In Re the Risk Level Determination of S.S.

OPINION ROSS, Judge. This appeal arises from an administrative law judge’s order affirming a decision by the department of corrections’ end-of-confinement review committee to designate S.S., a female convicted of criminal sexual conduct, a risk-level-III predatory offender. S.S. argues that the committee failed to follow the requirements of Minnesota Statutes section 244.052 in making its risk-level determination. Because the procedure the committee followed to assess S.S.did not comply with the statutory requirement to apply a weighted risk-assessment tool, we reverse and remand for a redetermination of S.S.’s risk level. FACTS 5.5. is an adult female who was convicted of criminal sexual conduct for her sexual behavior with an adolescent male. In this challenge to an end-of-confinement review committee’s (ECRC) risk-level determination, S.S. does not dispute the administrative law judge’s (ALJ) factual findings but instead contests the procedure through which the committee reached its determination. 5.5.’s underlying misconduct occurred when she worked as a cook and youth supervisor at a residential juvenile treatment facility in 1998. Five teenage boys at the facility alleged that, during her employment, S.S. engaged in sexual intercourse and oral sex with them. They stated that she provided them with marijuana and cigarettes and, when one of them tried to end their sexual relationship, she became angry and threatened to deny him home visits. She also would expose her breasts to some of the boys. In February 1999, the state charged S.S.with seven counts of criminal sexual conduct, two counts of indecent exposure, seven counts of contributing to the delinquency of a minor, and seven counts of furnishing tobacco to a minor. After hearing testimony from several staff members and residents who observed or experienced inappropriate behavior by S.S., a jury convicted her of three counts of criminal sexual conduct for her acts against one of the boys, five counts of contributing to the delinquency of a minor, and seven *123 counts of providing tobacco to a minor. In August 1999, the district court sentenced S.S.to 86 months’ incarceration. S.S.continued to misbehave after she arrived in prison. She was disciplined several times for rule violations, some major. 5.5. entered the prison’s sex-offender treatment program in February 2001 but was terminated from the program six months later for sexual misconduct. S.S. had flashed her breasts at other inmates in the food-service area and offered to perform sexual favors in exchange for money. She maintains that she was joking. She reentered the treatment program in January 2003 and completed it in September 2003. S.S.also engaged in misconduct after her release. Several months before her scheduled-release date, a correctional department psychologist reviewed her case file and recommended a risk-level-II assignment. The ECRC adopted this recommendation, and S.S. did not challenge it. She was released from prison in June 2004, but the district court terminated her conditional release in December 2004 after she violated several conditions of the release. She had unauthorized contact with minors, failed to maintain employment, and was terminated from her outpatient sex-offender treatment program. Most notably, S.S.’s friend and the friend’s two children, a thirteen-year-old boy and a twelve-year-old girl, had moved into S.S.’s one-bedroom apartment. S.S. had not obtained permission from her supervised release agent, and she failed to mention her living arrangement to the agent. Based on the violations, the court ordered. S.S. to return to prison for 150 days. Before S.S.’s new scheduled-release date, the department’s psychologist, who is also a member of the ECRC, reassessed 5.5.’s- risk-level determination. He reviewed her file and evaluated her according to the statutory factors set forth in Minn.Stat. § 244.052, subd. 3(g) (2004). He also noted that S.S.’s high-risk behavior while on release was a special concern. The psychologist recommended a risk-level-III assignment, and, in February 2005, the ECRC agreed. It is the process of reaching this risk-level assignment that is the subject of this appeal. S.S.sought administrative review of the committee’s assignment decision, challenging the procedure through which the committee reached its decision and its failure to apply a weighted risk-assessment tool in addition to the statutory factors. The department has a risk-assessment tool — an evaluative psychological and behavioral test — that is applied to male evaluees, but it has none specifically for women. The ALJ affirmed the ECRC’s determination. The ALJ held that the department developed the statutorily required risk-assessment tool, but that the controlling statutes do not require application' of a weighted tool when assessing a female offender’s risk level. She reasoned that it is not currently feasible to develop a tool for females comparable to the Minnesota Sex Offender Screening Test-Revised (MnSOST-R), which the ECRC uses when determining a male offender’s risk level. The ALJ then found that the ECRC properly reviewed S.S. in light of other statutory factors and a department of corrections’ policy. S.S. appeals by writ of certiorari. She asserts that the ECRC did not follow the applicable statutory procedure when assessing her risk level. ISSUE Did a department of corrections’ end-of-confinement review committee comply with Minn.Stat. § 244.052 (2004) when it failed to apply a weighted risk-assessment tool to assess the risk level of a female offender? *124 ANALYSIS S.S. challenges the ALJ’s affirmance of the risk-level-III assignment, which the ALJ based on the ECRC’s evaluative process. An offender may seek administrative review of an ECRC’s risk-level determination. Minn.Stat. § 244.052, subd. 6(a) (2004). The offender bears the burden of proving, by a preponderance of the evidence, that the committee’s determination was erroneous. Id., subd. 6(b) (2004). On certiorari appeal, we will affirm an ALJ’s decision unless we find that the offender’s substantial rights have been prejudiced because the agency decision was made upon unlawful procedure, affected by an error of law, or otherwise unsupported by substantial evidence. Id. § 14.69 (2004) (listing circumstances in which court of appeals may reverse or modify agency decision). S.S. argues that the ECRC made its risk-level determination without following the statutory procedures set forth in section 244.052. We agree. When an agency’s authority is questioned, we independently review the enabling statute. In re Risk Level Determination of R.B.P., 640 N.W.2d 351, 353 (Minn.App.2002), review denied (Minn. May 14, 2002). Section 244.052 of the Minnesota Statutes establishes a five-person end-of-confinement review committee within the department of corrections to assess, on a case-by-case basis, the public risk posed by predatory offenders who will soon be released from confinement. Minn.Stat. § 244.052, subd. 3(a)-(b) (2004); see also id., subd. 1(5) (2004) (defining predatory offender to ⅛ elude persons required to register as a predatory offender under Minn.Stat. § 243.166 (2004)). The committee must assign one of three risk levels to an offender according to the offender’s risk-assessment score. Id., subd. 3(e) (2004). A risk-level-I designation indicates that an offender has a low risk of reoffending, a risk-level-II designation indicates a moderate risk of reoffending, and a risk-level-III designation indicates a high risk of reof-fending. Id. The committee must make its determination by using a nonexclusive list of statutorily enumerated risk factors and a risk-assessment scale that assigns weight to each of the factors. Id., subd. 3(d)(i), (g) (2004). S.S. contends that the ECRC failed to follow the express requirement of the statute by not employing a risk-assessment scale. We look to the statute “to ascertain and effectuate the intention of the legislature.” Minn.Stat. § 645.16 (2004). If a statute’s language is unambiguous, we apply its plain meaning. Id.; In re Risk Level Determination of C.M., 578 N.W.2d 391, 395 (Minn.App.1998). On that review we agree that the ECRC’s procedure for assessing S.S.’s risk level did not comply with the plain meaning of the applicable statutes. Section 244.052 was enacted in 1996. 1996 Minn. Laws Ch. 408, art. 5, § 1, at 660-64. Subdivision 2 of this section required the commissioner of corrections to develop, by January 1,1997, “a risk assessment scale which assigns weights to the various risk factors listed in subdivision 3, paragraph (g), and specifies the risk level to which offenders with various risk assessment scores shall be assigned.” Minn. Stat. § 244.052, subd. 2 (2004). The subdivision required the commissioner to consult with various professionals to develop the scale. Id. Subdivision 3(d)(i) states, “The committee shall use the risk factors described in paragraph (g) and the risk assessment scale developed under subdivision 2 to determine the offender’s risk assessment score and risk level.” The plain language of the statute requires the creation and application of the weighted risk-assessment scale and application of the statutory factors. *125 To implement section 244.052, the department adopted policy 205.220 and uses the MnSOST-R as its- weighted risk-assessment tool. The MnSOST-R is an actuarial risk-assessment instrument that compares characteristics of an individual offender to those of a group of documented offenders to calculate the subject offender’s risk to reoffend. The test replaced an earlier version, the MnSOST, which also weighted various factors but was not statistically significant. Under the policy in effect when the ECRC designated S.S. to be a risk-level-III offender, an offender evaluated with the MnSOST-R would be assigned a presumptive risk level based on the results, but the committee reserved the discretion to adjust the level upward or downward based on an offender’s individual circumstances that are not adequately measured by the MnSOST-R. The department supplemented the policy with a list of fourteen special concerns that may result in an upward adjustment to the presumptive risk level. At issue in this appeal is the ECRC’s failure to apply the MnSOST-R or an alternative weighted risk-assessment tool to female offenders, like S.S., who are subject to the committee’s evaluation for risk-level assignment. The department of corrections’ psychologist testified that, when evaluating male sex offenders, he follows the procedure stated in the policy and applies the MnSOST-R. The committee, however, did not employ the MnSOST-R in determining S.S.’s risk level because the test is statistically valid only as applied to male sex offenders, and the test has never been used for female offenders. The psychologist stated that, to his knowledge, a statistically valid test does not exist to assess females for their risk of reoffense. He noted that the relatively low number of female sex offenders likely contributes to the difficulty in developing a test similar to the MnSOST-R to be applied to females. The lack of female sex offenders in the corrections population is supported by his 'testimony that, in his five years with the ECRC, he has evaluated over 2,000 males but only 30 to 35 females. In the absence of a version of the MnSOST-R for female offenders, the psychologist testified that the department has an unwritten informal policy of “applying] the policy for males, as much as possible, as fairly and as accurately as possible, to females.” In this case, the committee considered its list of special concerns (that is, those fourteen factors it considers when deciding whether to apply an upward adjustment to a male’s presumptive risk level) and found that one applied because S.S. had engaged in high-risk behavior during release. The committee also considered the statutory factors, which address the potential danger of reoffense, behavior on release, prior criminal history, offender characteristics, support in the offender’s community, intent to reoffend, and physical condition. Minn.Stat. § 244.052, subd. 3(g). When asked if the factors were weighted in any way, the psychologist responded, “I don’t know if I can answer that.” The ECRC erred in its procedure for determining S.S.’s risk level because the committee did not use any weighted risk-assessment tool as required by statute. The department argues that the failure to apply a weighted tool as directed by subdivision 2 of section 244.052 is inconsequential because the committee addressed the factors listed in subdivision 3(g), along with its. own list of special concerns. But subdivisions 2 and 3, which were enacted at the same time, outline complementary, not alternative, obligations. The use of the mandatory “shall” coupled with the conjunctive “and” in subdivision 3(d)(i) indicates that the legislature intended the *126 committee to apply both the factors and a weighted risk-assessment tool. We hold that applying the factors alone is insufficient under the plain meaning of the statute. The ALJ also relied on the discretion the ECRC is given in section 244.052 because, in addition to the risk-assessment scale results, the committee may consider the nonexclusive list of factors listed in subdivision 3(g). We acknowledge that the statutory provisions establishing the ECRC and the procedures it should follow vest the committee with some discretion, but this discretion does not relieve it of the obligation to comply with the provisions of section 244.052. In R.B.P., we concluded that the legislature intended to vest some discretion in the ECRC and held that, while the MnSOST-R score provides a presumptive score, the ECRC can rebut this presumption by accounting for individual circumstances not otherwise reflected in the score. R.B.P., 640 N.W.2d at 355-56. But although we have recognized the ECRC’s authority to undertake a subjective analysis, it is only to augment, not replace, its application of an objective analysis. See id. at 355 (recognizing presence of objective component while noting legislative intent “to vest some discretion in the ECRC” (emphasis added)). Section 244.052 requires application of a weighted tool in addition to the factors. We also note that the department’s policy states that the ECRC “will assign risk levels based on the totality of circumstances and articulable facts, including but not limited to the score indicated on the MnSOST-R.” Its policy to include the MnSOST-R score when imposing a reof-fense risk level underscores our conclusion that the test must, by statute, play a central role in establishing a risk level. The psychologist testified that MnSOST-R does not capture the nature and quality of the events it measures, and the special concerns and statutory factors therefore allow the committee to consider the presence of aggravating or mitigating factors. But the procedure the committee applied in determining S.S.’s risk level was wholly subjective and did not follow the statutory requirement to apply a weighted risk-assessment tool. The testimony in this case indicates that the department’s present ability to develop a statistically significant scale for females comparable to the MnSOST-R is dubious. Section 244.052, subdivision 2, however, does not require the department to develop a scale based on statistically determined criteria. The statute requires input from, professionals who serve in specific fields to develop the weighted assessment, but it does not include statisticians or social scientists among them. Minn. Stat. § 244.052, subd. 2 (requiring department to consult with county attorneys, treatment professionals, law-enforcement officials, and probation officers). We also note that the weighted risk-assessment tool the committee used before the MnSOST-R was not statistically valid. By all indications, the department took reasonable steps not only to comply with the statute but to improve its former assessment tool for greater, statistically based accuracy. This improved procedure may exceed the. requirements of section 244.052 in the vast majority of cases, which involve men, but the department failed to implement a statutorily sound procedure for the comparably few female offenders it must also assess. The department has since incorporated different procedures based on gender, but only after S.S.’s assessment. Whether the current policy satisfies section 244.052 is therefore not at issue in this appeal. Because we hold that the ECRC erred in its procedure for determining S.S.’s risk *127 level, we do not address S.S.’s argument that the ALJ’s interpretation of section 244.052 violates principles of equal protection. See State v. Hoyt, 304 N.W.2d 884, 888 (Minn.1981) (holding that courts do not decide constitutional questions unless issue is dispositive). DECISION Because Minnesota Statutes section 244.052 required the department of corrections to develop and apply a weighted risk-assessment tool for assessing offenders’ risk levels and the end-of-confinement review committee did not apply any weighted risk-assessment tool when assessing S.S.’s risk level, we reverse and remand for a redetermination of S.S.’s risk level. Reversed and remanded.