640 N.W.2d 351 (2002)
In the Matter of the Risk Level Determination of R.B.P.
No. C4-01-808.
Court of Appeals of Minnesota.
March 5, 2002.
John M. Stuart, State Public Defender, Cathryn Middlebrook, Assistant Public Defender, Minneapolis, MN, (for relator R.B.P.).
Mike Hatch, Attorney General, Angela M. Helseth, Assistant Attorney General, Minneapolis, MN, (for respondent).
Considered and decided by HANSON, Presiding Judge, LANSING, Judge, and KALITOWSKI, Judge.
OPINION
KALITOWSKI, Judge.
By writ of certiorari, relator challenges the Administrative Law Judge's order affirming the End of Confinement Review Committee's classification of him as a risk level III arguing (1) it was clearly erroneous because the End of Confinement Review Committee exceeded its authority under the statute by assigning him to a risk level two levels higher than the level based on his Sex Offender Screening Tool score; and (2) the evidence does not support the determination that relator should be assigned a risk level III.
FACTS
The End of Confinement Review Committee (ECRC) had evidence that relator R.B.P.'s criminal sexual conduct began when he was 13 years old. Beginning in 1979 relator began abusing a four-year-old and a seven-year-old whom he was babysitting. Because the victims' parents refused to cooperate with the investigation, R.B.P. was never charged with these offenses.
In 1983, at age 17, relator broke into his neighbor's home and, while wearing a nylon stocking over his head, threatened the victim with a knife and raped her approximately four times over the course of two and a half hours. Relator was convicted of two counts of first-degree criminal sexual conduct and burglary and sentenced to 90 months. While incarcerated, relator acquired 12 misconduct charges and refused to participate in sex offender rehabilitation. Relator was released from custody in June 1988, but was placed back in custody due to serious chemical abuse. He was again discharged in 1989.
In 1992, relator called a friend of his girlfriend and told her he needed help bailing his girlfriend out of jail. When the friend came over relator grabbed her and dragged her into the house while the friend's small child remained in her car. Relator threatened her with a long serrated knife and when the victim attempted to escape he caught her by her neck with his shirt and choked her. He cut her several times but eventually let her go. There were ropes tied to the bed and relator subsequently admitted he intended to tie the victim up and rape her.
Relator was convicted of first-degree attempted criminal sexual conduct and sentenced to 134 months. Both sex offender and chemical dependency treatment programs were recommended. In 1997, he was accepted into a sex offender treatment program, which he quit in February 1999. And in January 2000, when he reapplied to the program, the evidence indicates he lied several times on the intake questionnaire.
While in treatment, relator told his psychologist that he had physically abused other women. In addition, he admitted choking and forcing sex on his girlfriend. One of relator's psychologists stated that relator described his assaults in a "very cold and distant manner[.]" The evidence indicates that relator was considered for civil commitment at this time, but due to the length of time between the offenses he was not committed, despite a determination that he was likely to reoffend.
Based on relator's score on the Minnesota Sex Offender Screening Tool-Revised (MnSOST-R) relator was presumptively a risk level I. But because there were "special concerns" about relator's risk to the community, a Department of Corrections psychologist recommended that the ECRC assign relator a risk level of II. Following its meeting in December 2000, the ECRC, citing additional special concerns, assigned relator a risk level of III. Relator challenged this assessment and was afforded a hearing before an Administrative Law Judge (ALJ) in February 2001. The ALJ issued an order in April 2001 affirming the ECRC's risk level determination.
ISSUES
1. Did the End of Confinement Review Committee exceed its statutory authority by assigning relator a risk level III where his MnSOST-R score placed him at a risk level I?
2. Does the evidence considered by the End of Confinement Review Committee warrant the committee's determination that relator is a risk level III?
ANALYSIS
I.
Questions of statutory interpretation are reviewed de novo. Matter of Risk Level Determination of C.M., 578 N.W.2d 391, 395 (Minn.App.1998). But, a reviewing court affords substantial deference to an administrative agency's interpretation of its own rules and regulations. St. Otto's Home v. Minn. Dep't of Human Servs., 437 N.W.2d 35, 40 (Minn.1989). "If an administrative agency's authority is questioned, [a reviewing] court independently reviews the enabling statute." Weber v. Hvass, 626 N.W.2d 426, 431 (Minn.App.2001).
This case arises under the Sex Offender Community Notification Act, Minn. Stat. § 244.052 (1996 & Supp.1997). The Act provides that at least 90 days before an offender's release from prison, an ECRC shall convene to determine the offender's risk level. Id. at subd. 3(d). The ECRC determines the public risk posed by the offender on a case-by-case basis. Id. at subd. 3(a). "An offender assigned * * * to risk level II or III * * * has the right to seek administrative review of [the ECRC's] risk assessment determination," but on review the offender has "the burden of proof to show, by a preponderance of the evidence, that the end-of-confinement review committee's risk assessment determination was erroneous." Id. at subd. 6(a), (b).
The notification statute empowers the law enforcement agency in "the area where the sex offender resides, expects to reside, is employed, or is regularly found," to disclose to the public, any information that the agency deems "relevant and necessary to protect the public and to counteract the offender's dangerousness." Id. at subd. 4(a). The statute mandates that the agency consider the statute's guidelines that information on level I offenders be disclosed to other law enforcement agencies and victims of or witnesses to the offender's crime; that information on level II offenders further be disclosed to schools, daycare centers, "establishments and organizations that primarily serve individuals likely to be victimized by the offender," and individuals who fit the offender's pattern of victim preference; and that in the case of a level III offender, "the agency also may disclose the information to other members of the community whom the offender is likely to encounter." Id. at subd. 4(b).
To administer the statute, the Department of Corrections developed and revised a risk assessment scale to assign weights to the statutory risk factors, which is known as the MnSOST-R. Minn. DOC, Policy Number 205.220 (Dec. 20, 1999). Under the MnSOST-R, an offender who has a score of 3 or lower is a risk level I and an offender with a score of 3 or lower who causes "special concern" is a risk level II. Id. An offender starts the process with a presumptive score based on MnSOST-R from which the committee can depart downward or upward. According to the ECRC treatment professional, if there is an upward determination, the committee notes "special concerns," which are factors that were not considered in arriving at the offender's MnSOST-R score. In addition, the treatment professional testified that after the offender is assigned a MnSOST-R score, the committee interviews the offender to gather information not contained or fully explained in the offender's file. Following the interview, the ECRC prepares a risk assessment report specifying the offender's risk level and the reasons for the committee's risk assessment decision. Minn. Stat. § 244.052, subd. 3(f).
Here relator scored a 1 on the MnSOST-R but had numerous factors cited by the committee as causing "special concern." Relator argues that in light of his MnSOST-R score, the ECRC exceeded its authority and violated its own policy by assigning him a risk level III. We disagree.
The object of all interpretation and construction of laws is to effectuate the legislature's intent. Minn.Stat. § 645.16 (2000). "Every law shall be construed, if possible, to give effect to all its provisions." Id. Absurd and unreasonable results are presumed to be against the legislature's intent. Minn.Stat. § 645.17(1) (2000).
Minnesota courts have not interpreted the Sex Offender Community Notification Act to determine the degree of discretion the ECRC has in determining an offender's risk level. It is significant that the Act does not state that the committee has no discretion in assigning offenders to risk levels. Nor does the statute provide that the committee is limited to using an offender's MnSOST-R score in assigning an offender to a risk level.
We reject relator's argument that this court's holding in Matter of Risk Level Determination of C.M., 578 N.W.2d 391, severely limits the ECRC's level of discretion. That case is not controlling because it involves the issue of providing community notification concerning an individual who has not been proven to be a sex offender. Here, relator is a convicted repeat sex offender whose specific situation is not accurately reflected by his MnSOST-R score.
Because the statute states that risk levels are to be determined on a case-by-case basis in consideration of the risk the offender poses to the public, it is apparent the legislature intended that the committee have the authority to exercise some discretion. In addition, we note that the committee is made up of the head of the correctional facility where the offender is confined, a law enforcement officer, a caseworker experienced in assessing sex offenders, and a victim's services professional. If the sole function of the committee was merely to compute a numerical score based on objective criteria, then a committee with the statutorily specified expertise and experience would not be necessary. Moreover, if the process was purely objective, the committee's interview of the offender, the purpose of which is to gain information of a subjective nature, would also be unnecessary. We thus conclude that based on the statutory scheme, the intent of the legislature was to vest some discretion in the ECRC.
A review of how courts in other jurisdictions have addressed this issue further supports our conclusion. New York has a statute similar to Minnesota's establishing a "board of examiners of sex offenders," N.Y. Correct. Law § 168-l(1) (McKinney Supp.1999-2000), which evaluates offenders prior to their release from custody. Id. § 168-l(6). That board recommends whether offenders will be deemed sexually violent predators subject to community notification. Id. Similar to the Minnesota Notification Act, the New York statute established three classifications of offenders, with corresponding degrees of notification. Id.
The New York Superior Court in People v. Lombardo, 167 Misc.2d 942, 640 N.Y.S.2d 995 (1996), stated the numerical values based on risk factors are presumptive only. Id. at 996. The court went on to state:
"The ability to depart is premised on a recognition that an objective instrument, no matter how well designed, will not fully capture the nuances of every case * * *;" a court is therefore permitted to bring its sound judgment and expertise to bear on an otherwise coldly objective exercise which seeks to quantify that which may prove to be highly subjective.
Id. (quotation omitted).
In addition, the Lombardo court found the objective instrument does not account for conduct unless it falls completely within the boundaries defined by a particular risk factor, and therefore the similar conduct does not factor into the sex offender's risk level score. Id. at 996-97.
Similarly, a New Jersey court found its notification statute was not a scientific device, but rather merely a useful tool to help determine an offender's risk level. Matter of C.A., 146 N.J. 71, 679 A.2d 1153, 1171 (1996). It held that while the classification should be afforded deference, courts should not rely solely on an offender's point total when conducting a judicial review of a risk level classification or notification decisions because it is impossible to create an all-inclusive scale. Id.
[T]here is still a value judgment that must be made when determining a registrant's risk of re-offense and proper classification. * * * [T]here may be cases in which the registrant presents subjective criteria that would support a court not relying on the tier classification recommended by the Scale. In those cases, we do not expect the court to blindly follow the numerical calculation provided by the Scale, but rather to enter the appropriate tier classification.
Id.
We are persuaded by the reasoning of the New York and New Jersey courts. While the MnSOST-R score provides a presumptive risk level, this presumption can be rebutted by individual circumstances not taken into account by it. And the ECRC necessarily has the authority to consider these circumstances when making its ultimate risk assessment. Because the legislature intended the ECRC to have discretion to determine the offender's appropriate risk level, the ALJ properly determined the ECRC did not exceed its statutory authority in assigning relator to a risk level two levels higher than the presumptive level based on his MnSOSTR score.
II.
We also reject relator's argument that the information regarding relator considered by the ECRC was not sufficient to justify the committee's assignment of a risk level III. In arriving at its determination the committee cited several factors that are appropriate as special concerns.
First, the ECRC asked relator about two prior uncharged offenses, including a charge of molesting children he babysat. This additional information regarding relator's criminal sexual history was not included in relator's MnSOST-R score. In addition, the committee noted a special concern with relator's ability to remain sober. This is significant because there was evidence that chemical abuse played an important role in relator's past criminal behavior.
The committee also noted that both of relator's criminal convictions were a result of very serious offenses that involved weapons. On one occasion he entered a house wearing a mask and sexually assaulted the victim for several hours. And on another occasion he lured an acquaintance into his home and sexually assaulted her. Breaking into the victim's house and assaulting a known victim are both highly predatory offenses that support a risk level III. Moreover, because during the second assault he cut the victim with a knife, the committee appropriately expressed concern that any future victim could be seriously injured.
In addition, because one of relator's assaults included a break-in of the victim's residence, the committee reasoned that the statute would not be beneficial to public safety if the neighborhood could not be alerted and therefore relator should be assigned a risk level III.
Finally, the ECRC's review raised additional questions about the accuracy of relator's MnSOST-R score. Specifically, the committee was concerned that relator's MnSOST-R score was lowered because his particular offense did not fit into the predetermined categories. For example, relator received a score of 0 out of a possible 7 because he had been charged or convicted of sexual offenses against only one age group. But the committee had evidence that relator did sexually assault more than one age group but that because the victims' parents would not help with the prosecution, the charges were dropped. Also, although relator received a "negative-one" for not committing a sexual assault against a stranger, the two adult offenses were perpetrated against persons who were only slight acquaintances: a neighbor and a friend of a friend. Finally, relator received another "negative-one" out of a possible 15 for completing the sex offender program. But the ECRC had information that just a few months before his release date relator had stated that "it doesn't matter what you do with slutty women."
Given the potential gravity of a reoffense, relator's apparent unchanged lack of concern and respect for women, the inherent difficulty of applying an objective screening tool like the MnSOST-R, and the nature of relator's past criminal sexual conduct, we conclude that the ALJ did not err in determining that the ECRC acted within its discretion in concluding that a broader notification was necessary and that, therefore, relator should be classified as a risk level III.
DECISION
Under the Sex Offender Community Notification Act, the End of Confinement Review Committee, when it finds and documents special concerns, has the authority to exercise its discretion and assign the appropriate risk level for an offender notwithstanding the presumptive risk level indicated by the Sex Offender Screening Tool. The evidence considered by the ECRC supports its determination that relator should be assigned a risk level III.
Affirmed.