(dissenting, with whom Spina and Cowin, JJ., join). I agree with the court that the defendant was sentenced to probation after the effective date of the statute, G. L. c. 265, § 47, inserted by St. 2006, c. 308, § 8 (Act).1 Ante at 562-563. I write separately because I disagree with the court’s conclusion that the statute should be deemed penal and, therefore, that it would be a violation of the ex post facto clause to apply it to this defendant, who was convicted of committing a sex offense before the statute’s enactment. Ante at 560, 572.
As the court notes, in determining whether a statute constitutes an ex post facto law, we must discern whether the Legislature *574intended the statute to be a civil remedy or a criminal penalty. Ante at 565. In its analysis, the court concludes, ante at 567, that it is not clear whether the Legislature intended the statute to be criminal or civil and, therefore, the defendant should be relieved of his burden to provide “ ‘the clearest proof’ that ‘the statutory scheme [is] so punitive either in purpose or effect as to negate [the State’s] intention’ to deem it ‘civil.’ ” Commonwealth v. Bruno, 432 Mass. 489, 500 (2000), quoting Kansas v. Hendricks, 521 U.S. 346, 361 (1997). Although I agree with the court that the Act creating the statute contains both civil and criminal elements, I do not agree that the legislative intent is so unclear that we cannot conclude that the intent was regulatory and remedial.
General Laws c. 265, § 47, concerns a term of probation. The principal goals of probation itself are the rehabilitation of the probationer and the protection of the public. Commonwealth v. Lapointe, 435 Mass. 455, 459 (2001), citing Commonwealth v. Pike, 428 Mass. 393, 403 (1998). An important aspect of the statute is that it does not apply to all probationers, but only to those who have committed certain sex offenses, as defined in the sex offender registration law, G. L. c. 6, §§ 178C-178Q. Both the Legislature and this court have recognized that sex offenders present a serious threat to the public and have a high rate of recidivism. Doe, Sex Offender Registry Bd. No. 8725 v. Sex Offender Registry Bd., 450 Mass. 780, 789 (2008). Doe, Sex Offender Registry Bd. No. 3844 v. Sex Offender Registry Bd., 447 Mass. 768, 769 (2006). Coe v. Sex Offender Registry Bd., 442 Mass. 250, 258-260 (2004). Commonwealth v. Knapp, 441 Mass. 157, 159, 164 (2004). Determining the most effective way to control problems affecting public safety is left to the Legislature. Luk v. Commonwealth, 421 Mass. 415, 429 (1995).
Although it is true, as the court states, ante at 566, that the statute lacks a statement of civil intent, is imposed only in the criminal context, is placed in the criminal code, and is uniformly imposed on every defendant who is placed on probation for certain sex offenses, I conclude that these elements alone are not dispositive of whether the Legislature’s intent is so “unclear” as to relieve the defendant of his burden.
That a statute “may be ‘tied to criminal activity’ is ‘insuf*575ficient to render the statute punitive.’ ” Commonwealth v. Bruno, supra at 501, quoting Kansas v. Hendricks, supra at 362. The location and labels of a statute do not by themselves transform a civil remedy into a criminal one. Smith v. Doe, 538 U.S. 84, 94 (2003) {Smith). Even if the objective of a statute is consistent with the purposes of a State’s criminal justice system, “the State’s pursuit of it in a regulatory scheme does not make the objective punitive.” Id. In addition, a legislative restriction that is incident to the State’s power to protect the public will be considered regulatory rather than punitive. Smith, supra at 93-94, quoting Flemming v. Nestor, 363 U.S. 603, 616 (1960). Protecting the public from sex offenders is a nonpunitive objective. Smith, supra at 102-103.
Here, even though the location of the Act under which this statute falls is the criminal code and the Act’s title, “An Act increasing the statute of limitations for sexual crimes against children” could be read as criminal, one provision establishes criminal penalties for level three sex offenders who knowingly and willingly establish residence in certain convalescent or nursing homes or facilities for the mentally retarded. Another provision concerns a statute of limitations. See respectively St. 2006, c. 303, §§ 6, 9. Four provisions can be read as regulatory rather than criminal. See St. 2006, c. 303, §§ 1, 2, 3, 10 (reducing number of days for certain sex offender classification and registration requirements; raising fee). One section (§ 5) mandates deleting a subsection of G. L. c. 178H, § 3, concerning registration by homeless persons, and arguably is administrative in nature. The remaining three sections concern community parole supervision (§ 4) and GPS monitoring for parolees and probationers (§§ 7 and 8), which are tied, at least in part, to the offender’s victims.
The Act’s title references children. General Laws c. 265, § 47, itself (§ 8 of the Act), ties the global positioning system (GPS) requirement to the offender’s victim by stating that the Commissioner of Probation must establish “geographic exclusion zones” that must include the “victim’s residence, place of employment, school and other areas defined to minimize the probationer’s contact with children, if applicable.” This language clearly indicates that the Legislature’s concern was protection of the victim. Moreover, although it does not have its own statement *576of civil intent, the statute references G. L. c. 6, § 178C, for the definitions of sex offenders who fall under its purview. The preamble to the sex offender registry statute states that the statute’s purpose is to protect the public. St. 1999, c. 74, preamble & § 1. This court has found that purpose to be remedial. Doe v. Attorney Gen., 430 Mass. 155, 166, 168 (1999). The presumption that the Legislature is aware of the content of other statutes holds particularly true of statutes the Legislature specifically references. See Thurdin v. SEI Boston, LLC, 452 Mass. 436, 444 (2008); Suli-veres v. Commonwealth, 449 Mass. 112, 116 (2007), citing Commonwealth v. Callahan, 440 Mass. 436, 440-441 (2003).
In addition, other laws that establish regulations for sex offenders who have finished serving their sentences have been deemed to be remedial, including the confinement of those persons determined to be sexually dangerous. Commonwealth v. Bruno, 432 Mass. 489, 501-502 (2000) (commitment of sexually dangerous persons remedial). See, e.g., Doe, Sex Offender Registry Bd. No. 8725 v. Sex Offender Registry Bd., 450 Mass. 780, 787-789 (2008) (noting that sex offender registry laws are generally regulatory); Coe v. Sex Offender Registry Bd., 442 Mass. 250, 259-260 (2004), citing Smith, supra at 105 (posting information concerning level three sex offenders on Internet nonpunitive); Opinion of the Justices, 423 Mass. 1201, 1227 (1996) (proposed community notification of sex offender information regulatory and remedial).
Moreover, I do not agree with the court’s conclusion that, because the statute applies to all sex offender probationers without regard to individual dangerousness, the statute is therefore criminal. Ante at 572. “The State’s determination to legislate with respect to convicted sex offenders as a class, rather than require individual determination of their dangerousness, does not make a statute punitive under the Ex Post Facto Clause.” Smith, supra at 104.2 “The risk of recidivism posed by sex offenders is ‘frightening and high’ ”; “[w]hen convicted sex offenders reenter *577society, they are much more likely than any other type of offender to be rearrested for a new rape or sexual assault.” Id. at 103, quoting McKune v. Lile, 536 U.S. 24, 33, 34 (2002). Here, therefore, the Legislature could decide that, of all probationers, those convicted of sex offenses have a substantial risk of recidivism and, accordingly, mandate the use of a tool that assists law enforcement in protecting the public while the offender is completing his sentence. Smith, supra at 103. Moreover, the United States Supreme Court has upheld other laws without a corresponding risk assessment. Smith, supra at 104, citing Hawker v. New York, 170 U.S. 189, 197 (1898), and De Veau v. Braisted, 363 U.S. 144, 160 (1960).
I conclude that the legislative intent was to establish a regulatory rather than criminal regime, and I would not relieve the defendant of his burden to show that the statute is so punitive as to be penal. However, even assuming the Legislature’s intent is not clear, I conclude that, on balance, this statute establishes a nonpunitive regime to protect the public.
This court’s analysis of the ex post facto clause has been identical to that of the United States Supreme Court. Commonwealth v. Bruno, 432 Mass. 489, 492 n.4 (2000). The Supreme Court has stated that, where there is no conclusive evidence of legislative intent, courts should analyze the statute on its face using the seven factors listed in Kennedy v. Mendoza-Martinez, 372 U.S. 144, 168-169 (1963). However, absent an express showing of an intent to punish, the determination whether a statute is so punitive in purpose and effect that it negates any remedial purpose turns on the last two Mendoza-Martinez factors, whether the statute is rationally related to a purpose other than punishment and whether the effects are excessive in rela*578tion to that purpose. See Schall v. Martin, 467 U.S. 253, 269 (1984), quoting Kennedy v. Mendoza-Martinez, supra; Bell v. Wolfish, 441 U.S. 520, 538 (1979). See also Smith, supra at 102-103 (stating that “most significant” factor in determining that Alaska’s sex offender registration statute was not ex post facto law was its rational connection to nonpunitive purpose of protecting public from sex offenders).
Furthermore, this court has noted that, although the Mendoza-Martinez factors can be useful guidelines, without weight and priorities assigned to them, there are risks that the factors will be unmanageable for their “indefiniteness.” Opinion of the Justices, 423 Mass. 1201, 1222 (1996). The court also noted, however, that it considered that the more harshly a statute bears on an individual, the more closely it resembles a criminal sanction, and the more urgent the regulatory concern and the more soundly the statute is rooted in fact and not conjecture, the more regulatory the aim. Id. at 1224.
Although our analysis of the ex post facto clause is identical to that of the Supreme Court, here, relying on dissenting opinions in Smith, this court concludes that the finding of scienter and whether the behavior to which the statute applies is a crime weigh heavily in favor a determination of punitive effect. Ante at 569. However, in the Smith case, the majority stated that its assessment of Alaska’s sex offender registration law was not aided by these two factors because the scheme applied to past conduct that was and is a crime. Smith, supra at 105. In any event, I conclude that the statute does not create culpability for prior conduct; nor does it require a finding of scienter to subject a person to the statute’s terms.
The determination whether the statute is an affirmative restraint must be made in the context that all probation has a supervisory element to ensure that the probationer abides by its terms. Commonwealth v. Taylor, 428 Mass. 623, 626 (1999). In any event, restraint is not automatically punitive, and even non-punitive confinement to protect the public is a nonpunitive, legitimate government objective. Kansas v. Hendricks, 521 U.S. 346, 363 (1997). Indeed, the Court has upheld pretrial confinement as nonpunitive, where the objective was to ensure that a potentially dangerous defendant appeared for trial. See Schall v. Martin, *579supra at 272-274; Bell v. Wolfish, supra at 536-537 (restraint not necessarily punitive).
Here, the probationer must wear the device only during the term of his probation, a condition that is limited in duration and which already has restrictions that are related to the supervisory responsibility of the probation department.3 See Commonwealth v. Durling, 407 Mass. 108, 111 (1990). Although a probationer may be restricted from some geographic areas related to public safety, the statute does not otherwise prohibit the probationer from going about his or her daily business while on probation. Moreover, although entrance into an exclusion zone is a violation of probation, it is not a crime. See Commonwealth v. Wilcox, 446 Mass. 61, 65 (2006).
Concerning whether the statute promotes the traditional aims of punishment, retribution, and deterrence, the statute would function as a deterrent. However, statutes that have a deterrent or retributive effect may be regulatory, and burdens on the defendant that result from their imposition “only violate the ex post facto clauses ... if they must be deemed punishment.” Opinion of the Justices, supra at 1226. Virtually all civil regimes have some deterrent effect. Id. at 1222-1223, citing United States v. Ursery, 518 U.S. 267, 284 n.2 (1996). “To hold that the mere presence of a deterrent purpose renders such sanctions ‘criminal’ . . . would severely undermine the Government’s ability to engage in effective regulation.” Smith, supra at 102, quoting Hudson v. United States, 522 U.S. 93, 105 (1997).
It is true, as the court points out, ante at 570 n.18, that the GPS device could be seen by the public,4 and retribution could *580be visited on the probationer. However, retribution must be the purpose of the statute in order for it to be deemed punitive. Opinion of the Justices, supra at 1227. We have held that posting information about level three sex offenders on the Internet, although it could invite retribution, served a public safety goal and was not punitive. Coe v. Sex Offender Registry Bd., 442 Mass. 250, 259-261 & n.9 (2004), citing Smith, supra at 91, 105 (actual audience are those who would be offenders’ potential victims). Unlike posting information identifying certain sex offenders on the Internet, here, it is not at all certain that the public would know that the reason the offender was wearing the GPS device was because he was a sex offender, if they were to see the device at all. See generally Commonwealth v. Donohue, 452 Mass. 256, 257, 269 (2008) (sheriff has authority to discharge or parole prisoners in special program where they wear GPS device); Commonwealth v. Morasse, 446 Mass. 113, 113-114 (2006) (GPS used for pretrial home confinement of defendant indicted on various charges); Commonwealth v. Hector H., 69 Mass. App. Ct. 43, 44 (2007) (juvenile found delinquent by reason of trespassing placed on probation with electronic monitoring for one year).5
Turning now to the most important of the Mendoza-Martinez factors, I agree with the court that the “GPS requirement, with its instruction to designate exclusionary zones that are off limits to the sex offender, supports the rational purpose of keeping offenders away from their past victims and, more generally, the purpose of hmiting opportunities for reoffense.” Ante at 572. However, I do not agree with the court’s conclusion that because it applies, without exception, to convicted sex offenders without regard to whether they are individually dangerous, the burden on the probationer is excessive in relation to the purpose of protecting the public. Id., citing Smith, supra at 116-117 (Ginsburg, J., *581dissenting). As discussed, the majority in the Smith case stated that to treat sex offenders as a class is not punitive where it is rational to conclude that there is a high rate of recidivism among sex offenders. Smith, supra at 103-104. This court has stated that recidivism among sex offenders is high and that protection of the public is a compelling State interest. See Coe v. Sex Offender Registry Bd., 442 Mass. 250, 259-260 (2004); Commonwealth v. Knapp, 441 Mass. 157, 159, 164 (2004). See also Doe, Sex Offender Registry Bd. No. 8725 v. Sex Offender Registry Bd., 450 Mass. 780, 789 (2008); Doe, Sex Offender Registry Bd. No. 3844 v. Sex Offender Registry Bd., 447 Mass. 768, 769 (2006). Moreover, “[a] statute is not deemed punitive simply because it lacks a close or perfect fit with the nonpunitive aim it seeks to advance . . . [unless] the nonpunitive purpose is a ‘sham or mere pretext.’ ” Smith supra at 103, quoting Kansas v. Hendricks, 521 U.S. 346, 371 (1997) (Kennedy, J., concurring). Here, there is no evidence of sham or pretext. See Luk v. Commonwealth, 421 Mass. 415, 422 (1995) (sanctions are punitive only if they cannot be explained by nonpunitive purpose).
Given that the urgency of the regulatory concern of protecting the public from sex offenders has been found by both the Legislature and this court, and that here the statute covers sex offenders who have yet to serve their sentences, I conclude that on balance this statute is regulatory, despite the burdens on the sex offender probationer, and thus does not violate the ex post facto clause. See Opinion of the Justices, supra at 1223-1224 (court should weigh statute’s harshness on individual against urgency of regulatory concern).6
In his memorandum of decision and order, the Superior Court judge held that the statute applied retroactively to sex offenders who were sentenced to probation before the statute’s effective date. Because the defendant was placed on probation after the statute’s effective date, I would leave to another day the issue whether the words “is placed on probation” applies to the day probation was imposed or when the probation commences.
The court states, ante at 566 n.ll, that my reliance on Smith v. Doe, 538 U.S. 84, 104 (2003) (Smith), is mistaken because, in the Smith case, in response to the reasoning of the lower court, the United States Supreme Court distinguished Alaska’s sex offender registration requirement from probation, stating, “Probation . . . entail[s] a series of mandatory conditions and allow[s] the supervising officer to seek the revocation of probation ... in case of *577infraction” but, in the case of registration, failure to report results in criminal prosecution where the “proceeding [is] separate from the individual’s original offense.” Id. at 102. The Court’s discussion of probation in the Smith case is not dispositive here. Specific conditions of probation were not at issue in the case. In addition, the Court did not say that any 'statute that imposes a condition of probation, the violation of which would result in a proceeding related to the defendant’s original offense, is per se a violation of the ex post facto clause.
Moreover, the Court’s determination that a legislature can identify classes of persons to regulate without violating the ex post facto clause is based on the police powers of a State. Id. at 104, quoting Hawker v. New York, 170 U.S. 189, 197 (1898). The Hawker case involved a law excluding convicted felons from practicing medicine. Hawker v. New York, supra at 190, 192-193.
I acknowledge that, as illustrated by this defendant’s first probation of twenty-five years for a sexual offense against a child, probation may be for a long duration. The length of probation is within the discretion of the judge. See Commonwealth v. Power, 420 Mass. 410, 413-414 (1995), cert, denied, 516 U.S. 1042 (1996), citing Commonwealth v. Goodwin, 414 Mass. 88, 92 (1993) (judge has great latitude in sentencing where sentence imposed is within limits of applicable statute). I also note that, elsewhere, the Legislature has determined that offenders who commit certain sex crimes are subject to more stringent requirements. See, e.g., G. L. c. 6, § 178K (2) (d), prohibiting offenders who commit a sex offense involving a child or a sexually violent offense from process whereby sex offender could be relieved “of any further obligation to register.”
The statute itself indicates that the terms of the GPS monitoring will be administered by the Commissioner of Probation and that the commissioner *580and one local police department would be informed of a probationer’s whereabouts only if a violation occurs. G. L. c. 265, § 47. There is nothing in the statute that authorizes making public any information concerning a probationer.
As the court notes, there is no historical version of this GPS statute to assist in resolving whether a GPS has been historically regarded as punishment. Ante at 571. See Smith v. Doe, 538 U.S. 84, 97-99 (2003) (rejecting argument that dissemination of sex offender information over Internet is similar to historic shaming or branding punishments that involved setting up direct confrontation between offender and public).
Because I conclude that this statute does not violate the ex post facto clause, a due process analysis is necessary because it applies to a crime for which the defendant was convicted before its enactment. See Doe, Sex Offender Registry Bd. No. 8725 v. Sex Offender Registry Bd., 450 Mass. 780, 788 (2008), quoting American Mfrs. Mut. Ins. Co. v. Commissioner of Ins., 374 Mass. 181, 191 (1978) (court examines statute by considering nature of public interest that motivated Legislature to enact it, nature of rights affected, and extent or scope of statutory effect). I conclude that the statute does not violate a probationer’s due process rights. I have already discussed the compelling nature of the public interest and the extent and scope of the effect on the probationer during his term of probation. Concerning the nature of the probationer’s rights, I conclude that they do not outweigh the compelling public interest where probationers have a limited liberty interest. Commonwealth v. Wilcox, 446 Mass. 61, 64 (2006).