Classification Of Sex Offenders
It has long been suspected that sex offenders have lower IQs than non-sex offenders. Studies on the topic go as far back as 80 years. A large number of these studies on the topic of sex offender IQ, however, use no classification scheme to break down the sex offenses by subcategories, such as the age group of victim or age group of the offender. In fact, all of the previous studies on the topic of sexual offender IQ lump all subjects into a single, heterogeneous sample with no distinguishing characteristics. As a result, only generalizations can be made without distinguishing certain types of sexual offending behaviors as related to intellectual capacity. In a quantitative review of 236 previous studies on offender IQ, authors Cantor,
…show more content…
Blanchard, Robichaud, and Christensen (2005) conducted a reanalysis to understand how “contradictory findings reported might be reflections of the varying sample compositions” (Cantor, Blanchard, Robichaud, and Christensen, 2005, p. 555-556). More specifically, the purpose of this study is to “assess whether sexual offenders score lower in IQ than nonsexual offenders and to explore which sexual offense characteristics relate to IQ” (Cantor, et al., 2005, p. 555).
Research Problem
According to Cantor, et. al., prolonged scientific interest in the brain studies of sex offenders relates to the implications of the findings when applied to the etiology of sexual offending (2005, p. 555). Since the 19th century, hypotheses have been formed that attribute biological issues in neonatal brain development and childhood head trauma to low IQ, and the authors of this study are able to correlate low IQ with pedophilia using existing literature samples (2005, p. 565). Five questions in this reanalysis were addressed by the authors using literature samples from 236 studies comprised of 25,146 sex offenders and control subjects which recorded such variables as age of offender, IQ score or description of intelligence, type of IQ test used, and year of the report sample. The five questions proposed by Cantor, et al., are as follows:
(1) Do males committing sexual crimes differ in IQ from males committing nonsexual crimes?
(2) Do sexual offenders against adults differ in IQ from sexual offenders against children?
(3) Of the sexual offenders against children, do those selecting intrafamilial children differ in IQ from those selecting extrafamilial children?
(4) Of the sexual offenders against children, do those selecting boys differ in IQ from those selecting girls?
(5) Do juvenile sexual offenders differ in IQ from adult sexual offenders? (Cantor, et al., 2005, p. 557)
Methodology
In order to answer the first of the five questions, samples were broken down into two main groups: sexual offenders or nonsexual offenders. Nonsexual offenders for the purpose of this study are defined as men who were convicted of crimes not sexual in nature (2005, p. 558). IQ information and characteristics of nonsexual offenders and men who had not committed crimes included in the 236 samples were used as a comparison population and control samples against sexual offenders. Each group was assigned a letter code consisting of (a) sexual offenders; (b) nonsexual offenders; (c) nonoffenders. In order to answer the second question, the authors further subdivided the sexual offender samples into two groups based upon the variables of sex offenders with adult victims and sex offenders with child victims. In some samples, the age of the victim was unknown or mixed. Rather than disregard these cases from the study, a third group was added to this variable that included unclassifiable samples (2005, p. 558). The same divisions between extrafamilial and intrafamilial children occurred within the group of sex offenders with child victims. The authors also included stepchildren into the intrafamilial sample subclassification (2005, p. 558). Again, the samples were reclassified further into offenses against boys and offenses against girls (2005, p. 558). In answering the fifth and final question,
Cantor, et al., used the cutoff age of majority of 18 to distinguish adult from juvenile offenders (2005, p. 559).
Statistical Analyses
This reanalysis is considered case-oriented research, which Maxfield and Babbie (2011) define as research “in which many cases are examined to understand a small number of variables” (Maxfield & Babbie, 2011, p. 210). For each case analysis, the dependent variable was the sample IQ, while the independent variables are the sample characteristics of offense type: age group of the victims, intrafamilial victims, extrafamilial victims, gender of child victims, and the age group of sample cases. Due to the age span of the samples, which cover nearly 80 years of research, the authors introduced two additional covariants: the year of the sample publication and the IQ reporting format (2005, p. 559). The reason for introducing the year covariant, according to Cantor et al., was to “control for what has been called the Flynn effect, that is, the demonstrated increase in population IQ scores over time” (2005, p. 559). The IQ data was adjusted using a mean of 85 to compensate for the increase in IQ scores over time. Finally, the authors were not able to determine from any of the 236 samples the gender of the victims of juvenile offenders. The samples simply did not provide this variable information in the original sample literature.
Results
The sample groups consisted of 165 adults and 71 juveniles.
The findings were presented clearly by the authors using six bar graph figures. Figure one displayed the mean IQ score of 85 as the base and comparing (a) sex offenders; (b) nonsexual offenders; and (c) nonoffenders. The results of this test found that the literature supports the author’s initial hypothesis that as a whole, men who commit sexual offenses have lower IQ scores than those who commit nonsexual offenses (2005, p. 559). Figure two compared the following using a mean IQ of 85: sexual offenders against children; sexual offenders against adults; nonsexual offenders; nonoffenders; and sexual offenders with victims of unknown age (2005, p. 560). Again, the results support the hypothesis that the IQ of sex offenders correlates with the age of the victim (2005, p. 560). Figure three demonstrated mean IQ scores of samples of adult sex offenders against children, but was difficult for the reader to interpret the results. Figure four compared the IQ scores for intrafamilial and extrafamiliar sex offenders, and these findings support the hypothesis that IQ of sex offenders who commit intrafamilial sex offenses have lower average IQ scores than those who commit extrafamilial offenses (2005, p. 562). Figures five and six demonstrate IQ scores of sex offenders against children, against adults, and those with unclassifiable victims. Once again, Cantor, et al., was able to support the hypothesis that sex offenders with male victims score slightly lower than those who have female victims (2005, p. 563). Finally, figure six gives a generalized comparison of mean IQ scores for the adult vs juvenile categories of (a) sex offenders; (b) nonsexual offenders; and (c) nonoffenders. In all three categories, adults scored significantly higher in IQ than juveniles (2005, p.
564).
Discussion and Conclusion This quantitative reanalysis of sex offender IQ data is well researched and the results are clearly stated by the authors. The authors also bring up some very intriguing discussion points in their conclusion. These results prove that sex offenders have lower average IQ scores than nonsexual offenders, and IQ scores drop with the age of the victim. This study moves the field of research into the next phase by asking why these types of offenders show lower IQ scores. Cantor, et al. explains that these “IQ differences are genuine and reflect an underlying deficiency of brain function” (2005, p. 565). By suggesting that further research should be conducted into the prenatal and childhood brain development of pedophiles, studies in this field may yield valuable results about the etiology of sex offenders. The authors validate the scope and relevance of this study as applied to future research on biological influences in the development of pedophiles.
Blanchard, Robichaud, and Christensen (2005) conducted a reanalysis to understand how “contradictory findings reported might be reflections of the varying sample compositions” (Cantor, Blanchard, Robichaud, and Christensen, 2005, p. 555-556). More specifically, the purpose of this study is to “assess whether sexual offenders score lower in IQ than nonsexual offenders and to explore which sexual offense characteristics relate to IQ” (Cantor, et al., 2005, p. 555).
Research Problem
According to Cantor, et. al., prolonged scientific interest in the brain studies of sex offenders relates to the implications of the findings when applied to the etiology of sexual offending (2005, p. 555). Since the 19th century, hypotheses have been formed that attribute biological issues in neonatal brain development and childhood head trauma to low IQ, and the authors of this study are able to correlate low IQ with pedophilia using existing literature samples (2005, p. 565). Five questions in this reanalysis were addressed by the authors using literature samples from 236 studies comprised of 25,146 sex offenders and control subjects which recorded such variables as age of offender, IQ score or description of intelligence, type of IQ test used, and year of the report sample. The five questions proposed by Cantor, et al., are as follows:
(1) Do males committing sexual crimes differ in IQ from males committing nonsexual crimes?
(2) Do sexual offenders against adults differ in IQ from sexual offenders against children?
(3) Of the sexual offenders against children, do those selecting intrafamilial children differ in IQ from those selecting extrafamilial children?
(4) Of the sexual offenders against children, do those selecting boys differ in IQ from those selecting girls?
(5) Do juvenile sexual offenders differ in IQ from adult sexual offenders? (Cantor, et al., 2005, p. 557)
Methodology
In order to answer the first of the five questions, samples were broken down into two main groups: sexual offenders or nonsexual offenders. Nonsexual offenders for the purpose of this study are defined as men who were convicted of crimes not sexual in nature (2005, p. 558). IQ information and characteristics of nonsexual offenders and men who had not committed crimes included in the 236 samples were used as a comparison population and control samples against sexual offenders. Each group was assigned a letter code consisting of (a) sexual offenders; (b) nonsexual offenders; (c) nonoffenders. In order to answer the second question, the authors further subdivided the sexual offender samples into two groups based upon the variables of sex offenders with adult victims and sex offenders with child victims. In some samples, the age of the victim was unknown or mixed. Rather than disregard these cases from the study, a third group was added to this variable that included unclassifiable samples (2005, p. 558). The same divisions between extrafamilial and intrafamilial children occurred within the group of sex offenders with child victims. The authors also included stepchildren into the intrafamilial sample subclassification (2005, p. 558). Again, the samples were reclassified further into offenses against boys and offenses against girls (2005, p. 558). In answering the fifth and final question,
Cantor, et al., used the cutoff age of majority of 18 to distinguish adult from juvenile offenders (2005, p. 559).
Statistical Analyses
This reanalysis is considered case-oriented research, which Maxfield and Babbie (2011) define as research “in which many cases are examined to understand a small number of variables” (Maxfield & Babbie, 2011, p. 210). For each case analysis, the dependent variable was the sample IQ, while the independent variables are the sample characteristics of offense type: age group of the victims, intrafamilial victims, extrafamilial victims, gender of child victims, and the age group of sample cases. Due to the age span of the samples, which cover nearly 80 years of research, the authors introduced two additional covariants: the year of the sample publication and the IQ reporting format (2005, p. 559). The reason for introducing the year covariant, according to Cantor et al., was to “control for what has been called the Flynn effect, that is, the demonstrated increase in population IQ scores over time” (2005, p. 559). The IQ data was adjusted using a mean of 85 to compensate for the increase in IQ scores over time. Finally, the authors were not able to determine from any of the 236 samples the gender of the victims of juvenile offenders. The samples simply did not provide this variable information in the original sample literature.
Results
The sample groups consisted of 165 adults and 71 juveniles.
The findings were presented clearly by the authors using six bar graph figures. Figure one displayed the mean IQ score of 85 as the base and comparing (a) sex offenders; (b) nonsexual offenders; and (c) nonoffenders. The results of this test found that the literature supports the author’s initial hypothesis that as a whole, men who commit sexual offenses have lower IQ scores than those who commit nonsexual offenses (2005, p. 559). Figure two compared the following using a mean IQ of 85: sexual offenders against children; sexual offenders against adults; nonsexual offenders; nonoffenders; and sexual offenders with victims of unknown age (2005, p. 560). Again, the results support the hypothesis that the IQ of sex offenders correlates with the age of the victim (2005, p. 560). Figure three demonstrated mean IQ scores of samples of adult sex offenders against children, but was difficult for the reader to interpret the results. Figure four compared the IQ scores for intrafamilial and extrafamiliar sex offenders, and these findings support the hypothesis that IQ of sex offenders who commit intrafamilial sex offenses have lower average IQ scores than those who commit extrafamilial offenses (2005, p. 562). Figures five and six demonstrate IQ scores of sex offenders against children, against adults, and those with unclassifiable victims. Once again, Cantor, et al., was able to support the hypothesis that sex offenders with male victims score slightly lower than those who have female victims (2005, p. 563). Finally, figure six gives a generalized comparison of mean IQ scores for the adult vs juvenile categories of (a) sex offenders; (b) nonsexual offenders; and (c) nonoffenders. In all three categories, adults scored significantly higher in IQ than juveniles (2005, p.
564).
Discussion and Conclusion This quantitative reanalysis of sex offender IQ data is well researched and the results are clearly stated by the authors. The authors also bring up some very intriguing discussion points in their conclusion. These results prove that sex offenders have lower average IQ scores than nonsexual offenders, and IQ scores drop with the age of the victim. This study moves the field of research into the next phase by asking why these types of offenders show lower IQ scores. Cantor, et al. explains that these “IQ differences are genuine and reflect an underlying deficiency of brain function” (2005, p. 565). By suggesting that further research should be conducted into the prenatal and childhood brain development of pedophiles, studies in this field may yield valuable results about the etiology of sex offenders. The authors validate the scope and relevance of this study as applied to future research on biological influences in the development of pedophiles.