JUSTICE BYRNE delivered the opinion of the court:

The trial court found respondent, Alan Field, to be a sexually violent person pursuant to the Sexually Violent Persons Commitment Act (the Act) ( 725 ILCS 207/1 et seq. (West 2000)) and ordered him committed to a Department of Human Services (DHS) treatment facility indefinitely. Before the trial, the court granted respondent a hearing under Frye v. United States, 293 F. 1013 (D.C. Cir.1923), to determine the admissibility of certain actuarial instruments the State's experts used to predict the likelihood that respondent would reoffend. The court found the instruments admissible and permitted the experts to testify to their results. On appeal, respondent contends that (1) the trial court abused its discretion in denying his motion to exclude the evidence under Frye and (2) trial counsel was ineffective for failing to present evidence at the hearing to show that the instruments were unreliable and experimental. We affirm, concluding that the trial court committed harmless error in admitting the evidence of the actuarial instruments at trial.

FACTS

Respondent pleaded guilty to aggravated criminal sexual abuse and aggravated criminal sexual assault, and the trial court imposed concurrent five- and eight-year prison terms. On July 22, 1999, three days before respondent's scheduled mandatory supervised release, the State filed a petition to commit respondent pursuant to section 40 of the Act ( 725 ILCS 207/40 (West 2000)). The petition alleged that respondent suffered from the following mental disorders: (1) pedophilia, sexually attracted to females, nonexclusive type and (2) antisocial personality disorder. The State alleged that respondent was dangerous to others because his mental disorders created a substantial probability that he would engage in future acts of sexual violence. The State supplemented the petition with a mental health evaluation prepared by Dr. Marc Levinson, a clinical psychologist.

The trial court appointed the Boone County public defender's office as respondent's counsel and Dr. Donald Pearson as respondent's clinical expert. On January 19, 2001, respondent moved to bar expert testimony regarding the results of certain actuarial instruments utilized to predict the likelihood that respondent would reoffend. Respondent challenged the Hanson-Bussiere meta-analysis, the Rapid Risk Assessment of Sexual Offense Recidivism (RRASOR), the Static-99, and the Minnesota Sex Offender Screening Tool-Revised (MnSOST-R). Respondent argued that the validity and reliability of these tests have not been established and that the tests have not been accepted by the scientific community. Respondent concluded that the tests failed to meet the standards for admissibility set forth in Frye.

The trial court conducted a Frye hearing on January 25, 2001. Only one witness testified: Dr. Levinson, the State's expert who prepared the evaluation on which the petition was based. Dr. Pearson, respondent's expert, and Dr. Heaton, another State expert, were present, but neither testified. We need not discuss the reliability of the MnSOST-R results because they were not introduced at the subsequent hearing on the petition.

Dr. Levinson testified that meta-analysis refers to an actuarial instrument developed by Drs. Hanson and Bussiere. The instrument became available in 1996. In 1998, it was published in a peer review journal, the Journal of Consulting and Clinical Psychology. Levinson opined that the meta-analysis is generally accepted in the field of assessing the risk of sex offender recidivism. Levinson used the instrument as part of respondent's evaluation on May 27, 1999.

The meta-analysis is a table of 22 factors that statistically correlate with sexual offense recidivism. The correlations are based on a survey of 61 studies completed from 1943 to 1995, which involved approximately 29,000 sexual offenders in several countries. According to Levinson, each of the 22 factors had been shown to be statistically significant in at least three of the studies. Levinson acknowledged on cross-examination that all assessment instruments are susceptible to reporting false positives and that the instruments are likely to be revised and improved over time.

Dr. Levinson also used the RRASOR to evaluate respondent. The instrument became available in 1997 and was devised by the same Dr. Hanson who published the meta-analysis with Dr. Bussiere. The RRASOR is a list of 4 of the 22 meta-analysis factors that Dr. Hanson designated as easily identifiable and reliable risk predictors. The RRASOR factors are (1) the number of prior charges or convictions of sexual offenses; (2) the offender's age; (3) whether the sexual offense included male victims; and (4) whether the offender has ever been married. The RRASOR was standardized with a sample of 500 offenders and then cross-validated with another sample to determine whether the results were replicable and reliable. Dr. Levinson believed that the test was considered reliable in the scientific community. He also stated that the RRASOR is widely used by professionals who evaluate recidivism risk in sex offenders.

On cross-examination, respondent's counsel referred Dr. Levinson to a 1997 book entitled "Manual for Sexual Violence Risk 20, Guidelines for Assessing Risk of Sexual Violence." Dr. Levinson acknowledged that the book stated that there was no evidence that the RRASOR has "predictive validity with respect to sexual violence."

Dr. Levinson testified that the Static-99 is another actuarial instrument that he used to evaluate respondent on May 27, 1999. The Static-99 was developed by Drs. Hanson and Thornton, and it became available in 1999. It combines the RRASOR factors with a scale that Dr. Thornton developed in England. The authors cross-validated the Static-99 scores with an independent data sample to show that the predictions were reliable across different samples. Dr. Levinson believed that the Static-99 slightly improved upon the RRASOR as to the correlation between the scores and recidivism. He opined that the Static-99 is generally accepted in the community of professionals who evaluate sex offenders for risk of recidivism.

The trial court found that the instruments were admissible under Frye, concluding that each is generally accepted in the psychological community that deals with sex offenders. The court then heard the petition to declare respondent sexually violent.

Dr. Levinson testified that he reviewed respondent's master file before the evaluation. Respondent told Dr. Levinson that he understood the purpose of the interview and knew that he could decline to participate. Dr. Levinson based his evaluation on the Hare Psychopathy Checklist-Revised. Dr. Levinson learned that respondent had grown up in the Rockford area, attended public schools, and graduated from high school after struggling academically. As an adult, respondent was consistently employed, and his parents were hard-working, responsible members of the community.

Respondent told Dr. Levinson that he committed his first sexual offenses at age 12. Respondent told another psychologist that he victimized approximately 30 children, but he told Dr. Levinson that he "directly offended [only 15 or 20] in a hands-on manner." The victims were girls from 4 to 12 years old. Respondent was sexually aroused from fondling their vaginas. Respondent stated that he would "back off' if the girl resisted overtly.

Respondent was 42 years old when he was arrested in September 1994 for the first sexual offense of which he was convicted: the aggravated criminal sexual abuse of two girls who were seven and eight years old. Respondent placed his hands on their vaginas at least 10 times during several visits to their home. The presentence investigation report indicated that respondent acknowledged that he was sick and could not stop offending.

Respondent was arrested again in October 1995 for the aggravated criminal sexual assault of a girl less than 12 years old. Respondent befriended her family and took the girl on outings over a three-year period. He was convicted of inserting his finger into the girl's anus.

Respondent had had two adult girlfriends and proposed marriage to each, but both declined. Respondent never married or formed a long-term commitment to a sexual partner of his own age. Respondent told Dr. Levinson that his decision not to marry was partially motivated by his fear that he would molest daughters who might be born to the relationship.

Based on the Diagnostic and Statistical Manual of Mental Disorders (DSM IV), Dr. Levinson diagnosed respondent as suffering from pedophilia, sexually attracted to females, nonexclusive type. Dr. Levinson also diagnosed respondent as having antisocial personality disorder, which is characterized by a pattern of disregard for social rules and conventions, and a long-standing history of behavior that could be grounds for arrest. Dr. Levinson cited respondent's history of molesting children since the age of 12, his two arrests for driving under the influence (DUI) committed during his late teens and early twenties, and his juvenile arrests for sniffing glue and for breaking into a factory to steal soda and candy bars. Dr. Levinson did not diagnose respondent as alcohol or substance dependent, but he believed that respondent suffers from a mental disorder that affects his emotional or volitional capacity and predisposes him to future sexual offenses. Respondent told Dr. Levinson that he had tried to stop his physical attraction to young people.

In reaching his conclusions, Dr. Levinson "looked particularly" at the 22 risk factors of the meta-analysis. Respondent exhibited seven factors: (1) admitted sexual offenses before his current convictions; (2) deviant sexual arousal from prepubescent girls; (3) sexual offenses against persons unrelated to him; (4) the absence of marriage to or cohabitation with an age-appropriate sexual partner; (5) first offense at a very young age; (6) failure to complete a sex offender treatment program; and (7) a personality disorder, particularly antisocial personality disorder. Dr. Levinson opined that the meta-analysis indicated that it is "substantially probable" that respondent will engage in future sexual violence because he suffers from a mental disorder that predisposes him. "Substantially probable" means "much more likely than not."

On February 1, 2001, Dr. Heaton, another expert for the State, testified that he reviewed respondent's records, administered psychological tests on August 9, 1999, and interviewed respondent on August 21, 1999. Dr. Heaton concluded that respondent is self-centered, immature, needy, compliant, defensive, and impulsive. Respondent displays average intelligence, low self-esteem, and possibly an alcohol or substance abuse problem. Respondent admitted molesting numerous children, and Dr. Heaton believed that respondent had limited relationships with adult women because he had suffered from erectile dysfunction since he was 25 or 26 years old. Dr. Heaton diagnosed respondent as suffering from pedophilia, sexually attracted to females, nonexclusive type, which meant that respondent would potentially pursue sexual relations with adult women. Dr. Heaton also diagnosed respondent with antisocial personality disorder, based on respondent's glue sniffing, his DUI offenses, and his history of deceitful offenses against children. Dr. Heaton used the Hanson-Bussiere meta-analysis and the RRASOR in reaching his conclusions.

Dr. Heaton stated that respondent exhibited at least six of the Hanson-Bussiere meta-analysis risk factors: (1) prior sexual offenses, (2) prior criminal record of any kind, (3) early onset of sexual offending, (4) extrafamilial victims, (5) existence of any personality disorder, and (6) existence of antisocial personality disorder, in particular. Dr. Heaton found that respondent exhibited two of the four RRASOR risk factors: (1) prior sexual offenses, and (2) extrafamilial victims. In his August 23, 1999, report, Dr. Heaton stated that there was a "substantial probability" that respondent would sexually reoffend. Since submitting the report, Dr. Heaton reviewed respondent's treatment progress notes and spoke with respondent and the treatment providers. Dr. Heaton's opinion had not changed, because the risk factors persisted and nothing had been done that would significantly reduce respondent's risk of reoffending.

On cross-examination, Dr. Heaton acknowledged an article written by Drs. Hanson and Bussiere in which they wrote that the penile plethysmograph test, which measures a person's sexual arousal to various stimuli, is the strongest predictor of recidivism among the meta-analysis factors. Dr. Heaton did not subject respondent to the test because respondent's history of erectile dysfunction would obscure the results. Dr. Heaton believed that respondent can control his conduct at a specific moment, but that his fantasies and urges affect his ability to choose not to molest prepubescent girls.

Dr. Pearson, respondent's expert, testified that he evaluated respondent in March 2000. Respondent submitted to a penile plethysmograph test, during which he responded to mature women but not young children. Dr. Pearson ordered the test, but he did not know who administered it. The testers reported that they thought that respondent faked his response and that the results were ambiguous because respondent had an erection that did not dissipate.

Dr. Pearson administered several psychological tests, including the Bumby, the Bays, and the Freeman-Longo tests. The Bays test and respondent's record indicated that he posed a moderate risk of reoffending. Respondent fit the criteria for pedophilia, but Dr. Pearson could not be certain until an accurate plethysmograph could be completed. Dr. Pearson did not use the RRASOR because its validity and reliability were questioned in the profession. Dr. Pearson believed that respondent needed help and supervision but could control his urges and behavior. On cross-examination, the doctor characterized the risk of reoffending as "moderate to minimal."

The trial court found that the State had proved beyond a reasonable doubt that respondent was a sexually violent person. After the court denied respondent's motion for a new trial and ordered him committed to the DHS, respondent timely appealed.

ANALYSIS

Respondent initially contends that the trial court abused its discretion in permitting the State's experts to testify about certain actuarial instruments they used to predict the likelihood that respondent would reoffend. The State responds that the instruments are not subject to a Frye analysis, and that, even if Frye applies, the State established the instruments' admissibility.

In Illinois, the admission of expert testimony is governed by the Frye standard: whether the methodology or scientific principle upon which the opinion is based is sufficiently established to have gained general acceptance in the particular field in which it belongs. Donaldson v. Central Illinois Public Service Co., 199 Ill. 2d 63, 77 (2002). General acceptance of a methodology does not require universal acceptance, and the Frye test applies only if the scientific principle, technique, or test offered by the expert to support his or her conclusion is "new" or "novel." Donaldson, 199 Ill. 2d at 78-79. General acceptance and reliability are not two separate questions. The determination of the reliability of an expert's methodology is naturally subsumed by the inquiry into its general acceptance in the scientific community. Donaldson, 199 Ill. 2d at 81.

This court has held that the actuarial instruments at issue in this case are scientific methodologies that are admissible only if they pass the Frye test. In re Commitment of Lourash, 347 Ill. App. 3d 680, 686 (2004) (Hanson-Bussiere meta-analysis and MnSOST-R); In re Detention of Traynoff 338 Ill. App. 3d 949, 964 (2003) (RRASOR, Static-99, and MnSOST-R); People v. Taylor, 335 Ill. App. 3d 965, 977 (2002) (RRASOR and Static-99). In Taylor, we concluded that the actuarial instruments, which supplemented and went beyond the experts' "clinical judgment, training, and expertise," qualified as scientific methodologies that are admissible only if generally accepted in the relevant scientific community. Taylor, 335 Ill. App. 3d at 977. We held that the State had not met its burden ( Taylor, 335 Ill. App. 3d at 978), but we acknowledged that in a future case, the State might be able to prove that the actuarial instruments meet the Frye criteria. Taylor, 335 Ill. App. 3d at 980.

We decline the State's invitation to reconsider Taylor and its progeny and to hold that the actuarial instruments are not subject to the Frye test. The Third District and at least one panel of the Fourth District have adopted our reasoning (see In re Detention of Hargett, 338 Ill. App. 3d 669, 675 (2003) (Third District); In re Detention of Bolton, 343 Ill. App. 3d 1223, 1229 (2003) (Fourth District)). However, other panels of the Fourth District have departed from Taylor. See In re Commitment of Stevens, 345 Ill. App. 3d 1050 (2004); In re Detention of Erbe, 344 Ill. App. 3d 350 (2003). In the interest of stare decisis, we adhere to our earlier holdings, noting that the supreme court is poised to resolve the split in authority (see In re Commitment of Simmons, No. 5-02-0579 (2003) (unpublished order under Supreme Court Rule 23), appeal allowed, 206 Ill. 2d 622 (2003)).

We next address respondent's contention that the State failed to meet its burden of proving that the RRASOR and the Static-99 instruments had achieved general acceptance in the scientific community. We note that, on appeal, respondent does not challenge the admission of the Hanson-Bussiere meta-analysis evidence. Like in this case, the trial court in Taylor granted the respondent a Frye hearing during which the State's experts opined that the RRASOR and the Static-99 were generally accepted in the professional community. The evidence in Taylor indicated that the RRASOR and the Static-99 had been the subject of only one article in a peer review journal, and one expert asserted that the Static-99 had been cross-validated by other researchers. Because the State failed to offer "evidence of any significant peer review of the validity and reliability of [the] instruments," this court concluded that the RRASOR, the Static-99, the MnSOST, and the MnSOST-R were still in experimental stages and that the validity of the instruments had not been established. Taylor, 335 Ill. App. 3d at 978.

At the Frye hearing in this case, Dr. Levinson broadly opined that the RRASOR and the Static-99 had gained general acceptance among psychologists who evaluate the risk of sex offenders' recidivism. He testified that the authors of the RRASOR standardized the instrument with a sample of 500 offenders and then cross-validated it with another sample. Dr. Levinson also stated that the authors of the Static-99 cross-validated scores from that instrument with an independent data sample to show that the predictions were reliable across different samples. However, the State offered no evidence of peer review of either the RRASOR or the Static-99. Moreover, during the cross-examination of Dr. Levinson, respondent's counsel introduced a 1997 text stating that there was no evidence of the RRASOR's predicative validity. In agreement with Taylor, we conclude that the State failed to establish that the RRASOR and Static-99 had gained general acceptance in the psychological community that evaluates the risk of sex offenders' recidivism.

However, the erroneous admission of the actuarial instruments will not require a reversal if the error was harmless beyond a reasonable doubt. Lourash, 347 Ill. App. 3d at 687. In Lourash, the State's experts opined at trial that the respondent, a sex offender, was substantially likely to reoffend. The experts cited the Hanson-Bussiere meta-analysis, the Static-99, and the MnSOST-R, and we held that the trial court erred in denying the respondent a Frye hearing to evaluate the instruments. However, we further concluded that the error was harmless because, even if the respondent had prevailed at a Frye hearing, the result of the proceedings would have been the same because the instruments were not "crucial" to the expert's conclusion, and the evidence based on undisputed scientific methodologies amply supported the conclusion. Lourash, 347 Ill. App. 3d at 687.

Dr. Levinson testified that he diagnosed respondent with antisocial personality disorder and pedophilia, sexually attracted to females, nonexclusive type. Dr. Levinson also relied upon the Hanson-Bussiere meta-analysis, which is an actuarial instrument that respondent does not challenge. Dr. Levinson opined that respondent exhibited 7 of the 22 meta-analysis factors: (1) admission of sexual offenses before his current convictions; (2) deviant sexual arousal from prepubescent girls; (3) sexual offenses against persons unrelated to him; (4) the absence of marriage to or cohabitation with an age-appropriate sexual partner; (5) first offense at a very young age; (6) failure to complete a sex offender treatment program; and (7) a personality disorder, particularly antisocial personality disorder. Respondent admitted a 30-year pattern of molestation during which he fondled the vaginas of 15 to 20 girls who were 4 to 12 years old at the time. Respondent's presentence investigation report indicated that he told a psychologist that he knew he was sick and could not help offending. On redirect examination, Dr. Levinson stated that, in addition to the meta-analysis, he relied upon all of his experience in interviewing respondent and evaluating his clinical history.

Dr. Heaton also diagnosed respondent with pedophilia and antisocial personality disorder, and the doctor identified six meta-analysis factors indicating that respondent was very likely to reoffend if released. Moreover, respondent's own expert admitted that respondent fit the criteria for pedophilia and that he presented a minimal to moderate risk of reoffending. Therefore, we hold that, although the trial court erroneously admitted evidence regarding the RRASOR and the Static-99, the error was harmless beyond a reasonable doubt because the evidence unrelated to those instruments supported the opinions of the State's experts. See Lourash, 347 Ill. App. 3d at 687.

Finally, respondent argues that his trial counsel was ineffective for failing to present evidence at the Frye hearing to show that the RRASOR and the Static-99 were unreliable and experimental. To prevail on a claim of ineffective assistance of counsel, a defendant must show both that (1) counsel's performance was deficient, and (2) the deficient performance prejudiced the defendant such that he was deprived of a fair trial. In re Detention of Hughes, 346 Ill. App. 3d 637, 646 (2004). The failure to satisfy either prong of the test precludes a finding of ineffective assistance of counsel. Hughes, 346 Ill. App. 3d at 646. We agree with respondent that the instruments were inadmissible. However, we need not address his claim of ineffective assistance of counsel because we also conclude, for the reasons already stated, that the admission of the instruments was harmless, and therefore respondent was not prejudiced by his attorney's failure to better challenge their admissibility.

For the preceding reasons, the judgment of the circuit court of Boone County is affirmed.

Affirmed.

HUTCHINSON and GILLERAN JOHNSON, JJ., concur.