there were a place with an important need to reduce juvenile crime and victimization, it is the District of Columbia. The district court considered the 1995 Kids Count Data Book, which indicates that in 1992, the District of Columbia's violent crime arrest rate for youths aged ten to seventeen was the worst in the nation, and more than three times the national average, at 1,487 violent crime arrests per 100,000 youths. See id. at 49. The District of Columbia also had the worst violent death rate in the nation for teens aged fifteen to nineteen; the District's rate of 269 violent deaths per 100,000 teens was more than four times the national average. See id. Moreover, the problem was worsening: according to statistics from the Juvenile Division of the Office of the Corporation Counsel, between 1987 and 1995, juvenile arrests for aggravated assault increased by 89.8%, for murder by 157%, and for carrying a dangerous weapon by 282.7%. Also, the number of referrals to court for juveniles "in need of supervision" increased by 181.9% between 1990 and 1994. In addition to statistical evidence prepared by the executive and judicial branches of the District government, the D.C. Council received statistics from an opponent of the curfew legislation that showed an alarming number of murders, shootings, and assaults during curfew hours in areas covered by the curfew, although the ages of the perpetrators are not identified and most of the incidents involving minor victims did not occur during curfew hours. Further, elected representatives of Advisory Neighborhood Commissions ("ANCs") and other District residents testified before the D.C. Council about violence plaguing the streets, gunfire from early evening through early morning, children counting the new bullet holes every morning in the doors to their kindergartens, the worsening of teen violence, the gang victimization of youths, and murder becoming sport. Finally, upon removing the sunset provision of the Act, see supra note 4, the D.C. Council had new evidence that juvenile violence was "ever-increasing" and juvenile victimization was "skyrocketing."

Although the evidence upon which the D.C. Council relied was flawed, see infra subsection III.C., it nevertheless reflects a serious problem with juvenile crime and victimization in the District of Columbia. A district corut in a neighboring jurisdiction observed that "the crime rae by jubeniles in the District of Columbia is staggering by any definition." Schleifer, 963 F. Supp. at 546. Because the District presented ample evidence that juvenile crime and victimization are crushing problems for the District of Columbia, it has demonstrated an important government interest sufficient to meet the first prong of intermediate scrutiny review.

C.

Whether the District has demonstrated that the Act survives the second prong of the intermediate scrutiny standard is more problematic. Under intermediate scrutiny, the Act must be "substantially related" to the goals of reducing juvenile crime and victimization. By requiring a very close relationship between purpose and remedy, the court ensures that the legislature enacted its juvenile curfew on the basis of reasoned analysis rather than assumptions. See Hogan, 458 U.S. at 725-26; cf. Wengler, 446 U.S. at 151-52; Carey, 431 U.S. at 696 (plurality opinion of Brennan, J.). The Supreme Court instructs that the court must conduct a "searching analysis" in order to make sure that the legislature has provided sufficient justification for the statute's key provisions. Hogan, 458 U.S. at 728. At this point, the District has failed to provide the necessary justification and the Act thus fails intermediate scrutiny analysis.

Were the rational basis test applied, the Act would undoubtedly pass muster because the District's curfew regime is unquestionably rationally related to its goals of reducing juvenile crime and violence. By requiring that minors in public during curfew hours be accompanied by an adult, the D.C. Council reasonably assumed that adults will normally protect minors in their care and prevent them from victimizing others. In addition, the experience of other jurisdictions facing increases in juvenile crime and victimization indicated to the D.C. Council that a curfew can be a useful tool in fighting such problems. Reports on the specific experiences of Dallas and San Antonio, Texas, and New Orleans, Louisiana, showed that after a juvenile curfew became effective, the number of juvenile arrests for violent offenses decreased, and the Dallas and San Antonio reports also showed reductions in juvenile victimization. The D.C. Council could properly rely on studies by other cities so long as there was a reasonable basis on which to conclude that the studies were relevant. See City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 51-52 (1986).

Testimony before the D.C. Council further confirms that the Act is rationally related to the governmental interest in reducing juvenile crime and victimization. From the law enforcement community, the D.C. Council heard, through a representative of the Community Branch of Community Policing who has taken "ride-alongs" with the Metropolitan Police Department, that the juvenile curfew is "an important tool," although "not an all-inclusive cure," because "it disrupts the gang activity, the drug trade, the hanging out waiting for the right opportunity to commit the crime. It also removes potential drive-by victims from public places where they can be targets." The D.C. Council also received statistical information and reports from the Police Chief on the anticipated effect of a curfew statute. Following enactment of the curfew regime, in one of several reports on implementation of the Act, the Police Chief emphasized the value of the curfew in addressing the problems of truancy and runaways — early indicia of later criminal activity and victimization — and reported on juvenile arrests for violent offenses during curfew hours. In addition, an expert testified in the district court on the nature of peer pressure, opining that the large majority of delinquent acts committed by minors occur when minors are in the company of other minors, without adult supervision, and that, consequently, a curfew would reduce the number of such acts both by decreasing the amount of time minors are unsupervised and by encouraging parental supervision. The Act would certainly pass rational basis review.

Just as certainly, however, the evidence offered by the District to demonstrate that the curfew regime would accomplish its purposes is inadequate to survive strict scrutiny review. As the district court found, the District provided "only scant statistical information on crime in the District committed by and against minors under the age of seventeen." Hutchins, 942 F. Supp. at 675 (emphasis omitted). Such information as it provided was further flawed both because the statistics included persons up to age eighteen and because they did not show the time when incidents occurred, or the ages of the perpetrators or victims, or the places where incidents occurred. See id. at 675-76. As the district court noted, a chart of juvenile arrests during curfew hours prepared by the Metropolitan Police Department was overinclusive because it contained data on minors over age seventeen, was undated, and lacked corroboration from either its author, who was unidentified, or someone familiar with the methodology used to prepare the report. See id. at 677. The legislature did not act on evidence sufficient to withstand strict scrutiny review.

So too, the District failed to provide evidence of sufficient quality to support the Act under intermediate scrutiny review. Although the inquiry into the evidentiary basis for a legislature's impingement of constitutional rights may be somewhat less probing under intermediate scrutiny than under strict scrutiny to the extent that recognition must be given of the heightened governmental interest, the inquiry is still a serious one for a "substantial relat[ion]" must be demonstrated between means and purposes. Where statistical data is employed to justify the salient features of a statute, the court must ensure that the data shows that the "fit" between the statute and its goals is clear, and not "unduly tenuous." Craig, 429 U.S. at 200-03. Statistics establishing broad propositions may not suffice under the intermediate scrutiny test. Thus, in Craig, the Supreme Court held that flawed or only slightly relevant statistical studies could not form the basis for the use of gender as a classifying device. See id. at 200-04. In that case, the Court concluded that the statistical studies offered by the state, while graphically documenting the problem of underage driving while intoxicated, related little to the statute's key provisions, which barred consumption of alcohol by men but not women aged eighteen to twenty. See id. at 191-92, 202-03. Although courts must acknowledge that "matters of practical judgment and empirical calculation are for" the executive and legislative branches of government, since "the precise accuracy of [the state's] calculations is not a matter of specialized judicial competence," the court still must insist on "consistency and substantiality" in the evidentiary data relied on by the state to establish a sufficient fit between its means and goals. Lalli, 439 U.S. at 274 (plurality opinion of Powell, J.) (quoting Mathews v. Lucas, 427 U.S. 495, 516 (1976)) (internal quotation marks omitted).

Admittedly, it may not always be clear how closely a court engaging in intermediate scrutiny should probe the legislature's evidentiary findings. Most of the Supreme Court's decisions involving this standard outside of the First Amendment context have involved statutes or other state actions that have discriminated on the basis of gender rather than age; moreover, one Justice has questioned whether the Court is applying intermediate scrutiny according to its original terms in gender cases. See United States v. Virginia, 116 S.Ct. 2264, 2293-96, 2305-06 (1996) (Scalia, J., dissenting). Even without direct Supreme Court guidance, however, in the instant case it is clear that wherever the precise boundaries of the evidentiary nexus test under intermediate scrutiny are set, the evidentiary record supporting the Act does not satisfy them.

The evidentiary flaws identified by the district court illustrate how the District failed to show that the Act is "substantially related" to the District's goals. The first statistical problem is age-related, age being a key part of the curfew regime. The District's statistics on juvenile arrests and referrals to court for juveniles "in need of supervision," while indicating disturbing trends for all youths, are flawed in that they include youths aged seventeen. See id. at 675; see also D.C. Code § 16-2301(3). As the annual reports of the D.C. courts show, youths aged seventeen and older were responsible for 42% of juvenile referrals for the years 1990 through 1994; a curfew excluding this group lacks a close fit to the goal of reducing juvenile crime. Further, although the District's statistics indicate that its teen violent death rate was skyrocketing, those statistics pertain to youths aged fifteen to nineteen, see supra section III.B; because these statistics do not indicate what percentage of those dying are youths aged seventeen to nineteen, and thus unaffected by the curfew, the statistics offer only weak evidence that the curfew will much reduce the teen violent death rate.

The second statistical problem is temporal, time also being a key provision of the curfew regime. While the District's data on teen violent death demonstrate a devastating trend, neither they, nor the District's national teen victimization data, indicate what time of day or night minors are victimized. Uncontested evidence in the record indicates that most juvenile victimization nationwide occurs shortly after school, around 3:00 or 4:00 p.m. In addition, recent data from the Federal Bureau of Investigation shows that juvenile crime peaks between 3:00 and 8:00 p.m. The only chart before the D.C. Council addressing the time of day juvenile crime occurs, which indicates that half of juvenile arrests occur during curfew hours, was contradicted by other evidence before the district court and, even taken on its own terms, is flawed in that it also includes seventeen-year-olds, who will not be affected by the curfew. See Hutchins, 942 F. Supp. at 677. Further, the District's data do not indicate where juveniles are victimized or where juvenile crime occurs. If a substantial percentage of juvenile victimization and crime occurs within schools, homes, or youth recreation centers, a curfew that does not pertain to those locations will be of little assistance.

The faults in the District's juvenile crime and victimization data are not cured by other record evidence. The Police Chief's report that juvenile arrests for violent offenses declined during three of the months when the curfew was in effect is weakened by evidence that during those three months, the police budget and the size of the force were reduced, and limitations were placed on police overtime, which would logically suggest an alternate cause for the reduction in arrests. See id. at 676. No contrary conclusion was suggested in the record, as the District did not present expert testimony on this point. Cf. Schleifer, 963 F. Supp. at 545. Although properly gathered statistics might show that juvenile crime decreased even while police resources were, hypothetically, held constant, no such statistics are in the record. While the D.C. Council could properly rely on the experiences of other jurisdictions as evidence of the efficacy of juvenile curfews in general, see Renton, 475 U.S. at 51-52, the District presented no evidence that those other jurisdictions are sufficiently similar to the District of Columbia that a curfew designed like theirs will produce similar results here; indeed, specialized programs elsewhere may have also been a cause for other curfews' success, but the District offered no evidence either that the District has similar programs or that the absence of similar programs is insignificant. Without some evidentiary explanation for discrepancies, the District is impermissibly relying on bare assumptions. Cf. Hogan, 458 U.S. at 725-26; Wengler, 446 U.S. at 151-52; Carey, 431 U.S. at 696 (plurality opinion of Brennan, J.). Because the flaws in the legislative record pertain to such fundamental features of the curfew — the age of those covered and the hours of coverage — the court has no choice but to conclude that it cannot presume that the D.C. Council was acting in a sufficiently precise and reasoned fashion to withstand intermediate scrutiny.

Courts upholding similar curfews, albeit under a strict scrutiny standard, were not confronted with evidentiary deficiencies of this dimension. As the district court pointed out, the city of Dallas offered, in support of its curfew, evidence that juvenile crime increases proportionally with age between ten years old and sixteen years old, and evidence of the times of day and the places in which violent crime by all perpetrators occurs. See Hutchins, 942 F. Supp. at 679 (citing Qutb, 11 F.3d at 493). Although the city of Dallas was unable to provide data concerning the precise number of juveniles arrested or victimized during curfew hours, its data demonstrated a closer fit between its curfew and its goals because it at least included specific statistics concerning the age group affected by the curfew and about the times and places in which minors, as well as other victims, most commonly fall prey to violent crimes. See Qutb, 11 F.3d at 493 n. 7.

Similarly, the city of Charlottesville, Virginia, presented expert evidence describing national trends in juvenile crime and victimization, and confirming that Charlottesville's trends mirrored the national data and that the curfew law in Charlottesville was among the most modest and lenient nationwide. See Schleifer, 963 F. Supp. at 544. The Charlottesville Commonwealth Attorney supported the expert's testimony with data on juvenile crime in Charlottesville showing that the most serious crimes occur during nighttime hours, and explaining the link between the national statistics and juvenile crime and victimization in Charlottesville. See id. at 544-45. The Commonwealth Attorney also explained at a public hearing why the inclusion of seventeen-year-olds in crime and victimization data did not distort the statistics. See id. at 545. In addition, the city offered evidence comparing the incidents of violent crime during curfew and non-curfew hours and expert testimony that juveniles face more dangers during curfew hours than during the day. See Schleifer v. City of Charlottesville, No. Civ. A. 97-0021-C, 1997 WL 375542, at *2 (W.D. Va. May 20, 1997) (denying permanent injunctive relief). The city further demonstrated, to the satisfaction of the district court, that there was "considerable and careful deliberation and extensive research" underlying the enactment of the curfew. Id.

By contrast, the District offered no expert testimony explaining why the inclusion of seventeen-year-olds in crime data did not distort the data's relevance to a curfew applying to youths under the age of seventeen. Nor did the District offer an explanation for plainly contradictory statistics; although reasonable explanations of the discrepancies might exist, in their absence, the statistics appear largely meaningless. Unlike either Dallas or Charlottesville, the District also did not provide evidence of the places where violent crimes are likely to occur. Nor did the District provide expert testimony to explain the link between national and local statistics, or to explain the enhanced risks youths face in public at night.

Crafting a constitutional curfew is not an easy task. The tensions between minors' rights and governmental interests are interwoven and complex. Some of the "defenses" in the Act cause gaps that would appear to undermine achievement of the results sought by the District: a juvenile can be attacked even while running errands without detour or stop, exercising First Amendment rights, returning home from city-sponsored events, or, for that matter, being married. Yet filling these gaps would place greater restraints on movement than the Act imposes, and such "defenses" or exceptions have proven important to courts analyzing a curfew's constitutionality. In one sense, this paradox might lead to the conclusion that the close-fit requirement for curfews may always prove insufficiently precise. But even if the problem is vexing, courts have not suggested that a "sufficiently related" nexus cannot be achieved in a statute, much less that, under strict scrutiny, a "narrowly tailored" statute is impossible. In any event, the paradox is not presented here. The District may be able to present sufficient evidence to explain why the contradictions and under and overinclusiveness do not matter. For now, the Act is impaled on unexplained evidence suggesting that seventeen-year-olds are committing a large percentage of juvenile crime and that much crime involving juveniles occurs before 8 p.m. The Act would do virtually nothing to address these law violations. Although well-crafted defenses may minimize the intrusive effect of a curfew regime, see Qutb, 11 F.3d at 49395, the statistical and other data deficiencies going to the heart of the District's curfew regime force the conclusion that the Act may be overinclusive despite the ameliorative effect of its defenses.

Given the deficiencies in key evidence before the D.C. Council and the district court, the Act does not survive intermediate scrutiny as it affects minor appellees' equal protection and due process rights. This is not to say that the D.C. Council is precluded from relying on the experiences of other cities in formulating curfew policy choices, cf. Renton, 475 U.S. at 51-52, or that it must produce "scientifically certain criteria of legislation," Ginsberg, 390 U.S. at 642-43 (quoting Noble State Bank v. Haskell, 219 U.S. 104, 110 (1911)) (internal quotation marks omitted), or that it is not to be allowed a reasonable opportunity to experiment with solutions to serious problems, see Renton, 475 U.S. at 52. Neither is it to suggest other than that it is for the District government and not the courts to weigh alternative policies in light of supporting data. See Lalli, 439 U.S. at 274. But when a statute impinges upon minors' constitutionally protected rights, there must be persuasive evidence that the problem will be addressed by the legislative solution. See Hogan, 458 U.S. at 731; Lalli, 439 U.S. at 274. Otherwise, the statute has no more than a tenuous connection to the problem it seeks to address and the legislature has acted in an insufficiently reasoned fashion. Hence, notwithstanding any greater latitude afforded a legislature under an intermediate standard than under strict scrutiny, the loose ends — especially those relating to the age group that is most often victimized or arrested, and the time and place most juvenile crime and victimization occurs — leave a court to speculate that the curfew regime would likely achieve its goals. Intermediate scrutiny requires more.

Accordingly, in light of Judge Tatel's concurrence in the judgment upon concluding that the Act fails to survive strict scrutiny, we affirm the judgment of the district court holding the Act unconstitutional.