Leg Length Versus Torso Length in Pedophilia: Further Evidence of Atypical Physical Development Early in Life

Abstract

Adult men’s height results from an interaction among positive and negative influences, including genetic predisposition, conditions in utero, and influences during early development such as nutritional quality, pathogen exposure, and socioeconomic status. Decreased height, reflected specifically as a decreased leg length, is strongly associated with increased risk of poorer health outcomes. Although prior research has repeatedly shown that pedophiles are shorter than nonpedophiles, the largest study to date relied on self-reported height. In the present study, pedophiles demonstrated reduced measured height and reduced leg length as compared with teleiophiles. Given the prenatal and early childhood origins of height, these findings contribute additional evidence to a biological, developmental origin of pedophilia. In addition, the magnitude of this height difference was substantially larger than that found in children exposed to a variety of early environmental stressors, but similar to that seen in other biologically based neurodevelopmental disorders.

Keywords

anthropometry height pedophilia phallometry sex offenders

Introduction

Adult men’s height results from an interaction among positive and negative influences, including genetic predisposition, conditions in utero, and postnatal environmental influences such as nutritional quality, pathogen exposure, and socioeconomic status (Batty et al., 2009). In addition to peak adult height, rate and pattern of growth are also affected (Howe et al., 2012; Li, Dangour, & Power, 2007; Wadsworth, Hardy, Paul, Marshall, & Cole, 2002). These relationships between early hardship and height relate to the negative correlations between height and diseases including heart disease (Batty et al., 2009), stroke (Lee et al., 2009), schizophrenia (Nopoulos, Flaum, Arndt, & Andreasen, 1998; Zammit et al., 2007), and Alzheimer’s disease (Beeri et al., 2005; Petot et al., 2007), as well as mortality (Silventoinen et al., 2006). Large samples of twin pairs suggest that these associations (i.e., between height and heart disease) are largely due to biological factors in the environment (such as childhood diseases, differential diets, etc.), rather than to genetic factors (Silventoinen et al., 2006).

Prior research has identified a height difference between pedophilic and nonpedophilic men. In the largest study reported thus far (Cantor et al., 2007), pedophilic sexual offenders, on average, were 2.1 cm shorter than teleiophilic nonoffenders and 1.3 cm shorter than teleiophilic sexual offenders. This difference persisted after controlling for age. McPhail and Cantor (2015) conducted a meta-analysis of four extant studies concerning height and pedophilia. In addition to each study consistently showing a similar effect size, when their samples were combined, pedophiles were significantly (1.7 cm) shorter than nonpedophiles. Mean height for pedophiles was 174.1 cm, as compared with 175.8 cm for nonpedophiles (McPhail & Cantor, 2015).

Although reduced overall height reflects atypical development, the effect is carried specifically by the length of the legs and not the trunk of the body. Leg length is a sensitive marker of health as increase in the height of populations with improved diet is manifested almost entirely as an increase in leg length, whereas torso length has shown almost no significant correlation with any improvement in health or longevity (Davey Smith et al., 2001; Gunnell et al., 1998). These correlations between increased leg length and improved health outcomes have been shown when leg length was measured both in children (Davey Smith et al., 2001; Gunnell et al., 1998) and in adults (Gunnell et al., 2003; Skidmore, Hardy, Kuh, Langenberg, & Wadsworth, 2007). The strongest positive correlation between improved health and increased leg length is in the domain of coronary heart disease, as a number of studies have shown increased leg length to be linked with more favorable insulin sensitivity, lipid levels, and a variety of cardiovascular outcomes, including reduced cardiovascular mortality (Davey Smith et al., 2001; Skidmore et al., 2007; Whitley, Martin, Davey Smith, Holly, & Gunnell, 2012). The relationship between leg length and improved health remains after adjustment for sociodemographic and health factors, such as childhood and adult socioeconomic status, whether individuals were breastfed as infants, birth weight, maternal smoking status, physical activity, smoking, alcohol consumption, and waist-to-hip ratio (Fraser, Ebrahim, Davey Smith, & Lawlor, 2008; Liu, Akseer, Faught, Cairney, & Hay, 2012; Whitley et al., 2012).

Interestingly, leg length and torso length were previously examined in a sample of gay men (i.e., homosexual teleiophiles), revealing the gay men to be shorter than their heterosexual counterparts and that this difference was attributable to differences in leg length only (Martin & Nguyen, 2004). This indicates that height may be not just related to later negative health outcomes, but may be a generalized marker sensitive to atypical development.

Given the aforementioned correlations of height with indicators of developmental hardship and subsequent atypical adult outcomes, as well as the research indicating pedophiles are of shorter stature than teleiophilic males, it is plausible that height may be a useful marker of developmental events which are pedophilogenic (Cantor et al., 2007). In other words, some circumstances of pre- or perinatal life affect the development of the brain, producing a cluster of neurobehavioral traits including pedophilia. Other evidence supports this hypothesis. For example, pedophiles are approximately 3 times more likely to be non-right-handed than teleiophilic men (Cantor et al., 2004; Cantor, Klassen, et al., 2005), and non-right-handedness has been linked with multiple developmental disorders and stressors (see Blanchard et al., 2007, for a review). Pedophiles have also demonstrated lower IQs than teleiophiles in individual studies as well as a large meta-analysis (Blanchard et al., 2007; Cantor et al., 2004; Cantor, Blanchard, et al., 2005).

Whereas height may be a useful biomarker of pedophilogenesis, an unfortunate limitation of the extant research concerning height as a biological marker of developmental hardship has been its reliance on self-reported, rather than objectively measured, height. Self-reported height is, on average, 2 cm greater than the measured height in males (Flood, Webb, Lazarus, & Pang, 2000; Gunnell et al., 2000). In one study, fewer than 25% of the sample reported their height within 1 cm of their measured height, and almost 20% of males overreported their height by at least 5 cm (Flood et al., 2000). In another study, more than 50% of men overreported their height by 2 cm or more (A. Taylor et al., 2006). Moreover, individuals who are male, shorter, and older tend to overreport their height to a greater extent (Gunnell et al., 2000), as do those who are White and Black (rather than Hispanic) and those who are overweight or obese (Merrill & Richardson, 2009). This tendency has been found in several Western countries, including the United States, Canada, France, and Australia (Flood et al., 2000; Jain, 2010; Merrill & Richardson, 2009; Niedhammer, Bugel, Bonenfant, Goldberg, & Leclerc, 2000; Shields, Connor Gorber, & Tremblay, 2008; A. Taylor et al., 2006), although the magnitude varies by country.

To rule out self-report bias, the present study directly measured height, separately measuring leg length and torso length. Whereas D. Taylor, Myers, Robbins, and Barnard (1993) found a difference in height of more than 4 cm (and utilized measured height), the rest of the literature (Cantor et al., 2007; Jung, Klaver, & Pham, 2014; Mellan, Nedoma, & Pondĕlíčková, 1969) has produced differences in height of about 1 to 2 cm, which is well within the typical range of overreporting of height. Cantor et al. (2007) explicitly used self-reported height, and the methodology of Jung and colleagues (2014) was not made explicit, making the results of these studies potentially influenced by self-report bias. Alternately, given that pedophiles were found to be shorter in these studies and that those who are shorter systematically tend to overreport their height to a greater degree, it is possible that were their height measured, pedophiles would be even shorter yet than those with nonparaphilic sexual preferences.

The present study set out to investigate the relationship between measured height and pedophilia in a group of men referred to a sexual behaviors clinic for a variety of problematic sexual behaviors, including pedophilia. Considering the well-established negative relationship between exaggeration of height and lesser stature, it was hypothesized that the previously established relationship between pedophilia and stature would again emerge and, moreover, would be specific to lesser leg length due to its association with early developmental hardship.

Method

The present investigation involves three erotic age preferences: Pedophilia refers to erotic interest in prepubescent children (von Krafft-Ebing, 1886/1965; with Tanner Stage 1 features, typically under age 11), hebephilia refers to erotic interest in pubescent children (Glueck, 1955; with Tanner Stage 2-3 features, typically ages 11 through 14), and teleiophilia refers to erotic interest in adults (Blanchard et al., 2000; with Tanner Stage 5 features, typically age 17 or above).

Participants

Participants were recruited from the Kurt Freund Laboratory of the Centre for Addiction and Mental Health (Toronto, Ontario, Canada), which provides evaluation services to male patients referred as a result of illegal or clinically significant sexual behaviors or interests. The primary source of referrals to the facility is parole and probation officers, with physicians and lawyers providing others. The standard assessment consists of a psychophysiological (phallometric) examination of the patients’ erotic preferences, a semistructured interview of their sexual history and interests, a review of supplementary psychiatric and legal documents supplied by the referral source, and a brief questionnaire that includes questions about the patient’s height. At the completion of the standard assessment, patients were invited to consent to the use of data for research purposes. Those doing so (and who were fluent English speakers, at least 18 years of age, and not actively psychotic or under the influence of any recreational substances) were invited to participate in an additional assessment in return for a remuneration of Can$90. Interested participants were scheduled for an appointment during which a research assistant (masked to the results of the other tests) conducted an assessment of minor physical anomalies (detailed in Dyshniku, Murray, Fazio, Lykins, & Cantor, 2015) and also measured standing and sitting height.

Two hundred five participants were recruited for the study. Although prior research has shown racial difference in height (Jain, 2010), there were too few non-Caucasian participants in this sample to conduct reliable statistical comparisons. As such, 53 non-Caucasian participants were excluded from further analysis. In addition, 1 participant self-reporting a compressed spine was also excluded. This resulted in 151 cases available for analysis. Phallometric testing reliably classified 142 of those participants. Based on the group assignment criteria stated later, 102 cases were submitted to analyses (see Figure 1).

Figure 1.

Flowchart explaining participant group assignment. Penile plethysmography. (PPG)

The final sample had a mean age of 37.02 years (SD = 12.03; N = 102), ranging from 18 to 74. The median education level was high school graduation. Fifty-two participants (51.0%) were known to have committed a sexual offense against one or more victims age 11 or under, 40 (39.2%) against one or more victims ages 12 to 14, 11 (10.8%) against one or more victims ages 15 to 16, and 27 (26.5%) against one or more victims age 17 or above; 9 participants (8.8%) had no known victims and received assessments following charges of possession of child pornography or because of the patient’s own concern regarding his sexual urges. The sum of these percentages exceeded 100% due to some participants having victims in more than one category. In this sample, no distinction was made between incest and nonincest offenses.

Measures

Sexual offense history

A standardized form was used to record each participant’s history of sexual offenses. The coding of this information included each participant’s number of victims in the aforementioned age ranges. The information came primarily from documents that accompanied the participant’s referral, such as reports from police, probation, or parole officers. Some participants themselves reported additional information regarding offenses that were not included in their files and for which they had not been charged.

Self-reported physical height

On the day of their assessment, participants completed a standard questionnaire of demographic and family background information. The questionnaire included questions regarding participants’ height and permitted participants to respond with either metric or imperial units. For purposes of analysis, all responses were converted into centimeters.

Measured physical height

Participants had their height measured while in their socks on a scale with a height rod. Similar to prior research (e.g., Davey Smith et al., 2001; Fraser et al., 2008; Gunnell et al., 2003), a stool was placed in front of the scale and participants’ sitting height was also measured so that torso length and leg length could be calculated separately. Torso length was computed by taking [sitting height + the height of the scale base from the ground (8.255 cm)] − the height of the stool, which was 72.2 cm. Leg length was then calculated by subtracting torso length from standing height. All participants were measured on the same scale using a standard protocol by a trained clinic staff member.

Phallometric assessment (penile plethysmography)

Blanchard et al. (2007) described the phallometric procedure and data processing techniques in detail. Briefly, a computer records an examinee’s penile blood volume while the examinee observes a standardized set of stimuli that depict a variety of activities and persons of potential erotic interest to the examinee. Change in the examinee’s penile blood volume (i.e., his degree of penile erection) indicates his relative erotic interest in each class of stimuli. The specific phallometric protocol used at the Kurt Freund Laboratory reliably distinguishes pedophilic from teleiophilic men (Blanchard, Klassen, Dickey, Kuban, & Blak, 2001).

The stimuli used in the phallometric test are audio-taped narratives presented through headphones and accompanied by slides. There are seven categories of narratives. They describe sexual interactions with female prepubescent children, female pubescent children, female adults, male prepubescent children, male pubescent children, male adults, or erotically neutral (i.e., nonsexual) activities. Neutral narratives are accompanied by slides of landscapes. The data reduction process (Blanchard et al., 2007) yields seven category scores, one to reflect each of the six combinations of the age group and sex of the stimuli, plus the neutral category.

Group assignment

Participants were divided into three discrete groups for data analysis: pedophiles (n = 20), hebephiles (n = 36), and teleiophiles (n = 46; Blanchard et al., 2007; Cantor et al., 2007). For pedophiles, the categorization was made on the basis of (a) one or more sexual offenses against prepubescent victims and (b) their phallometry results (i.e., participant responded more to prepubescent males or females than to any other category). In the absence of a valid phallometry result, self-reported higher sexual attraction to prepubescent males or females than to any other gender-age category was used to classify the participant as a pedophile. Regarding the self-report variable, participants rated their sexual attraction on a 5-point scale (1 = strongest sexual interest; 5 = no sexual interest) on each gender-age category. If they ranked sexual attraction to prepubescent males or females as lower on the 5-point scale (i.e., lower scores correspond to stronger sexual interest) relative to their ranking of the other gender-age categories on this same scale, then that participant was considered to possess a pedophilic interest. Due to the small number of nonoffending pedophiles (n = 6), a comparison between them and offending pedophiles would have been statistically underpowered and unreliable (see Figure 1).

The corresponding process categorized hebephiles. The teleiophilic group was composed of participants who (a) showed the most erotic arousal to adult males or females on the phallometry assessment or (absent a valid phallometric result) self-reported a sexual preference for adult males and/or females and (b) had no charges or admissions of possessing child pornography (i.e., no history that contradicted the phallometry results or the self-report). This group included both sex offending teleiophiles who had committed one or more sexual offenses against individuals age 17 or above, as well as nonoffending teleiophiles originally referred (through a physician) due to concerns of hypersexuality (see Cantor et al., 2013; also Sutton, Stratton, Pytyck, Kolla, & Cantor, 2015).

Results

Groups’ self-reported and objectively measured heights appear in Table 1. Statistical comparisons were conducted using age and age-squared as covariates to control for any effects of increasing height at younger ages and decreasing height with natural aging (Borkan, Hults, & Glynn, 1983; Gunnell et al., 2000).

Table 1.

Mean Self-Reported and Measured Physical Height by Group.

Group	n	Mean measured height (SD)	n	Mean self-reported height (SD)	Weighted mean difference	t
Pedophiles	20	172.95 (7.89)	20	174.87 (8.26)	1.93	3.94*
Hebephiles	35	174.32 (5.99)	36	178.24 (6.31)	3.64	5.35*
Teleiophiles	46	175.94 (6.85)	46	178.43 (7.40)	2.49	5.91*

p < .01, two-tailed.

The mean self-reported height of the whole sample combined was 177.66 cm (SD = 7.27 cm; N = 102; range = 160.02-198.12 cm). The mean height when measured directly, however, was 174.79 cm (SD = 6.82 cm; n = 101; range = 161.30-192.00 cm), representing a significant decrease, t(100)¹ = 8.60, p < .001, two-tailed. All three groups significantly overestimated their heights (Table 1).

Planned contrasts of self-reported height, with age and age-squared covaried, replicated the prior finding of pedophiles self-reporting lesser height than teleiophiles, with a weighted mean difference of 3.55 cm, d = .47, F(1, 97) = 3.31, p = .036 (one-tailed; Keppel & Wickens, 2004). The height of the hebephilic group was intermediate between that of the pedophilic and teleiophilic groups, and not significantly different from either. Self-reported height was not significantly related to age, F(1, 97) = .08, p = .777, or age-squared, F(1, 97) = .04, p = .838.

Planned contrasts of measured height, with age and age-squared covaried, confirmed the pedophilic group to be significantly shorter than the teleiophilic group, with a weighted mean difference of 3.09 cm, d = .42, F(1, 96) = 2.92, p = .046 (one-tailed). The height of the hebephilic group was again intermediate between that of the pedophilic and teleiophilic groups, and not significantly different from either. The effects of age and age-squared, respectively, were F(1, 96) = 3.42, p = .068, and F(1, 96) = 3.18, p = .078.

Controlling for age arises out of a need to account for potential height gain in younger participants and potential shrinking in older participants (Borkan et al., 1983; Gunnell et al., 2000). The first regression analysis was set up such that three dichotomous variables (dummy coded as follows: pedophilic sex offenders vs. reference category; hebephilic sex offenders vs. reference category; teleiophilic sex offenders vs. reference category; Cohen, Cohen, West, & Aiken, 2003) were simultaneously entered as predictors in Step 1, with age and age-squared entered in Step 2. The second regression was set up in a similar way, with the only difference being the inclusion of the teleiophilic sex offender group, instead of teleiophilic nonoffenders, as the reference category.

Table 2 shows the breakdown of the height measurements into its two components, leg length and torso length. Planned contrasts, with age and age-squared covaried, revealed that the pedophilic group had significantly less leg length than the teleiophilic group, with a weighted mean deficit of 2.37 cm, d = .54, F(1, 85) = 3.56, p = .032 (one-tailed). The effects of age and age-squared, respectively, were F(1, 85) = 3.11, p = .081, and F(1, 85) = 1.79, p = .179.

Table 2.

Mean Measured Leg Length and Torso Length by Group.

Group	n	Mean leg length (SD)	Mean torso length (SD)
Pedophiles	17	79.06 (4.00)	92.80 (4.37)
Hebephiles	32	80.93 (4.14)	93.55 (3.19)
Teleiophiles	41	81.49 (4.80)	94.18 (4.09)

The analogous planned contrast of torso length between the pedophilic and teleiophilic groups was not significant, with a weighted mean difference of 1.55 cm, d = .34, F(1, 85) = 1.95, p = .084 (one-tailed). Neither age, F(1, 85) = .80, p = .375, nor age-squared, F(1, 85) = 1.35, p = .249, was related to torso length.

To explore the data regarding prior findings of shorter leg length among gay men (Martin & Nguyen, 2004), gender orientation (heterosexual vs. homosexual), and age orientation (pedophilic vs. teleiophilic) were submitted to a 2 × 2 ANOVA. The analysis revealed a significant main effect for age orientation, F(1, 46) = 4.13, p = .048 (two-tailed), but not gender orientation, F(1, 46) = .59, p = .448 (two-tailed). The interaction Gender Orientation × Age Orientation was not significant, F(1, 46) = 1.17, p = .285 (two-tailed).

Discussion

Analysis of height was broken down into its two components, leg length and torso length. The leg length of pedophiles was, on average, 2.37 cm shorter than the leg length of teleiophiles, even after adjusting for age. Although the prior finding of shorter leg length among gay men relative to their heterosexual peers (Martin & Nguyen, 2004) was not detected here, pedophiles demonstrated significantly shorter leg length than teleiophiles, consistent with the proposition that shorter leg length reflects early atypical development.

In contrast with leg length, torso length did not differ significantly between pedophiles and teleiophiles. The observed difference in torso length of 1.55 cm corresponded to an effect size that was, although not ignorable, smaller in magnitude to the one observed in leg length. These results are is in line with prior suggestions that, whereas leg length is sensitive to early developmental events, torso length is only weakly associated with correlates of health, longevity, and developmental hardships (Davey Smith et al., 2001; Gunnell et al., 1998; Wadsworth et al., 2002). Alternately, stature in general and leg length in particular are affected by events occurring prenatally (e.g., mothers smoking during pregnancy; Fogelman & Manor, 1988) and during infancy (e.g., during breastfeeding; Wadsworth et al., 2002). Because the shortest stature and leg length were observed among pedophiles, it seems reasonable to suspect that pedophiles may have suffered developmental hardships during infancy or even in utero complications, which could be pedophilogenic in nature. Namely, conditions affecting height pre- or perinatally may heighten the likelihood of developing and/or expressing pedophilia. Along with other findings, such as elevated rates of non-right-handedness (Cantor et al., 2004; Cantor, Klassen, et al., 2005), more head injuries before but not after age 13 (Blanchard et al., 2003), and lower IQ (Blanchard et al., 2007; Cantor et al., 2004), the present results add further evidence to a neurodevelopmental basis of pedophilia.

The present investigation extended and replicated the previously reported relationship between height and pedophilia in a number of ways. First, similar to Cantor et al. (2007) and McPhail and Cantor (2015), the present study found that pedophiles reported being significantly shorter than teleiophiles (by 3.55 cm). Second, pedophiles were, on average, 3.09 cm shorter than teleiophiles. Overall, participants overestimated their actual height by approximately 3 cm, which is comparable with Flood et al.’s (2000) 2 cm difference (on average). The significant discrepancy in stature between pedophiles and teleiophiles persisted even when measured height was examined.

Prior effect size estimates from Cantor et al. (2007) and McPhail and Cantor (2015) have ranged from d = .21 to d = .25 (see Table 3). In the current investigation, the height difference of 3.09 cm between the pedophilic and nonpedophilic group corresponds to a small-to-medium effect, d = .42 (Cohen, 1992). To contextualize the magnitude of this height discrepancy, the height deficit in pedophiles exceeds that of male offspring born to mothers who smoke (1.1 cm shorter; Fogelman & Manor, 1988), mothers who drink during their pregnancy (0.9 cm shorter; Silventoinen, Lahelma, & Rahkonen, 1999), or children who suffer from a serious illness prior to age 6 (1.8 cm shorter; Kuh & Wadsworth, 1989).

Table 3.

Effect Size Estimates of Mean Height Differences Between Pedophilic Sex Offenders and the Nonpedophilic Group.

Study	Mean height of pedophilic sex offenders (SD)	Mean height of nonpedophilic group^a (SD)	Mean difference in cm	Cohen’s d	95% CI	n (pedophilic vs. nonpedophilic group)
McPhail and Cantor (2015; self-reported height)	174.13 (9.26)	175.84 (7.73)	1.71	0.21	[0.08, 0.34]	423 vs. 559
Cantor et al. (2007; self-reported height)	175.60 (8.06)	177.27 (7.62)	1.67	0.22	[−0.43, 0.86]	223 vs. 343
Current manuscript (self-reported height)	174.87 (8.26)	178.43 (7.40)	3.55^b	0.47	[−1.35, 2.29]	20 vs. 46
Current manuscript (measured height)	172.95 (7.89)	175.94 (6.85)	3.09^b	0.42	[−1.28, 2.13]	20 vs. 46

Note. CI = confidence interval.

As per Cantor et al. (2007), the nonpedophilic group is comprised of teleiophilic sex offenders and nonoffenders.

Weighted mean difference in centimeters.

Despite this accumulating line of evidence, further research will be needed to explicate the hypothesized neurodevelopmental basis of pedophilia. For example, research has shown that greater vulnerability to schizophrenia may be acquired through prenatal and perinatal factors including obstetrical complications, asphyxia, and exposure to viruses or infections (Brennan & Walker, 2010). It is also of note that some research has demonstrated differences in height of comparable magnitude among those with schizophrenia to those presented here (2.3-5.1 cm; Houston & Bloom, 1975; Nopoulos et al., 1998). Although the data were not presented here due to the small sample size, if teleiophilic nonoffenders only were used as the control group, the height difference was even greater, and very similar to that in another manuscript evaluating stature in those with schizophrenia (6.8 cm; Brooksbank et al., 1970). Along those lines, it would be of empirical interest to examine whether vulnerability to pedophilia could also be traced back to obstetrical complication and maternal exposure to stressors. Given its relatively well-established neurodevelopmental basis, it is tempting to further follow in the footsteps of schizophrenia research and investigate whether and how the nature and frequency of minor physical anomalies in pedophiles compares with rates observed among individuals with schizophrenia (Compton, Chan, Walker, & Buckley, 2011). Research assessing these hypotheses is underway (see Dyshniku et al., 2015).

Although height is an important anthropometric marker of neurodevelopmental hardship, it is not specific to pedophilogenic factors. It could, instead, serve as a tertiary marker for factors more germane to pedophilogenesis. For example, King (2010) argued that given the correlation of height with IQ and educational level, it is cognitive ability rather than height, which would be the mediating factor. Future studies may also wish to examine potential height difference between offending and nonoffending pedophiles to elucidate other possible mediating factors. This kind of comparison has traditionally been elusive due to the mandatory reporting laws in North America, which have limited the recruitment of nonoffending self-identifying pedophiles. Research originating from the Prevention Project Dunkelfeld in Germany, however, may help shed some light into how offending pedophiles differ from nonoffending pedophiles.

Another line of future research could be exploring early environmental factors that might contribute more specifically to the development and/or expression of pedophilia. Given the observed torso difference of 1.55 cm (d = .34) between teleiophiles and pedophiles, and the findings that a shorter trunk is uniquely associated with childhood serious illness and parental separation (Wadsworth et al., 2002), it is plausible that pedophiles may have been exposed to comparatively more stressful childhood circumstances. In support of this argument, retrospective studies of pedophiles have extensively reported on early abusive experiences and strained parental relationships (i.e., Weeks & Widom, 1998). Furthermore, families that struggle financially may be unable to provide optimal conditions (e.g., nutritious food, medical care during pregnancy, medical care during childhood, etc.) for adequate physical growth during both gestation and early childhood, which could then produce the height deficits in both leg length and torso length observed among the pedophiles in this sample.

These findings are most readily applicable to clinical interventions aimed at individuals with an erotic interest in children. The robust finding of shorter stature among pedophiles, coupled with evidence that height is primarily determined by genetics, conditions in utero, and early developmental hardships (Batty et al., 2009), undermines social learning theories regarding the development of pedophilia. Instead of attempting to “cure” pedophilia through resolving childhood issues, counterconditioning, or conversion, the more viable approach may be management of sexual urges and relapse prevention.

In summary, this article contributes to the growing evidence supporting an early neurodevelopmental origin of pedophilia. Not only were the pedophiles shorter than the other groups, they also demonstrated less of a bias in their self-reported height, which may prove an interesting avenue for future research, as may the especially apparent discrepancies in leg length as compared with torso length or overall height. That the magnitude of these differences was most similar to that seen in schizophrenia—a well-established neurodevelopmental disorder—also raises the question of what other neurological or anthropometric commonalities these disorders may share. Pedophilia was previously compared with autism-spectrum disorders because of the similarities between the two’s rates of non-right-handedness (Cantor, Klassen, et al., 2005). That a number of features demonstrated in those with pedophilia are aligned closely with various neurodevelopmental disorders should serve to shift the focus from environmental or learning theories of pedophilia to a more biological approach.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by Canadian Institutes of Health Research Grants 79276 and 89719 to James M. Cantor.

Notes

References

Batty

Shipley

Gunnell

Huxley

Kivimaki

Woodward

. . . Smith

(2009). Height, wealth, and health: An overview with new data from three longitudinal studies. Economics and Human Biology, 7, 137-152.

Beeri

Davidson

Silverman

Noy

Schmeidler

Goldbourt

(2005). Relationship between body height and dementia. American Journal of Geriatric Psychiatry, 13, 116-123.

Blanchard

Barbaree

Bogaert

Dickey

Klassen

Kuban

Zucker

(2000). Fraternal birth order and sexual orientation in pedophiles. Archives of Sexual Behavior, 29, 463-478.

Blanchard

Klassen

Dickey

Kuban

Blak

(2001). Sensitivity and specificity of the phallometric test for pedophilia in adult male patients stratified by referral source. Psychological Assessment, 13, 118-126.

Blanchard

Kolla

Cantor

J. M.

Klassen

Dickey

Kuban

Blak

(2007). IQ, handedness, and pedophilia in adult male patients stratified by referral source. Sexual Abuse: A Journal of Research and Treatment, 19, 285-309.

Blanchard

Kuban

Klassen

Dickey

Christensen

Cantor

J. M.

Blak

(2003). Self-reported head injuries before and after age 13 in pedophilic and nonpedophilic men referred for clinical assessment. Archives of Sexual Behavior, 32, 573-581.

Borkan

G. A.

Hults

D. E.

Glynn

R. J.

(1983). Role of longitudinal change and secular trend in age differences in male body dimensions. Human Biology, 55, 629-641.

Brennan

P. A.

Walker

E. F.

(2010). Vulnerability to schizophrenia in childhood and adolescence. In Ingram

R. E.

Price

J. M.

(Eds.), Vulnerability to psychopathology: Risk across the lifespan (pp. 363-388). New York, NY: Guilford Press.

Brooksbank

B. W. L.

MacSweeney

D. A.

Johnson

A. L.

Cunningham

A. E.

Wilson

D. A.

Coppen

(1970). Androgen excretion and physique in schizophrenia. British Journal of Psychiatry, 117, 413-420.

10.

Cantor

J. M.

Blanchard

Christensen

Dickey

Klassen

Beckstead

. . . Kuban

(2004). Intelligence, memory, and handedness in pedophilia. Neuropsychology, 18, 3-14.

11.

Cantor

J. M.

Blanchard

Robichaud

Christensen

(2005). Quantitative reanalysis of aggregate data on IQ in sexual offenders. Psychological Bulletin, 131, 555-568.

12.

Cantor

J. M.

Klassen

Dickey

Christensen

Kuban

Blak

. . . Blanchard

(2005). Handedness in pedophilia and hebephilia. Archives of Sexual Behavior, 34, 447-459.

13.

Cantor

J. M.

Klein

Lykins

A. D.

Rullo

J. E.

Thaler

Walling

B. R.

(2013). A treatment-oriented typology of self-identified hypersexuality referrals. Archives of Sexual Behavior, 42, 883-893.

14.

Cantor

J. M.

Kuban

Blak

Klassen

Dickey

Blanchard

(2007). Physical height in pedophilic and hebephilic sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 19, 395-407.

15.

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155-159.

16.

Cohen

West

S. G.

Aiken

L. S.

(2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates.

17.

Compton

M. T.

Chan

R. C.

Walker

E. F.

Buckley

P. F.

(2011). Minor physical anomalies: Potentially informative vestiges of fetal developmental disruptions in schizophrenia. International Journal of Developmental Neuroscience, 29, 245-250.

18.

Davey Smith

Greenwood

Gunnell

Sweetnam

Yarnell

Elwood

(2001). Leg length, insulin resistance, and coronary heart disease risk: The Caerphilly Study. Journal of Epidemiology & Community Health, 55, 867-872.

19.

Dyshniku

Murray

M. E.

Fazio

R. L.

Lykins

A. D.

Cantor

J. M.

(2015). Minor physical anomalies as a window into the prenatal origins of pedophilia. Archives of Sexual Behavior, 44(8):2151–2159.

20.

Flood

Webb

Lazarus

Pang

(2000). Use of self-report to monitor overweight and obesity in populations: Some issues for consideration. Australian and New Zealand Journal of Public Health, 24, 96-99.

21.

Fogelman

K. R.

Manor

(1988). Smoking in pregnancy and development into early adulthood. BMJ: British Medical Journal, 297, 1233-1236.

22.

Fraser

Ebrahim

Davey Smith

Lawlor

D. A.

(2008). The associations between height components (leg and trunk length) and adult levels of liver enzymes. Journal of Epidemiology & Community Health, 62, 48-53.

23.

Glueck

B. C.

Jr. (1955). Final report: Research project for the study and treatment of persons convicted of crimes involving sexual aberrations (June 1952 to June 1955). New York: New York State Department of Mental Hygiene.

24.

Gunnell

Berney

Holland

Maynard

Blane

Frankel

Smith

(2000). How accurately are height, weight and leg length reported by the elderly, and how closely are they related to measurements recorded in childhood? International Epidemiological Association, 29, 456-464.

25.

Gunnell

Davey Smith

Frankel

Nanchahal

Braddon

Pemberton

Peters

(1998). Childhood leg length and adult mortality: Follow up of the Carnegie (Boyd Orr) survey of diet and health in pre-war Britain. Journal of Epidemiology & Community Health, 52, 142-152.

26.

Gunnell

Whitley

Upton

McConnachie

Davey Smith

Watt

(2003). Associations of height, leg length, and lung function with cardiovascular risk factors in the Midspan Family Study. Journal of Epidemiology & Community Health, 57, 141-146.

27.

Houston

B. K.

Bloom

L. J.

(1975). Constitutional factors, stature, and chronic schizophrenia. Journal of Clinical Psychology, 31, 26-29.

28.

Howe

Tilling

Galobardes

Davey Smith

Gunnell

Lawlor

(2012). Socioeconomic differences in childhood growth trajectories: At what age to height inequalities emerge? Journal of Epidemiology & Community Health, 66, 143-148.

29.

Jain

(2010). Regression models to predict corrected weight, height and obesity prevalence from self-reported data: Data from BRFSS 1999-2007. International Journal of Obesity, 34, 1655-1664.

30.

Jung

Klaver

Pham

(2014). Pedophilia and its (non-) association with physical height among sex offenders. Deviant Behavior, 35, 323-331.

31.

Keppel

Wickens

T. D.

(2004). Design and analysis: A researcher’s handbook (4th ed.). Upper Saddle River, NJ: Pearson Education International.

32.

King

(2010). Neurodevelopmental vulnerability for pedophilia: A taxonometric investigation of its latent structure. The University of Maine, Dissertation Abstracts International: Section B. (Sciences and Engineering, 71(8)).

33.

Kuh

Wadsworth

(1989). Parental height: Childhood environment and subsequent adult height in a national birth cohort. International Journal of Epidemiology, 18, 663-668.

34.

Lee

C. M. L.

Barzi

Woodward

Batty

G. D.

Giles

G. G.

Wong

J. W.

. . . Huxley

R. R.

(2009). Adult height and the risks of cardiovascular disease and major causes of death in the Asia-Pacific region: 21 000 deaths in 510 000 men and women. International Journal of Epidemiology, 38, 1060-1071.

35.

Dangour

Power

(2007). Early life influences on adult leg and trunk length in the 1958 British birth cohort. American Journal of Human Biology, 19, 836-843.

36.

Liu

Akseer

Faught

B. E.

Cairney

Hay

(2012). Use of leg length to height ratio to assess the risk of childhood overweight and obesity: Results from a longitudinal cohort study. Annals of Epidemiology, 22, 120-125.

37.

Martin

J. T.

Nguyen

D. H.

(2004). Anthropometric analysis of homosexuals and heterosexuals: Implications for early hormone exposure. Hormones and Behavior, 45, 31-39.

38.

McPhail

I. V.

Cantor

J. M.

(2015). Pedophilia, height, and the magnitude of the association: A research note. Deviant Behavior, 36, 288-292.

39.

Mellan

Nedoma

Pondĕlíčková

(1969). Somatosexuální nálezy u pedofilních mužů [Somatosexual findings in pedophilic men]. Československá Psychiatrie, 65, 30-33.

40.

Merrill

Richardson

(2009). Validity of self-reported height, weight, and body mass index: Findings from the National Health and Nutrition Examination Survey, 2001-2006. Preventing Chronic Disease: Public Health Research, Practice, and Policy, 6, 1-10.

41.

Niedhammer

Bugel

Bonenfant

Goldberg

Leclerc

(2000). Validity of self-reported weight and height in the French GAZEL cohort. International Journal of Obesity, 24, 1111-1118.

42.

Nopoulos

Flaum

Arndt

Andreasen

(1998). Morphometry in schizophrenia revisited: Height and its relationship to premorbid function. Psychological Medicine, 28, 655-663.

43.

Petot

Vega

Traore

Fritsch

Debanne

Friedland

Lerner

(2007). Height and Alzheimer’s disease: Findings from a case-controlled study. Journal of Alzheimer’s Disease, 11, 337-341.

44.

Shields

Connor Gorber

Tremblay

M. S.

(2008). Estimates of obesity based on self-report versus direct measures (Statistics Canada Catalog No. 82-003). Health Reports, 19, 1-16.

45.

Silventoinen

Lahelma

Rahkonen

(1999). Social background, adult body-height and health. International Journal of Epidemiology, 28, 911-918.

46.

Silventoinen

Zdravkovic

Skytthe

McCarron

Herskind

Koskenvue

. . . Kaprio

(2006). Association between height and coronary heart disease mortality: A prospective study of 35,000 twin pairs. American Journal of Epidemiology, 163, 615-621.

47.

Skidmore

Hardy

Kuh

Langenberg

Wadsworth

(2007). Life course body size and lipid levels at 53 years in a British birth cohort. Journal of Epidemiology & Community Health, 61, 215-220.

48.

Sutton

K. S.

Stratton

Pytyck

Kolla

N. J.

Cantor

J. M.

(2015). Patient characteristics by type of hypersexuality referral: A quantitative chart review of 115 consecutive male cases. Journal of Sex & Marital Therapy, 41, 563-580.

49.

Taylor

Dal Grande

Gill

Chittleborough

Wilson

Adams

. . . Ruffin

(2006). How valid are self-reported height and weight? A comparison between CATI self-report and clinic measurements using a large cohort study. Australian and New Zealand Journal of Public Health, 30, 238-246.

50.

Taylor

Myers

Robbins

Barnard

(1993). An anthropometric study of pedophiles and rapists. Journal of Forensic Sciences, 38, 765-768.

51.

von Krafft-Ebing

(1965). Psychopathia sexualis: A medico-forensic study ( Wedeck

H. E.

, Trans.). New York, NY: Putnam. (Original work published 1886).

52.

Wadsworth

Hardy

Paul

Marshall

Cole

(2002). Leg and trunk length at 43 years in relation to childhood health, diet and family circumstances; Evidence from the 1946 national birth cohort. International Journal of Epidemiology, 31, 383-390.

53.

Weeks

Widom

C. S.

(1998). Self-reports of early childhood victimization among incarcerated adult male felons. Journal of Interpersonal Violence, 13, 346-361.

54.

Whitley

Martin

R. M.

Davey Smith

Holly

J. M. P.

Gunnell

(2012). The association of childhood height, leg length and other measures of skeletal growth with adult cardiovascular disease: The Boyd-Orr cohort. Journal of Epidemiology & Community Health, 66, 18-23.

55.

Zammit

Rasmussen

Farahmand

Gunnell

Lewis

Tynelius

Brobert

(2007). Height and body mass index in young adulthood and risk of schizophrenia: A longitudinal study of 1 347 520 Swedish men. Acta Psychiatrica Scandinavica, 116, 378-385.