Supreme Court, New York County, July 31, 2012

APPEARANCES OF COUNSEL

Cyrus R. Vance, Jr., District Attorney (Jennifer Sculco of counsel), for plaintiff. Center for Appellate Litigation (Carl S. Kaplan of counsel), for defendant.

Daniel P. Conviser, J.

This decision addresses an issue of first impression under New York's Sex Offender Registration Act (SORA): Does the "Position Statement" dated June 1, 2012 by the New York State Board of Examiners of Sex Offenders (the Board) regarding the scoring of child pornography cases under the New York sex offender risk assessment instrument (the RAI) change the manner in which such cases must be scored under the instrument? This court concludes that it does. While prior to the Position Statement, courts were clearly required to score defendants in typical child pornography cases with points because victims were strangers and because there were multiple victims, the Position Statement has now apparently eliminated the direction for such points. The court also concludes that while the Position Statement may result in an improvement over the manner in which child pornography defendants were previously scored, the new system leaves courts with inadequate tools with which to make informed RAI departure decisions.^[FN1]

This court conducted a SORA risk assessment hearing in this case on July 7, 2012. Santiago Marrero was convicted by plea of guilty in July of 2009 of one count of possession of [*2]child pornography under federal law for which he received a sentence of two years' imprisonment and five years of supervised release. Mr. Marrero was apprehended in a "sting" operation in which he responded to a solicitation to buy child pornographic videos by postal inspectors. He was subsequently arrested and a search of his home computer uncovered 300 to 600 images primarily of prepubescent girls clothed, naked or being sexually abused. He was released from prison to federal supervised release in December of 2011.{**37 Misc 3d at 431}

Mr. Marrero is 33 years old and has had no other contacts with the criminal justice system. He has suffered from severe problems with depression and social anxiety arising from prior physical abuse by his father. Psychiatric evaluations indicated that he posed little risk of committing a contact sex offense. His prospects for committing another child pornography crime, however, were less certain. Mr. Marrero's counsel argued that the defendant did not view the child pornographic images on his computer for sexual pleasure but rather because he identified with the pain and abuse the children in the videos had suffered. In the court's view, however, the evidence indicated that the defendant also may well have been motivated at least in part by sexual desire. The evidence indicated that although Mr. Marrero could not be diagnosed with pedophilia, such a diagnosis could also not be ruled out. Mr. Marrero has expressed remorse for his crime. He currently lives and cares for his ailing mother in public housing but would apparently have to move out of his mother's apartment if he were designated as anything other than a level one offender. He has been complying with the requirements of his supervision.

The court received an RAI scored by the People and the Board. The People scored the defendant with 30 points under RAI factor 3 for having three or more victims, 30 points under factor 5 because the victims of the offense were 10 years old or less and 20 points under factor 7 because the defendant and the victims were strangers. The People's total RAI score was thus 80 points, 5 points above the 75-point minimum threshold for a level two offender at a moderate risk to re-offend.

The Board presented the court with a radically different proposed score. The Board scored the defendant with 30 points because the victims were 10 years old or less. However, the Board did not include any points for stranger or multiple victims. The Board's presumptive score was thus 30 points. The Board, however, recommended an upward departure to level two. Following the hearing, the court scored the defendant with 20 points (rather than 30 points) under factor 5 because there was not sufficient evidence that the prepubescent girls in the images on the defendant's computer were 10 years old or less rather than 11 years old or more. In accordance with the Board's recommendation, the court did not score any points because the victims were strangers or because there were more than three victims, although the court did find that these circumstances literally existed. The court also denied the People's motion to{**37 Misc 3d at 432} depart to a level two and thus scored Mr. Marrero as a level one offender at a low risk to re-offend.

The question of whether child pornography defendants should be scored with RAI points because the "relationship" between the defendant and his victims is one of a "stranger" was addressed by the Court of Appeals in People v Johnson (11 NY3d 416 [2008]). The defendant in Johnson argued that such points were not justified because the "stranger" risk factor was intended to apply only to contact offenses. In analyzing the issue, the Court first looked at the Sex Offender Registration Act: Risk Assessment Guidelines and Commentary (2006) to the RAI written by the Board. The Commentary asserts that one of the primary [*3]reasons for adding points for stranger victims is that "[t]he need for community notification . . . is generally greater when the offender strikes at persons who do not know him well" (at 12 n 8).

Commenting on that proviso and the overall issue, the Court noted that

"the truth of this generalization is not at all obvious when the offense in question is the possession of child pornography. Most people who commit such crimes never have any contact with the children whose images they look at. The unusual case where the offender and the children are acquainted would seem to present a greater threat to the community, not a lesser one. Bad as this defendant's conduct was, it would surely be worse—and defendant would seem a significantly more dangerous man—if he had been looking at pictures of his friends' or neighbors' children. Yet, under factor 7, previous acquaintance with the children would (unless one of the other facts listed in factor 7 were present) decrease defendant's risk score, not increase it. It does not seem that factor 7 was written with possessors of child pornography in mind." (11 NY3d at 419-420.)

The Court went on to note that the assessment of points for stranger victims in child pornography cases produces "a seemingly anomalous result, one the authors of the Guidelines may not have intended or foreseen." (11 NY3d at 421.)

In McFarland I, this court explained why the "stranger" factor was not written with a view towards child pornography cases. The RAI was written in January of 1996 and slightly{**37 Misc 3d at 433} revised in a manner not relevant here in 1997. The current version of the Commentary was published in 2006 but the 2006 Commentary modified the original instrument only to include updated statutory language and clarification. The 2006 version of the Commentary lists 36 articles upon which the RAI is based. The most recent of these articles was published in 1995. The most significant research and learning in the field of sex offender risk assessment, however, has occurred during the ensuing 17 years. (See McFarland I, 2010 NY Slip Op 51705[U], *8, 18-19.) New York's child pornography statute was enacted shortly after the RAI was promulgated.^[FN2] The simple possession of child pornography was not a crime under New York law when the "stranger" factor was written. The instrument thus was obviously not written, as the Court of Appeals surmised, with "possessors of child pornography in mind."

Despite its concerns, the Court of Appeals in Johnson held that points for stranger victims had to be assessed under the RAI in the case before it because to do otherwise would be to "distort the text of [RAI] factor 7 to avoid an unjust result in cases like this." (11 NY3d at 420.) The remedy, the Court held, in cases where the RAI 's point scoring system led to an inappropriate result was for a court to depart downward to an appropriate risk level.

Board Child Pornography Position Statement

The Board's June 1, 2012 Position Statement on the scoring of child pornography cases reads in relevant part as follows:

"Case law (People v Johnson, 11 NY3d 416, 420-421 (2008) & People v Poole, [*4]90 AD3d 1550 (2011)) has held that child pornography offenders properly have points assessed on Factors 3 (Number of Victims) and 7 (Stranger Relationship). However, as the Court in Johnson notes, scoring all child pornography cases for stranger relationship (and similarly in Poole scoring for three or more victims) produces an unintended, anomalous result as the majority of offenders convicted of child pornography offenses will be scored the same when there are clearly vast differences amongst these types of offenders. To ad{**37 Misc 3d at 434}dress the Court's concern and to more accurately reflect the risk of a repeat offense and threat posed to public safety, the Board:

"Will continue to score either 20 or 30 points for the youngest age depicted in the images under the 'Current Offense' category, and will depart from the presumptive level when appropriate based upon factors including but not limited to."

The Position Statement goes on to list nine factors, then says it will recommend an "automatic override" to level three "where clinical documentation exists detailing a mental abnormality that decreases the ability to control impulsive sexual behavior, such as Pedophilia or Hebephlia." Finally, the Position Statement expresses the Board's concerns about child pornography crimes, says that in the majority of cases such offenders have a sexually deviant interest in children which "poses a significant risk to public safety" but says that the Board also recognizes that each offender poses "unique" risks.^[FN3]

The Meaning of the Position Statement

Initially, there are two factual statements in the Position Statement which, in the court's view, are misleading.^[FN4] The first is the notion that either the Johnson or Poole courts expressed any concerns about scoring child pornography offenders for the number of victims. Neither of these courts expressed any concerns in that regard. The issue was not addressed in Johnson. The Poole Court explicitly endorsed such scoring. (90 AD3d at 1550.) The second misleading assertion is that the concern expressed by the Johnson Court about scoring for stranger victims was that the "majority of offenders convicted of child pornography offenses will be scored the same when there are clearly vast differences amongst these types of offenders." The concern the Court of Appeals expressed in Johnson was not inappropriate consistency. It was that such scoring made no sense.

The more significant question is what the Statement is intended to say. After outlining concerns regarding stranger and{**37 Misc 3d at 435} victim number scoring, the Statement asserts that it will "address" these concerns and provide more accurate assessments by continuing to score points [*5]for the age of victims and then depart from presumptive risk levels based on a variety of newly articulated factors. What is conspicuously absent is any mention of what the Board will do henceforth in scoring for stranger or multiple victims. The implication seems to be that child pornography offenders should no longer be scored for such factors. But that basic point is cryptically omitted. Any ambiguity in that regard was resolved in this case, however, by the Board's Case Summary and proposed RAI. Those Board documents scored the defendant only with 30 points for victim age. What was implied in the Position Statement, therefore, was clarified by the Board's proposed RAI score in this case, although the Board provided no reason in this case for not scoring for stranger or multiple victims.

The final puzzling aspect of this pronouncement is why it was written as a "Position Statement." The Board promulgates the Sex Offender Registration Act: Risk Assessment Guidelines and Commentary. This document has multiple sections. It consists of a Guidelines summary, a Commentary, a detailed explanation of the way in which the specific Guidelines should be applied, a "Note on Overrides," a bibliography and a narrative appendix (the sum of these narrative interpretive aids is hereinafter referred to as the "Commentary"). For 16 years, courts have consistently cited the Commentary as controlling authority both in interpreting the RAI and in establishing the standards courts must follow in departing from the instrument's scoring parameters.

The Commentary is written by the Board. The Board may revise it at any time. Multiple sections of the Commentary contain risk assessment policy discussions of precisely the kind which comprise the Position Statement. This is the first Position Statement the Board has ever issued. It reflects the first substantive change in the Board's written policies since a minor amendment made to the RAI in 1997. Why it was promulgated as a Position Statement (which is physically not part of the document entitled "Sex Offender Registration Act: Risk Assessment Guidelines and Commentary") rather than an amendment to the existing Commentary, however, is unclear. What is clear is that the Board no longer believes that child pornography offenders should be scored with points for stranger or multiple victims. Given that, should our courts continue to score such points? This court believes the answer is no.{**37 Misc 3d at 436}

Rationale for "Stranger" and "Multiple Victim" Scores in Child Pornography Cases

The reason courts have assigned points in child pornography cases for stranger or multiple victims is because the RAI does so. A brief examination of these two factors, however, indicates that there is no justification whatsoever for assessing points for "stranger" victims in child pornography cases. Points for multiple child pornography victims, on the other hand, have no known relationship to risk, but are relevant to harm.

Stranger Victims

As this court outlined in McFarland I, one of the many significant problems with the RAI is that it is not a risk assessment instrument based on actuarial data which indicate that its factors as they are scored are correlated to re-offense risk. (See McFarland I, 2010 NY Slip Op 51705[U], *14, 17-38.) One factor in the RAI which is strongly related to risk, however, is that of stranger victims. Ironically, this appears to be inadvertent. As outlined, supra, the RAI Commentary indicates that the instrument assigns extra points for stranger victims not because such offenders are thought to pose a heightened risk to re-offend but because such offenders present an increased need for community notification. [*6]

Actuarial data which indicate that offenders against stranger victims have a heightened risk to re-offend, however, are derived from studies of contact offenders. This court is not aware of any data which indicate that child pornography offenders against stranger victims face a heightened risk to re-offend when compared to offenders who possess child pornography depicting a family member, friend or acquaintance. With respect to offenders who possess but have not produced child pornography, such cases are likely rare in any event. A 2011 meta-analysis of sex offender recidivism studies cited in the Position Statement notes that the well-established "stranger" risk factor applicable to contact offenses may have no relevance in child pornography cases.^[FN5] Indeed, as the Court of Appeals perceptively observed in Johnson, offenders with sexual images of children the offender knows may pose a greater risk to re-offend.

The assessment of such points also, obviously, makes no sense at all with respect to the need for community notification under{**37 Misc 3d at 437} SORA. As noted, supra, the RAI adds points for stranger offenders because of the reasonable view that such offenders present a more urgent need for community notification. With respect to child pornography crimes, however, an offender who possesses pornographic images of his children, neighbors or relatives may well be more threatening and deserving of community notification than an offender who downloads anonymous images from the Internet. In sum, it is clear that there is no policy justification for assigning stranger victim points in child pornography cases.

Multiple Victims

The policy rationale for scoring child pornography offenders for multiple victims is less clear. The expert evidence in the McFarland I case indicated that the number of victims of an instant contact offense was not predictive of an offender's re-offense risk unless there were a very large number of victims (more than 10). (McFarland I, 2010 NY Slip Op 51705[U], *3.) The court is unaware of any data which would suggest that child pornography offenders have varying re-offense risk characteristics, however, depending upon, for example, whether they might possess 2, 3 or 10 images. This factor, then, appears to not have any relationship to re-offense risk.

It does, however, obviously have a relationship to the harm caused by these offenses. The more images an offender possesses, the more the offender has contributed to creating a market for the sexual abuse of children. The RAI, however, does not and was not designed to effectively measure the relative degrees of harm caused by child pornography offenders. It provides 20 points for two victims and 30 points for three or more victims. In the court's experience, however, child pornography offenders are not typically prosecuted for having one or two child pornography images. The usual case, rather, is one where there are large numbers of such images. Thus, in the majority of cases, offenders have been scored with 30 points. The score has functioned less as a measure of how harmful a child pornography offender's crimes are when compared to like offenders and more as an assessment that the demand created by such crimes generally promotes the sexual abuse of multiple children. Since New York's child pornography statute was enacted after the RAI was created, however, none of these impacts could possibly [*7]have been intended by the instrument's drafters.

Acting Supreme Court Justice Mark Dwyer summarized a number of these points in what he described as dicta in a recent child pornography SORA scoring decision:{**37 Misc 3d at 438}

"The language of the sex offender classification rules assigns points to possessors of child pornography for the 'number' of victims, and the 'stranger' classification of victims, in a way that was intended by the authors of the guidelines to apply to physical contact, and not to defendants who possessed and shared child pornography . . . As a result, many possessors of child pornography who are not serious threats to the community will presumptively be classified as level two offenders. Since this court does not think that result would be consistent with the intent of the authors of the SORA guidelines it anticipates that many SORA applications made as to such defendants should result in downward departures to level one." (People v Yen, 33 Misc 3d 1234[A], 2011 NY Slip Op 52240[U], *4 [2011].)

Court Reliance on RAI Commentary

For 16 years, New York courts have consistently relied upon the Commentary to interpret the RAI, even, in this court's view, when the Commentary has not provided any evidence that an RAI factor was related to risk or harm. RAI factor 15, for example, assigns 10 points if an offender's "living or employment situation is inappropriate." In People v Alemany (13 NY3d 424 [2009]) the Court of Appeals held that these points were properly assigned to a homeless person. The reason, according to the Court, was that the RAI Commentary for this factor described an inappropriate living situation, inter alia, as one which provided a "reduced probability of detection" for law enforcement officers. Since homeless people were more difficult to detect than persons with a known address, the Court held, they could be scored with these points. The RAI does not indicate, however, how a "reduced probability of detection" may be related either to re-offense risk or harm. Courts score points in such cases primarily because the RAI says they should be scored.

Courts have consistently followed the dictates of the Commentary when they have reflected legal determinations or value judgments which are clearly at odds with those of the legislature. RAI factor 4, for example, assigns points for offenders who engage in a continuing course of sexual contact against a victim. The Penal Law also contains continuing course of sexual{**37 Misc 3d at 439} conduct crimes.^[FN6] The RAI Commentary for this factor, however, defines "continuing course of sexual contact" crimes in a manner wholly different from the way such crimes are defined by the Penal Law. (See McFarland I, 2010 NY Slip Op 51705[U], *23-24.) Case law treats the RAI Commentary, however, as the authority in determining whether such conduct should be scored. (See e.g. People v Wizes, 79 AD3d 1543 [3d Dept 2010]; People v Simmonds, 74 AD3d 1505 [3d Dept 2010].) As the Johnson case made plain, our courts have followed the literal dictates of the RAI and its Commentary when it was obvious that such directives produced unintended and [*8]anomalous results.^[FN7] The Commentary has even been held to conclusively define the legal standard which courts must follow in deciding whether to depart from the RAI. (See McFarland I, 2010 NY Slip Op 51705[U], *38-42.)

This court has long believed that our law has afforded undue deference to the RAI and Commentary. In the court's view, however, the instant Position Statement is entitled to the same deference as the Board's other policy pronouncements. It was written by the Board as an official policy statement. The fact that it is labeled as a Position Statement rather than a Commentary, appendix, note or other interpretive aid, in the court's view, is irrelevant to its controlling authority. Indeed, it is the only policy pronouncement by the Board which has not been informed solely by archaic scientific evidence.

Practical Impact of the Position Statement/Analysis of New Departure Considerations

Prior to the Position Statement, child pornography offenders who had no other aggravating factors would be given scores which placed them either slightly above or below the threshold number of points for a level two offender. The difference in such scores arose from whether there was sufficient proof that the victims displayed in images were 10 years old or less (which would result in a level two score) or whether such evidence was{**37 Misc 3d at 440} lacking (in which case, defendants would be presumptively scored at a level one).^[FN8]

The Position Statement now presumptively scores such defendants as level one offenders and invites courts to consider departures based on a range of factors. Three articles are cited in support of those factors.^[FN9] The first, by Jim Tanner, Ph.D. (the Tanner article) outlines 14 factors which the author discovered in her practice reflected an "investment in digital sexual content that is beyond the norm for convicted sex offenders." (Tanner article at 1.) The author asserts that these factors lead to treatment resistance and higher recidivism. No evidence is offered in the article, however, to link any of the listed factors to higher recidivism rates. The article concerns sex offenders generally and is not focused on child pornography offenders.

The Position Statement includes considerations equivalent to 3 of these 14 factors. The three factors in the Position Statement are "paid subscriptions to access child pornography" (which is similar to the article factor entitled "Membership in adult sites or [*9]organizations promoting sexual behavior"); "nature of images (i.e. sadomasochistic)" (which corresponds to the article factor entitled " 'Red Flag' Themes, if they have a significant number of images/files") and "categorized/organized material in their child pornography collection" (which corresponds to the article factor "Any cataloging of sexual content").^[FN10] The Position Statement also includes three other digital sexual content factors not included in the article: the number of{**37 Misc 3d at 441} images possessed, the length of time the offender has been collecting or viewing child pornography and masturbating to child pornographic images.

The second article, by Hanson and Morton-Bourgon, is a meta-analysis of general factors related to sex offender recidivism published in 2004 (the Hanson article). The Position Statement factor "absence of adult sexual relationships" corresponds with the Hanson article's recounting of the fact that a 1998 study determined that the lack of an intimate partner has been associated with increased recidivism. This court made the same point in McFarland I (2010 NY Slip Op 51705[U], *38). This factor is clearly related to recidivism. The odd part about its inclusion in the Position Statement is that it is not listed as a factor courts should rely upon for any other type of sexual offender under any other RAI or Commentary section. In fact, however, there is no evidence the court is aware of that this consideration is more significant for child pornography offenders than other sex offenders.

The "emotional identification with children" factor in the Position Statement is also derived from the Hanson article. The article notes, however, that this pattern in the literature is "most commonly found among extrafamilial child molesters." (Hanson article at 16.) Under the RAI and Commentary, however, it will now be a consideration only in child pornography cases. Many of the most well-documented factors which predict recidivism among sex offenders, meanwhile, including age, time spent offense-free in the community, the presence of male victims or the presence of extra-familial victims continue to be absent from the considerations courts are directed to consider in setting risk levels for any offenders. (See McFarland I, 2010 NY Slip Op 51705[U], *37-38.)

The final article by Seto, Hanson and Babchishin (the Seto article) examines prior contact offenses and re-offense rates for child pornography offenders through two meta-analyses. The{**37 Misc 3d at 442} recidivism study found that 4.6% of online child pornography offenders committed a new sexual offense during a 1.5- to six-year follow-up period. Two percent committed a contact offense and 3.4% committed a new child pornography offense. The meta-analysis included nine studies with 2,630 offenders. The authors caution that although recidivism rates for online offenders appear to be low, the time period measured in the studies was small and the statistics are subject to the caveat that sex crimes are typically underreported. The authors concluded, however, that[*10]"online offenders rarely go on to commit detected contact sexual offenses" and that the recidivism rates for this group were significantly lower than for other sex offenders. The study also indicated that "online offenders who had no history of contact offenses almost never committed contact sexual offenses." (Seto article at 136-137.) Finally, the Position Statement notes as a departure consideration whether an offender has had previous "allegations regarding sexual contact with children." Prior sex offenses against children are already scored by the RAI. The Position Statement, however, apparently now provides that an "allegation" regarding a child contact offense should also be used to inform departure determinations. It is not clear what degree of reliability such an allegation would be required to have in order to be considered relevant in making such a decision.

Conclusions Regarding Position Statement

When compared to prior practice, the Position Statement likely will result in more accurate initial scoring determinations. The recidivism rates for child pornography offenders when compared to other sex offenders appear to be low. The Position Statement, however, in this court's view, continues to leave courts with no useful guidance in determining whether to depart from presumptive RAI scores in setting risk levels.

The list of considerations courts are asked to review in making such determinations all have some relevance. But they do not provide a template courts can rely upon in predicting risk. The list includes a couple of significant factors related to recidivism. But it also omits many more clearly relevant factors which are not otherwise provided for by the RAI or Commentary. It includes a few factors which may indicate that an offender is particularly invested in online sexual content. This court is not aware of any empirical basis to conclude, however, that these discrete factors increase the risk to re-offend. Even assuming such factors do predict recidivism, moreover, the Position Statement{**37 Misc 3d at 443} excludes numerous other considerations which might indicate those identical tendencies.

Informed sex offender risk assessments, in this court's view, require psychiatric evaluations by trained professionals who have reviewed relevant data and a defendant's scores on a validated actuarial risk assessment instrument. Such evaluations are typically not available in SORA proceedings. A review of the list of factors in the Position Statement is no substitute for a valid risk assessment. Risk level determinations under SORA are increasingly used not only for criminal justice and community notification decisions but to dispositively determine fundamental issues about people's lives, like where a sex offender is permitted to live and whom that offender can live with. Yet, in this state, our courts continue to tolerate a system which bases these increasingly important decisions on outdated and inaccurate scoring systems and court assessments which are made in most cases with clearly inadequate information. This court continues to believe that we can and must do better.

Will continue to score either 20 or 30 points for the youngest age depicted in the images under the 'Current Offense' category, and will depart from the presumptive level when appropriate based upon factors including but not limited to:

the number of images possessed (10,000 is more concerning than {**37 Misc 3d at 444}the length of time the offender has been collecting/viewing child porn (i.e. "><100)

the length of time the offender has been collecting/viewing child porn (i.e. >6 months)

paid subscriptions to access child pornography

categorized/organized material in their child pornography collection

absence of adult sexual relationships

emotional identification with children

allegations regarding sexual contact with children

nature of images (i.e. sadomasochistic)

reinforcement of deviant sexual arousal to children by masturbating to these images

Will recommend an automatic override to Level 3 in cases where clinical documentation exists detailing a mental abnormality that decreases the ability to control impulsive sexual behavior, such as Pedophilia or Hebephilia, as provided for on page 22 of the Guidelines.

"The Board remains concerned about child pornography offenders, and in the majority of cases, believes that they have a sexually deviant interest in children which poses a significant risk to public safety; however, recognizes that each person convicted of a child pornography offense poses risks that are unique to that individual. These images are in essence crime scene photos of children being sexually abused, and the increased demand for these images results in further sexual victimization of children." (State of New York Board of Examiners of Sex Offenders, June 1, 2012.)

"(1) Surfing more than 10 hours a week of sexual content.

"(2) High ratio of sexual sites to general surfing, regardless of number of hours.

"(3) Saved versus cached material. As the ratio of saved to cached goes up, so does the risk . . .

"(5) Low ratio of 'Splash Page' to 'Inside Site; images . . .

"(7) Nude pictures of the offender on the offender's devices.

"(8) Pictures with sexual content taken by, created by, or altered by the offender.

"(9) Erotic literature written by the offender.

"(10) Trophy materials stored on the offender's devices.

"(11) Usegroup or Peer to Peer activity seeking sexually explicit materials . . .

"(13) Internet grooming or solicitation of minors using any medium.

"(14) Use of technology for sexual content which indicates a more heavily invested approach."

Footnote 1: See for example, Tanner, J. (2010). Digital Technology Use Factors Which Indicate Increased Sex Offender Investment in Digital Sexual Content. Retrieved on June 1, 2012 from Internet site http://www.kbsolutions.com/KBS14Factors.pdf; Hanson, and Morton-Bourgon (2004). Predictors of Sexual Recidivism: An Updated Meta-Analysis 2004-02. Retrieved from Canadian government Internet site http://www.publicsafety.gc.ca/res/cor/rep/_fl/2004-02-pred-se-eng.pdf; Seto, M.C., Hanson, R.K., & Babchishin, K. M. (2011). Contact sexual offending by men arrested for child pornography offenses. Sexual Abuse: A Journal of Research and Treatment, 23, 124-145.