<%BANNER%>

Sexual Deviance amongst Non-Sex Offenders

Permanent Link: http://ufdc.ufl.edu/UFE0022805/00001

Material Information

Title: Sexual Deviance amongst Non-Sex Offenders A Social Learning Analysis
Physical Description: 1 online resource (198 p.)
Language: english
Creator: Bowman, Jason
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: deviance, learning, normal, offender, sex, sexual, social, theory
Psychology -- Dissertations, Academic -- UF
Genre: Counseling Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In place of the familiar approach of examining sexual deviance through sex offender populations, this study targeted a non-sex offender population (i.e., college students). A new scale named the Sexual Deviance Inventory (SDI) was designed to measure a person?s overall proclivity toward inappropriate or ?deviant? sexual interests. Instead of finding that sexual deviance was statistically deviant, the SDI detected that the experience of sexual deviance was a relatively normal experience within this non-sex offender population as 90% of the 353 online survey participants reported having had experienced at least 1 of the 24 deviant sexual interests listed in the SDI and, on average, participants endorsed having had experienced 5 (or 22%) of the 24 deviant sexual interests listed in the survey. A social learning theory (SLT) model comprised of several SLT measures created specifically for this study accounted for the 53% of the variance in the SDI scores, R2 = .53, F (6, 240) = 44.20, p < .0005. This SLT model did a better job predicting SDI scores than sexual deviance measures that assessed only specific categories of interests (i.e., the SICSQ-R subscales). The SLT model only accounted for between 5% and 34% of the variances within the subscales. The SLT model also did a better job predicting SDI scores than a measure that assessed a combination of sexual interests restricted to the most extreme deviant sexual interests (i.e., the SICSQ-R). The SLT model only accounted for 38% of the variance for the SICSQ-R (males only), R2 = .38, F (6, 33) = p = .011 versus 49% of the SDI (males only), R2 = .49, F (6, 35) = p < .0005. Findings from this study suggest that sexual deviance should be studied within non-sex offender populations. Findings also suggest that the construct of sexual deviance may be more accurately represented by a scale that assesses a combination of sexual interests comprised of various levels of deviant sexual interests.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jason Bowman.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Neimeyer, Greg J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0022805:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022805/00001

Material Information

Title: Sexual Deviance amongst Non-Sex Offenders A Social Learning Analysis
Physical Description: 1 online resource (198 p.)
Language: english
Creator: Bowman, Jason
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: deviance, learning, normal, offender, sex, sexual, social, theory
Psychology -- Dissertations, Academic -- UF
Genre: Counseling Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In place of the familiar approach of examining sexual deviance through sex offender populations, this study targeted a non-sex offender population (i.e., college students). A new scale named the Sexual Deviance Inventory (SDI) was designed to measure a person?s overall proclivity toward inappropriate or ?deviant? sexual interests. Instead of finding that sexual deviance was statistically deviant, the SDI detected that the experience of sexual deviance was a relatively normal experience within this non-sex offender population as 90% of the 353 online survey participants reported having had experienced at least 1 of the 24 deviant sexual interests listed in the SDI and, on average, participants endorsed having had experienced 5 (or 22%) of the 24 deviant sexual interests listed in the survey. A social learning theory (SLT) model comprised of several SLT measures created specifically for this study accounted for the 53% of the variance in the SDI scores, R2 = .53, F (6, 240) = 44.20, p < .0005. This SLT model did a better job predicting SDI scores than sexual deviance measures that assessed only specific categories of interests (i.e., the SICSQ-R subscales). The SLT model only accounted for between 5% and 34% of the variances within the subscales. The SLT model also did a better job predicting SDI scores than a measure that assessed a combination of sexual interests restricted to the most extreme deviant sexual interests (i.e., the SICSQ-R). The SLT model only accounted for 38% of the variance for the SICSQ-R (males only), R2 = .38, F (6, 33) = p = .011 versus 49% of the SDI (males only), R2 = .49, F (6, 35) = p < .0005. Findings from this study suggest that sexual deviance should be studied within non-sex offender populations. Findings also suggest that the construct of sexual deviance may be more accurately represented by a scale that assesses a combination of sexual interests comprised of various levels of deviant sexual interests.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jason Bowman.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Neimeyer, Greg J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0022805:00001


This item has the following downloads:


Full Text

PAGE 1

1 SEXUAL DEVIANCE AMONGST NONSEX OFFENDERS: A SOCIAL LEARNING ANALYSIS By JASON BOWMAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Jason Bowman

PAGE 3

3 To Mia

PAGE 4

4 ACKNOWLEDGEMENTS I would like to thank the chair and members of my supervisory committee, and the participants in my surveys for being open and honest about extremely personal information. I would also like to thank my family for supporting me both emotionally and financially.

PAGE 5

5 TABLE OF CONTENTS page LIST OF TABLES ...........................................................................................................................9 LIST OF FIGURES .......................................................................................................................11 ABSTRACT ...................................................................................................................................12 CHAPTER 1 INTRODUCTION ..................................................................................................................14 Focus on the Sex Offender .....................................................................................................15 Fear of Victimization .......................................................................................................15 Medicalization of the Construct ......................................................................................16 Sampling Convenience ....................................................................................................17 Methodological and Statistical Problems ...............................................................................18 Construct Inaccuracy .......................................................................................................18 Sample Selection Bias .....................................................................................................19 Truncated Range of Scores ..............................................................................................20 Limited Data ....................................................................................................................21 2 REVIEW OF RELEVANT LITERATURE ...........................................................................23 Overview .................................................................................................................................23 Theories of Sexual Deviance ..................................................................................................23 Social Learning Theory ...................................................................................................24 Differential association ............................................................................................26 Dif ferential reinforcement ........................................................................................26 Modeling ..................................................................................................................26 Definition .................................................................................................................27 Other Theor ies of Sexual Deviance .................................................................................27 Expressions of Sexual Deviance .............................................................................................28 Sexual Behavior ...............................................................................................................29 Prohibited ........................................................................................................................29 Non -Prohibited ................................................................................................................33 Sexual Fantasy .................................................................................................................34 Pornography ....................................................................................................................36 Masturbation ....................................................................................................................39 Assessments ............................................................................................................................40 Historical Assessments ....................................................................................................42 Self reports ...............................................................................................................42 Criminal records .......................................................................................................42 Collateral intervie ws ................................................................................................42 Physiological Assessments ..............................................................................................43 Abel assessment for sexual interest (AASI) .............................................................43

PAGE 6

6 Vaginal and penile plethysmographs (PPG) ............................................................43 Polygraph .................................................................................................................44 Scales, Inventories and Questionnaires ...........................................................................45 Sexual interest card sort questionnaire revised (SICSQ R) ..................................45 Multiphasic sex inventory second edition .............................................................45 Clarke sexual history scale for males .......................................................................46 Wilson sexual fantasy questionnaire ........................................................................46 Psychopathy che ck list revised ..............................................................................46 3 METHODS AND MATERIALS ...........................................................................................51 Overview .................................................................................................................................51 Phas e 1 ....................................................................................................................................51 Social Learning Theory ...................................................................................................51 Construct items .........................................................................................................51 Eliminate it ems .........................................................................................................52 Develop scoring method ...........................................................................................52 Sexual Deviance Inventory ..............................................................................................52 Co nstruct items .........................................................................................................53 Collect community standards of deviance severity data ..........................................53 Eliminate items .........................................................................................................54 Format questionnaire ................................................................................................55 Develop scoring method ...........................................................................................55 Phase 2 ....................................................................................................................................56 Survey Administration .....................................................................................................56 Social learning theory scales ....................................................................................57 Sexual deviance measures ........................................................................................59 Other Measures ................................................................................................................60 Data Analyses .........................................................................................................................61 Pre -Study .........................................................................................................................61 Sample size ...............................................................................................................61 Phase 1 .............................................................................................................................61 Deviance severity data .............................................................................................61 Phase 2 .............................................................................................................................61 SLT data ...................................................................................................................61 Sexual deviance data ................................................................................................62 Relationship between SLT and sexual deviance ......................................................62 4 RESULTS ...............................................................................................................................65 Overview .................................................................................................................................65 Phase 1 ....................................................................................................................................65 Participants ......................................................................................................................65 SLT Measures ..................................................................................................................66 Community Standards of Sexual Deviance .....................................................................66 Phase 2 ....................................................................................................................................67 Participants ......................................................................................................................67

PAGE 7

7 SLT Data and Scale Scores .............................................................................................68 SLT scale construction .............................................................................................68 SLT item responses ..................................................................................................68 SLT scale scores .......................................................................................................68 Sexual Deviance Data and Measures ..............................................................................68 Sexual Deviance Inventory ..............................................................................................69 SDI scores ................................................................................................................71 Sexual Deviance and Social Learning Theory ........................................................................72 Correlations .....................................................................................................................72 Controlling for the SDS ...................................................................................................73 Multiple Regressions .......................................................................................................73 5 DISCUSSION .........................................................................................................................93 Overview and Hypothesis .......................................................................................................93 Conceptual Shift .....................................................................................................................93 Behavior versus Interest ..................................................................................................94 Pathological versus Normal .............................................................................................94 Specific versus Generalized ............................................................................................94 Dichotomou s versus Dimensional ...................................................................................94 What is Sexual Deviance? ......................................................................................................95 Deviance Severity ............................................................................................................95 Sexual Interest .................................................................................................................95 Prevalence of deviant sexual interest .......................................................................96 Gender differences in deviant sexual interests .........................................................97 Co occurrence of deviant sexual interests ..............................................................100 Deviance Severity and Sexual Interest ..........................................................................100 Sexual Deviance Inventory ...................................................................................................101 Social Learning Theory and Sexual Deviance ......................................................................103 Differential Association .................................................................................................104 Differential Reinforcement Peers ..................................................................................104 Differential Reinforcement Physiology .........................................................................105 Reinforcement Balance .................................................................................................105 Modeling ........................................................................................................................106 Definition .......................................................................................................................106 Li mitations ............................................................................................................................106 6 CONCLUSION .....................................................................................................................108 APPENDIX A PILOT STUDY .....................................................................................................................109 Participants ...........................................................................................................................109 Materials ...............................................................................................................................110 Procedures .............................................................................................................................111 Results ...................................................................................................................................112

PAGE 8

8 Sexual Deviance Inventory Beta (SDI ) ...................................................................112 Scoring ...................................................................................................................112 Non -deviant/deviant cutoff ....................................................................................112 Deviant sexual interest ...........................................................................................112 Personality correlates .....................................................................................................113 Conclusions ...........................................................................................................................113 B SURVE Y MATERIALS: PHASE I .....................................................................................117 C SURVEY MATERIALS: PHASE II ....................................................................................130 REFERENCES ............................................................................................................................186 BIOGRAPHICAL SKETCH .......................................................................................................197

PAGE 9

9 LIST OF TABLES Table page 2-1 Theories of sexual deviance ...............................................................................................48 2-2 Comparison of R-square for several studies examining the relationship between social learning theory and deviant behavior (Akers, 1998) ...............................................49 2-3 List of common sex crimes (Sex crim es, n.d.) ...................................................................49 2-4 List of paraphilias (DSM IV -TR, 2000) ............................................................................50 3-1 Calculating the SDI score ..................................................................................................64 4-1 Deviance severity ratings from Phase 1 .............................................................................80 4-2 Differential Association .....................................................................................................82 4-3 Differential Reinfo rcement Peer ........................................................................................83 4-4 Differential Reinforcement Physiology .............................................................................83 4-5 Reinforcement Balance ......................................................................................................84 4-6 Modeling ............................................................................................................................84 4-7 Definitions ..........................................................................................................................85 4-8 Descriptive statistics for social learning theory variables ..................................................85 4-9 Correlation matrix for social learning theory variables .....................................................85 4-10 SDI frequency from Phase 2 ..............................................................................................86 4-11 SDI lifetime prevalence from Phase 2 ...............................................................................88 4-12 Descriptive statistics for sexual deviance variables ...........................................................89 4-13 Correlation matrix for sexual deviance variables ..............................................................90 4-14 Correlation for the SLT variables for the sexual deviance variables controlling and not controlling for social desirab ility (across gender) .......................................................90 4-15 Correlation across sexual deviance variables for the SLT variables controlling for social desirability (males only) ..........................................................................................91 4-16 Multiple regression analyses testing the utility of the SLT model in expaining sexual deviance in the generalized sexual deviance variables (across gender) ............................91

PAGE 10

10 4-17 Mu ltiple regression analyses testing the utility of the SLT model in explaining sexual deviance in the sexual deviance variables (males only) ....................................................92 A-1 SDI lifetime prevalence from Pilot Study ....................................................................115

PAGE 11

11 LIST OF FIGURES Figure page 3-1 Sample of three completed SDI items ...............................................................................64 4-1 SDI score distribution (all participants). ............................................................................75 4-2 SDI score distribution (males). ..........................................................................................76 4-3 SICS QR score distribution (males). .................................................................................77 4-4 The relationship between deviance severity and frequency of sexual interest ..................78 4-5 The re lationship between deviance severity and lifetime prevalence for sexual interest ................................................................................................................................79

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SEXUAL DEVIANCE AMONGST NONSEX OFFENDERS: A SOCIAL LEARNING ANALYSIS By Jason Bowman May 2010 Chair: Greg Neimeyer Major: Counseling Psychology In place of the familiar approach of examining sexual deviance through sex offender populations, this study targeted a non-sex offender population (i.e., college students). A new scale named the Sexual Deviance Inventory (SDI) was designed to measure a persons overall proclivity toward inappropriate or deviant sexual interests. Instead of finding that sexual deviance was statistically deviant, the SDI detected that the experience of sexual deviance was a relatively normal experience within this non -sex offender population as 90% of the 353 online survey participants reported having had experienced at least 1 of the 24 deviant sexual interests listed in the SDI and, on average, participants endorsed having had experienced 5 (or 22%) of the 24 deviant sexual interests listed in the survey. A social learn ing theory (SLT) model comprised of several SLT measures created specifically for this study accounted for the 53% of the variance in the SDI scores, R2 = .53, F (6, 240) = 44.20, p < .0005. This SLT model did a better job predicting SDI scores than sexua l deviance measures that assessed only specific categories of interests (i.e., the SICSQ -R subscales). The SLT model only accounted for between 5% and 34% of the variances within the subscales. The SLT model also did a better job predicting SDI scores th an a measure that assessed a combination of sexual interests restricted to the most

PAGE 13

13 extreme deviant sexual interests (i.e., the SICSQ -R). The SLT model only accounted for 38% of the variance for the SICSQ -R (males only), R2 = .38, F (6, 33) = p = .011 versus 49% of the SDI (males only), R2 = .49, F (6, 35) = p < .0005. Findings from this study suggest that sexual deviance should be studied within non-sex offender populations. Findings also suggest that the construct of sexual deviance may be more accurately represented by a scale that assesses a combination of sexual interests comprised of various levels of deviant sexual interests.

PAGE 14

14 CHAPTER 1 INTRODUCTION With respect to the concept of inappropriate arousal itself, there are no theories of etiology which have other than laboratory demonstrations of plausibility or anecdotal support. We simply do not know how to account for individual differences in sexual arousal patterns. Clearly, developmental studies of the acquisition of sexual preferences should receive high priority. V.L. Quinsey & W.L. Marshall, Procedures for reducing inappropriate sexual arousal: An evaluation review There is no comprehensive theory that is agreed upon to explain the development of paraphilic behaviorResearch must be conducted on the development of sexual interest patterns if we are to understand and control paraphilic behavior. Attorney Generals Commission on Pornography The genesis of deviant sexual behavior in childhood, adolescence, and adulthood still remains obscure despite a wide range of studies. Levesque, Adolescents, sex, and the law: Preparing adolescents for responsible citizenship Empirical data continues to elude the field of sexual deviance research. Sexual deviance is generally understood to be the expression of deviant or inappropriate sexual interests through fantasies, masturbation, pornography, behaviors, or any combination thereof (Laws & ODonohue, 1997). As a result, the interest can be expressed in ways that are as innocuous as a fantasy or as conseque ntial as a felony. Using this broader conceptualization, sexual deviance becomes much more prevalent previously considered. In a study by Bowman and Schneekloth (1999), nearly 97% of college student population reported that they had experienced sexual a rousal toward a scenario that was considered sexually deviant Furthermore, participants endorsed having been sexually interested in an average of 131 of the 38 deviant sexual scenarios listed in the study. Individuals even endorsed being sexually intere sted in scenarios that paraphilic and/or criminal in nature. For example, 50% of males endorsed having being sexually 1 Categories of deviance were designated as part of the study, see Appendix A

PAGE 15

15 interested in Post -Pubescent Pedophilia2 and 34% of all participants endorsed having been sexually interested in Physically Coercive Sex.3 Based on data from this pilot study, it seems that instead of sexual deviance being statistically deviant it may be the norm. Focus on the Sex Offender Despite the prevalence of deviant sexual interests throughout the general population, researchers continue to limit the scope of sexual deviance research to sex offender populations (Abel, Jordan, Hand, Holland, & Phillips, 2001; Laws & ODonohue, 1997; Stinson & Becker, 2008). Three possible explanations for this pressured focus are the a) fear of vic timization b) the medicalization of the construct and c) sampling convenience. Fear of V ictimization Few crimes evoke as much fear as sexual offenses. Child molestation and sexual assault consistently rank as some of the most severe crimes that a person can commit. In fact, these two offenses are only surpassed in crime severity by first degree murder ( Advisory Commission on Sentencing, 2002) The number of people that experience victimization at the hands of sex crime makes these threats especially palpable as 27 percent of women and 16 percent of men report having been put through these demoralizing experiences (Finkelhor et al., 1990). Also, the individuals that commit these sexual acts seem to be especially culpable as it is the norm to be charged with multiple offenses. Astonishingly, the typical child sex offender is reported to molest an average of 117 children (Abel, Becker, Cunningham -Rathner, Mittelman, & Rouleau, 1988). It is believed that fear of victimization has incited a rush to cure se x offenders. It is almost as if researchers have crowded around these populations with magnifying glasses in 2 A child that has reached the age of puberty but not yet reached the age of consent 3 The use of physical f orce to induce sexual relations

PAGE 16

16 hopes of somehow finding the disease mechanism. Ironically, this intense focus on the sex offender is likely what is hindering theoretical progre ss Medicalization of the Construct The categorization of psychological constructs is a necessary practice that has aided treatment standardization and health care communications. Alongside the benefits of this classification system there have been concern s over institutional proscription. When socially problematic behaviors cannot be easily controlled by the criminal justice system, they are occasionally classified as mental disorders (Thomas Szasz, 1974) This practice allows for the control of socially disruptive behaviors that are not necessarily criminal. For example, a child that is unruly or disruptive and difficult to control may be given a diagnosis of Conduct Disorder or Oppositional Defiant Disorder. An individual that is hostile or physically abusive may be given the diagnosis of Impulse Control Disorder or Antisocial Personality Disorder. In extreme cases of social non compliance, such as when a person is considered a threat to him or herself or others, the civil commitment process allows the state to temporarily take control of a persons free will. The problem with using mental diagnoses as a means for social control is that it erodes the authenticity of what it means to be mentally ill. For example, within each set of diagnostic criteria there exists an im paired functioning criterion. In order to qualify, the condition needs to cause significant distress or impair one or more major areas of functioning such as work, interpersonal relations, or self -care (APA, 2000). A mental diagnosis i s questionable when the condition has not caused significant distress or impaired a persons func tioning. It is noteworthy that the impairment criterion is often altered or, at times, for conditions that could be considered socially problematic behaviors (APA, 2000). An argument against diagnosing socially disruptive behaviors as mental disorders would not suggest that these behaviors should go

PAGE 17

17 unchecked. It does, however, raise questions related to whether or not behaviors should be listed as a mental d isorder versus if they should just be designated as illegal and left up to the criminal justice system to reinforce. This controversy is especially applicable to sexual interests. Moser and Kleinplatz (2005) specifically catalogued the reasons specific sexual interests, a.k.a. paraphilias, do not fit the criteria for mental disorders and argued that they should be removed from psychiatric classification. The consequence of defining sexual deviance as a mental disorder has been that it has forced sexual de viance into a false dichotomy that requires sexual interests to be either categorized as normal or pathological. Sexual deviance research has followed suit and dedicated all of its resources to understanding sexual deviance as the pathology that is to bla me for sexual misconduct. Conceptualizing sexual deviance as pathology has prevented researchers from accepting that perhaps sexual deviance is normal. Maybe everyone experiences this proclivity and the major difference between sex offenders and nonsex offenders is not the proclivity toward deviant sexual interest but, rather, behavioral constraint and criminal intent. Belief in this false dichotomy has also prevented researchers from examining sexual deviance as a dimensional constructs. It is believ ed that empirical data will accumulate once researchers start exploring the spectrum of sexual deviance found within the general public. Sampling Convenience The prevalence and accessibility of sex offenders within sex offender treatment programs makes the se populations convenient to sample. According to the Safer Society Survey (McGrath, Cumming, & Burchard, 2003) there are 1,549 sex offender treatment programs in the U.S. with an average of about 70 adult sex offenders treated each year per program. Whi le in these programs, the incarcerated individuals tend to be captive audiences making data collection relatively simple. Although the practice of coercing incarcerated individuals into

PAGE 18

18 research participation is unethical, concerns have been cited because there is no national regulatory body that currently oversees ethics related to prison research (Kalmbach & Lyons, 2002). It is understood research conducted on sex offender populations is extremely important. These studies are useful in revealing facto rs associated with why certain individuals choose to act their sexual deviance in ways that are prohibited by law. However, it does not effectively reveal information related to why the deviant sexual interest exists. Limiting the study of sexual deviance to sex offenders has had several unintended consequences that will be described in the next section. Because of these, it is suggested that researchers begin to look toward nonsex offender populations to reveal nature of sexual deviance. Methodological and Statistical Problems Limiting the study of sexual deviance to sex offender populations has made sexual deviance research vulnerable to the following errors: a) construct inaccuracy, b) sample selection bias, c) truncated ranges of sexual deviance, and d) the limited scope of assessments. Construct Inaccuracy A prerequisite for theory testing is the accurate conceptualization and measurement of the construct in question (Coates, 1995). As referenced through several studies, an overall proclivity for se xual deviance exists above and beyond the interest toward any one deviant sexual act (Able, Becker, Cunningham-Rathner, Mittelman, & Rouleau, 1988; Bowman & Schneekloth, 1999; Leitenberg, 1995; Walker, 1998). It is believed that the overwhelming dominance of the medical model has actually altered the way most researchers define sexual deviance. Instead of conceptualizing sexual deviance as a dimensional trait, the construct has been dissected and understood to be the deviant sexual interests that are listed within the DSMIV -TR (APA, 2000). Limiting the scope of a dimensional construct to one or even several categories of interest can

PAGE 19

19 severely damage its construct validity (Netemeyer, Beardon, & Sharma, 2003; Whitley, 1996). This misrepresentation of the construct disallows weakens the measure and disallows any meaningful relationships to be interpreted. For instance, if research were to detect a significant correlation between the sexual interest toward children and jogging that relationship could be reported to exist. However, no inferences could be made between those findings and the broader construct of sexual deviance. Sample Selection Bias Exclusively examining sex offenders also makes sexual deviance studies susceptible to sample selection biases. Basic research principles suggest caution when sampling from subsets of any targeted population (Whitley, 1996). Data from a population subset is only acceptable if they are representative of everyone that expresses sexual deviance. This not being th e case, the data is vulnerable to selection biases (Bernard, 1996). These biases have the potential to distort matters of statistical significance and to also cause illusory artifacts (Lane, 2009; Wang, 2008). When these biases occur it is not known whether significant results are due to the dependent variable or to the unique characteristics of the population sampled. For example, if someone wanted to do a study on wealth, he or she would need to determine a representative sample to survey. It might seem reasonable to collect data from members of a yacht club since only the very rich can afford these memberships. However, if the sample is not representative of all that are wealthy it may yield illusory artifacts. For example, data from this sample may suggest that that all wealthy individuals have a passion for boating. These may or may not true characteristics of the wealthy; however, they are likely characteristics of yacht club members. Even though it is a necessary condition for yacht club members to be wealthy, this yacht club member population may confound the study because they have unique yacht club member characteristics that distinguish them from the rest of the wealthy.

PAGE 20

20 Acknowledging this potential for bias, the findings from any sexual deviance research conducted exclusively on sex offender populations should be questioned. For instance, a strong relationship has been reported between psychopathy and sexual deviance (Dorr, 1998; Raymond, et al., 1999; Seto, Harris, Rice, & Barbaree, 2004). However, since these studies were all done on sex offenders, an alternative explanation might be that there was a relationship between psychopathy and the willingness to sexually offend. In other words, psychopaths may be disproportionally represented in sex offender samples because their lack of empathy and remorse makes them more willing to victimize people through their deviant sexual interests than other nonpsychopaths that possess similar levels of sexual deviance. Other studies examining sex offenders suggest that there is a relationship between frontal lobe damage and sexual deviance (Joyal, Black, & Dassylva, 2007). Again, because the research was only conducted on sex offenders it is equally plausible that the relationship exists between frontal lobe damage and the population subset. In other words, frontal lobe damaged individuals may be disproportionally represented in a sex offenders population because their lack of impulse control and behavioral constraint makes them more likely to sexually offend than other nonfrontal lobe damaged individuals with similar levels of sexual deviance. Trunca ted Range of Scores Another potential issue with sex offender populations is that these individuals will likely possess a truncated range of sexual deviance scores (i.e., only high scores). Whenever the extent of variation within a sample is less than the extent of variation within the represented population it potentially diminishes the strength of test statistics through a reduction of effect size estimators and statistical power (Greene, 2005; Grissom, 2005; Wang, 2009) These statistical problems are devastating to research that is already overwhelmed with noise from privacy concerns, participant embarrassment, and ethical constraints (Ward, L aws & Hudson, 2003; Lee, 1994).

PAGE 21

21 An example of how restricted range can influence effect sizes can be seen by examining the reduced r found between SAT scores and grade point averages (GPA) at the universities with the highest admissions standards, (e.g., Ivy League schools). In general, students SAT scores are highly correlated with their GPA scores (e.g., r = .70), (Camara & Echeternacht, 2000). However, when the range of admissions standards is restricted to only those schools with the highest admissi on standards, r is reduced. There are also extreme versions of this phenomenon that occur when dependent variables are truncated at extremes ends of the continuum. In these rare occurrences, the relationships to a dependent variable will show reverse correlations (Groseclose, 2006). Limited Data Another problem related to the practice of only examining sexual deviance in sex is that there is likely a limited range of deviant sexual interests being assessed through current sexual deviance measures. Over the years sexual deviance research has become synonymous with the study of sex offenders (Sexual Interest Card Sort Questionnaire Revised; Abel Assessment Screening Measure; Multiphasic Sex Inventory). Concurrently, the instruments constructed to measure sexual deviance have been specifically designed for sex offenders. Instead of assessing the entire range of deviant sexual interests, sexual deviance studies have exclusively focused on detecting the presence of paraphilic interests that are commonly fou nd in sex offenders. By only collecting data associated with only the most extreme forms of sexual deviance it has inadvertently limited the range of the dependent variable in these studies. This study set out to test four main hypotheses. The first hypothesis was that sexual deviance, as defined by a persons overall proclivity toward sexual deviance, would be present and could be adequately measured in a non-sex offender population. The second hypothesis was that social learning theory would be able to predict a significant amount of a persons proclivity

PAGE 22

22 toward sexual deviance, as it does with other forms of cultural deviance. The third hypothesis was that social learning theory would be able to better predict sexual deviance represented as combination of categories of deviant sexual interest versus sexual deviance represented by only one category of deviant sexual interest. The fourth hypothesis was that social learning theory would be able to better predict sexual deviance represented as combination of categories of deviant sexual interests that varied in their levels of perceived deviance, (e.g., mild, moderate, and extreme), versus a combination of categories that only represented extreme forms of perceived deviance. The dependent variables for this study were the Sexual Deviance Inventory (SDI) and the Sexual Interest Card Sort Questionnaire Revised (SICSQ -R; Abel & Becker, 2002). The SDI is a scale designed to measure a persons generalized proclivity toward sexual deviance. The SICSQ R is a previously established measure designed to assess for the more extreme types of sexual deviance that tend to be common in sex offenders. The SICSQ-R can provide an overall sexual deviance score as well scores for the following categories of deviant sexual interests: a) Adult homosexuality, b) Adult heterosexuality, c) Voyeurism, d) Exhibitionism, e) Frotteurism, f) Extrafamilial molestation of girls, g) Intrafamilial molestation of girls, h) Extrafamilial molestation of boys, i) Intrafamilial molestation of boys, j) Rape of adult females, k) Sadism, l) Masochism, m) Female gender identity, n) Male gender identity, and o) Transvestic fetishism. The independent variables for this study were six scales developed to represent factors from Akers and Burgess (1966) version of social learning theory. In past studies, this theory has successfully explained many forms of deviance including, not limited to, delinquency, alcoholism, smoking, drug use marijuana usage, and sexual aggression (Akers, 1998).

PAGE 23

23 CHAPTER 2 RE VIEW OF RELEVANT LITERATURE Overview Sexual deviance is defined as the interest in sexual behaviors that diverges from the accepted norm or what is regarded as normal by a community (www.definition-of.com, n.d.). Currently, most studies that examine sexual deviance gather data limited to the commission of sexual behaviors. In contrast, this study utilizes a more liberal conceptualization of sexual deviance as it assesses for the recollection of sexual interest associated with a variety of sexual expressi ons. The first section of this literature review section summarizes social learning theory along with the most common theories associated with sexual deviance. The second section provides a brief overview of the various expressions of sexual deviance (i.e., sexual behavior, sexual fantasy, pornography usage, and masturbation). The third section details several of the instruments and methods used to assess sexual deviance. Theories of Sexual Deviance As was mentioned in the introduction, most of the theories associated with sexual deviance currently lack empirical support. It is this void that has been the impetus for this project. Despite the absence of data, many theorists have published works describing the possible origins of sexual deviance. The se theories are summarized in this section. Others studying sexual deviance have published works that describe the differences between sex offenders and nonsex offenders. Theoretical explanations that determine behavioral commission or behavioral constr aint are left out of this section, (e.g., cognitive distortions, victim empathy, impulse control, entitlement, narcissism and psychopathy). While these theories do an adequate job explaining why a sexual offense may or may not occur, they are not as usefu l in explaining the etiology of sexual deviance.

PAGE 24

24 Social learning theory was the primary theory used to explore sexual deviance within this study. It was chosen because of its success in explaining other forms of social deviance as well as its ability to provide a more generalized explanation for sexual deviance. Many theorists offer explanations for individual paraphilias or specific sexual offenses. These specific theories were abandoned for a more general theory because of the presence of co -occurring deviant sexual interests found throughout sexual deviance literature. Although, these polymorphous perversities have been frequently detected there is very little theoretical work describing why unique forms of deviant sexual interest occur together ( Ab el, Becker, CunninghamRathner, Mittelman, & Rouleau, 1988). Social Learning Theory Social learning theory is a theory that has been helpful in describing many socially deviant behaviors (Table 2.3). It has also been shown to be a good predictor of spec ific attitudes related to sexual offenses such as sexual aggression. The basic premise of the theory is that behavior operates in the context of social structure, interactions, and situations. Within these contexts, the behavior in question either increa ses, decreases, or remains stable as a result of the exposure. Although Sutherland (1947) generally receives credit for much of the groundwork associated with social learning theory, over the years the theory been linked to several prominent researchers t hrough its revisions. The version of social learning theory most pertinent to the present study is the revision by Akers and Burgess (1966). They primarily changed how reinforcement variables were conceptualized. According to Akers and Burgess (1966), r einforcement referred both to social and nonsocial reinforcers, whereas Sutherlands (1947) definition of reinforcement applied only to social reinforcers. Akers and Burgess more liberal definition of reinforcement suggests that, aside from social influe nces, people may also learn through reinforcers such as internal

PAGE 25

25 physiological feedback and nonsocial feedback from the physical environment. This theoretical shift allows social learning theory account for social behaviors that have strong ties to physiological reinforcers. Examples of these are drug usage and its relationship with intoxication, sensation seeking and its relationship to endorphin release, and sexual behavior and its relationship to orgasm. The Akers and Burgess revision of Sutherlands (1947) classic theory of differential association consists of the following seven propositions: 1. Deviant behavior is learned according to the principles of operant conditioning. 2. Deviant behavior is learned both in nonsocial situations that are reinforcing or discriminating and through social interaction in which the behavior or the other person is reinforcing or discriminating for such behavior. 3. The principle part of the learning of deviant behavior occurs in those groups that comprise or control the individu als major source of reinforcements. 4. The learning of deviant behavior, including specific techniques, attitudes, and avoidance procedures, is a function of the effective and available reinforcers and the existing reinforcement contingencies. 5. The specific class of behavior learned and its frequency of occurrence are a function of the effective and available reinforcers, and the deviant or nondeviant direction of the norms, rules, and definitions which in the past have accompanied the reinforcement. 6. The proba bility that a person will commit deviant behavior is increased in the presence of normative statements, definitions, and verbalizations, which, in the process of differential reinforcement of such behavior over conforming behavior, have acquired discrimina tive value. 7. The strength of deviant behavior is a direct function of the amount, frequency, and probability of its reinforcement. The modalities of association with deviant patterns are important insofar as they affect the source, amount, and scheduling of reinforcement (as cited in Boeringer, 1992, pp. 21-21). The propositions listed above can be divided into the following four concepts: D ifferential association, differential reinforcement, modeling, and personal definitions.

PAGE 26

26 Differential association Differential association is the assimilation that occurs when people spend disproportionately greater amounts of time with a specific type of people or groups of peers. As people frequently interact with deviant persons or deviant groups they are often affected by the normative climate of those persons and consequently behave in a similar manner. In relation to sexual deviance, this concept might suggest that a person who differentially associates with those who commit, model, and support violations of sexual norms may, as a result, possess a greater degree of sexual deviance. Differential reinforcement Differential reinforcement is the learning process that occurs when people are rewarded or punished for certain behaviors (Akers, 1998). The reinforcement can be either social or physiological in nature. Social reinforcers include such rewards as praise and inclusions whereas physiological reinforcers include rewards such as an endorphin release or an orgasm. The amount, frequency, and strength of reinforcements all affect the probability of future behaviors. In relation to sexual deviance, a person who is differentially reinforced for violating sexual norms may possess a greater degree of interest in deviant sexual behaviors. Modeling The modeling effect refers to the tendency of people to mimic behaviors they see others commit (Bandura, 1977). For example, this concept suggests that if a person observes another person committing an action he or she will be more likely to commit the activity in the future. Additionally, if the viewer perceives that the person committing the action is being rewarded for the activity the influence will be even stronger. This further increases the likelihood of the action being mimicked in the future. In relation to sexual deviance, one example would be that a person exposed to models that violate sexual norms would potentially be more likely to possess

PAGE 27

27 greater degrees of sexual deviance. It is important to note that these models may include friends, acquaintances, strangers, public figures, or even media representations. Therefore, it may follow suit that individuals and activities displayed via pornography could potentially influence viewers through this same type of modeling effect. Definition Personal definitions are the personally held beliefs that an individual embraces. If a person possesses more positive than negative definitions associated with participating in an activity then he or she will be more likely to participate in that activity (Akers, 1998). With regards to sexual deviance, this concept could imply that a person who possesses learned definitions that are favorable toward committing sexually deviant acts would be more likely to be interested in deviant sexual behaviors. Each of these variables will be developed into a scale in order to better investigate how social learning affects sexual deviance. Based on previous studies, the measurement of these social learning theory concepts can be best accomplished by constructing and administering the following six s cales: Differential Association, Differential Reinforcement Peers (reinforcement by peers), Differential Reinforcement Physiology (reinforcement by physiological reactions), Reinforcement Balance (overall balance of reward versus punishment), Modeling, and Definition. Together, these scales measure the degree that a behavior is learned through socialization processes. Other Theories of Sexual Deviance A number of theories associated with sexual deviance are summarized in Table 2.4. Many of these theori es are based on basic psychological processes such as conditioning, modeling, and habituation. Others are based on more tenuous concepts, such as, imprinting and Freudian principles. What they all have in common is that they offer an explanation as to why a

PAGE 28

28 person would develop deviant sexual interest. This contrasts many theories developed within the field of sexual deviance that only relate to behavioral constraint. Although a majority of these theories stand on their own merit, some have been taken out of larger compilation of integrated theories used to explain sexual offending. For example, Wards (2001) Pathways Model of Child Sexual Abuse lists deviant sexual script as just one of the five pathways a child molester may progress through to sexually offend. This subtheory was extracted because it potentially offered etiological information. Another subtheory named emotional congruence was extracted from Finkelhors Precondition Theory. This subtheory was just one of the four underlying factors he used to explain child sexual abuse (Finkelhor, 1984). Again, this subtheory was singled out because offered it accounted for the underlying motivation for the sexual interest; whereas, the other factors were more associated with behavioral constraint. Expressions of Sexual Deviance In conjunction with this studys definition of sexual deviance, literature related to each of the four expressions of sexual interest was reviewed. In reality, these expressions should not be viewed as discrete categories as they often occur in combination with one other. However, for simplicitys sake, the literature associated to each of the expressions will be kept separate and in their own categories. Literature related to behavioral expressions of sexual deviance was found to be the most prevalent. This was not surprising since the focus of most sexual deviance studies has been on individuals that commit the behaviors associated with sexual deviance (i.e., sex offenders). Sexual deviance expressed through sexual behaviors remains a topic of interest within our society because of the potential for victimization and legal proscriptions. The second largest category of literature related to sexual deviance as it has been expressed through pornography. Similar to the behavioral literature, it is common for the pornography literature to

PAGE 29

29 be litigious in nature. Articles citing trends in the prevalence of pornography along with the impact this has had on society have become increasingly common. Research associated with sexual deviance as it is expressed through masturbation and fantasy was relatively sparse. This is assumed to be due to the private nature of these expressions. Sexual Behavior As was mentioned previously, one of the current problems with sexual deviance research is that it has been too focused on sexual offenders. This is not meant to imply that sex offenders should be excluded from sexual deviance research. Rather, it suggests that sexual deviance research should include, but not be limited to, the study of sex offenders. This first section provides a brief summary of prohibited behaviors that have been associated with sexual deviance. The next several sections will describe the systems that control against these behaviors. The last section will sum marize several forms of permissible forms of sexual deviance that exist in current society. Prohibited The four primary ways that society controls prohibited forms of sexual deviance are through the criminalization, psychiatric classification, sex offender treatment programs, and sex offender classification. Society justifies these extensive and comprehensive efforts because the deleterious effect of sexual victimization is often long lasting and severe. Studies have shown that the effect of childhood sexual abuse includes increases in health problems, substance abuse, teen promiscuity and pregnancy, and crime (Noll, Tricket, & Putnam, 2003; Paolucci, Genuis, & Violato, 2001). Examining the effects of pedophilia alone, it is estimated that 20% of Americans are sexually abused before they reach the age of 18 (Felitti et al., 1998) and it is that there are approximately 39 million survivors of this type of abuse (Abel et al., 1987). Nationwide, these

PAGE 30

30 offenses come at a great cost to the correctional system, the economy, not to mention the personal well being of the victims. Sex c rimes: Criminalizing sexual behaviors is one way that society attempts to control unwanted sexual behaviors. It is job of the criminal justice system to detect and deter the occurr ence of sex crimes. These highly sanctioned behaviors are generally associated with one or more of the following: violence, age of consent, victimization, sexual harassment, exposure, incest, and cultural taboos. Although sexual crimes and their sentence s vary from state to state, a summary of some of the more common sex crimes are presented in Table 2.1 (Holmes & Holmes, 2008). Some, but not all, sexual crimes are also considered to be mental illnesses. For example, most individuals charged with child molestation would also be the diagnostic criteria for the mental disorder pedophilia. An example of a sex crime that does not generally receive a psychiatric diagnosis is rape. Paraphilias: The psychiatric classification is another way that society regulates deviant sexual behaviors. The DSMIV TR describes the essential feature of the paraphilias as a recurrent, intense sexually arousing fantasies, sexual urges, or behaviors generally involving 1) nonhuman objects, 2) the suffering or humiliation of oneself or ones partner, or 3) children or other non consenting persons, that occur over a period of at least 6 months (American Psychiatric Association, 2000, pp. Page 566). noitces suoiverp eht ni debircsed saw sA, si ereht sailihparap dna semirc lauxes neewteb palrevo emos. eviecer taht slaudividni eht fo tsoM yllanimirc degrahc eb nac taht roivaheb a dettimmoc evah osla sesongaid cilihparap. roF example, ma ny individuals diagnosed with exhibitionism have committed behaviors that, if caught, would also be charged with indecent exposure. However, the paraphilias that do not place individuals at criminal risk, such as Transvestic Fetishism and Sexual Masochism do not

PAGE 31

31 have corresponding criminal charges. For a complete list of the paraphilias found in the DSMIV -TR please see Table 2.2 (APA, 2000). Incarceration, treatment and registration: Mandatory psychiatric treatment assists society in curbing unwanted sexual behaviors. In todays society sexual offenses are conceptualized as hybrids of criminal behavior and psychopathology. This conceptualization allows society to incarcerate the offender for violating sexual laws while, at the same time, force treatm ent in an attempt to extinguish the mental disorder. Barbaree and Marshall (1998) see the conflict that arises when an offense is understood as a criminal infraction on one hand and a mental disorder on the other. They state that it frequently evokes both feelings of contempt and empathy for the offender in acknowledging the harm that befell the victim while, at the same time, consenting that the behavior may have been the result of issues they were unable to control or understand. Many treatments meth ods have been developed for sex offender treatment programs. Barbaree and Seto (1997) categorize these treatments into general psychotherapy, organic treatments, behavioral therapies, and relapse prevention. General therapies have been used but have been found to be relatively ineffective in reducing recidivism. Organic treatments focus on eliminating sexual urges and deter deviant sexual arousal in the process. Compliance issues and ethical concerns remain major obstacles to these treatments. Behavioral therapies focus on modifying maladaptive sexual response patterns and reorienting the offender to more healthy preferences. These include conditioning treatments such as masturbatory satiation, covert sensitization, and orgasmic reconditioning. Recidivism studies shows mixed results for the effectiveness of sex offender treatment. The Sex Offender Treatment and Evaluation Project (SOTEP) found no differences between

PAGE 32

32 treated and untreated groups in sexual or violent reoffending over an eight year follow-up period. However, when clients who had successfully completed treatment goals were compared with those who had not, the treated groups demonstrated significantly lower recidivism rates (Marques, Wiederanders, Day, Nelson, & van Ommeren, 2005). Another study with similar findings concluded that non-compliant sex offenders had a higher recidivism rate (17%) than those who complied with psychotherapy (10%) (Hanson et al., 2002). First time offenders seemed to respond even better to treatment as 9% of the treated offenders recidivated compared to 27% of the untreated offenders (Nicholaichuk, Gordon, Gu, & Wong, 2000). Therefore, one might assume that at least offenders that choose to participate in therapy will benefit from the process. The fourth way society controls deviant sexual behaviors is by classifying and labeling individuals as a sex offender. On a given day in 1994, there were 234,000 convicted sex offenders under the care of correctional facilities. The prevalence of actual sex offenders is difficult to ascertain because of chronic under reporting and accountability issues related to individuals that commit multiple offenses. Once individuals have been convicted of a sexual crime many limitations are placed on them regarding where they are a llowed to live, work and visit. Locations that are frequently prohibited are churches, schools, parks and other public venues, apartments, malls, stores, and shopping centers. Behavioral restrictions are also placed on being in the presence of minors, owning toys or other items of interest to minors, or using the Internet (Bureau of Justice Statistics, 2007). Sex offenders are also required to register as a sex offender in the national sex offender database. The Dru Sjodin National Sex Offender Publ ic Website (NSOPW) gives the general public access to the addresses and pictures of registered sex offenders. Using this system,

PAGE 33

33 citizens are provided a convenient way to detect the presence of convicted sex offenders living in their neighborhoods. Some s tudies note the positive effects of registration laws, including motivation to prevent recidivism and increased honesty with friends and family (Levenson, & Cotter, 2005). Others have found profound negative consequences for the offenders such as employme nt difficulties, relationship problems, harassment, stigmatization, and constant vulnerability (Tewksbury, & Lees, 2006). Recently, civil liberty issues associated with the banishment of sex offenders from their homes (Jacobs, 2009) and the ex post facto registration of offenders (Hooper, 2008) have been the focus of legal disputes. Non Prohibited In contrast to behaviors that frequently result in felony convictions, there are some forms of sexual deviance where all parties are involved consenting adults and the acts are legally permissible. It appears that for these behaviors, the inappropriateness of the activity has not reached the extent that institutional control is mandated. Throughout 1990s HBO produced a documentary series titled Real Sex that de tailed many fringe type sexual behaviors. The show brought to light several sexual subcultures and the people that participated in them. Specific episodes featured legalized prostitution, fetish clubs, swinger lifestyles, nudist and sex resorts, and sex toy completed with training seminars. In the Internet Age, these alternative lifestyles continue to thrive as clubs and organizations can be accessed through simple web searches. Although the classifieds have always existed as a way for individuals to anonymously match interests with one another, several internet dating sites have taken this activity to a new level. Adult friend finder is one of the most successful sex matching sites with 75,000 new users registering to listing erotic sexual preferences, such as, gay, straight, bisexual, couples, group sex, etc (Stein, 2007). In 2007, Adult Friend Finder reported earnings over $200,000 in revenue and generating twice the traffic as rival dating sites Yahoo Personals and Match.com. These

PAGE 34

34 websites continue to grow in popularity, offering people the ability to match sexual desires, sometimes deviant sexual desires with individuals possessing similar interests. In 2007, Penthouse bought Adult Friend Finder along with several of its sister websites for $500,000,000 in cash and securities (Kettner, 2007). Sexual Fantasy A sexual fantasy is any mental imagery that is sexually arousing or erotic to the individual. These types of fantasies are a relatively common experience as approximately 95% of men and women admit to experiencing sexual fantasies throughout the day both in the absence of and during sexual activities (Leitenberg & Henning, 1996). It is generally hypothesized that the presence of sexual fantasy precedes acting out most sexual behaviors. This is assumed to be true for most types of sexual scenarios, including deviant sexual behaviors (Ward, & Beech, 2006). McGuire et al. (1965) found that, in a group of 52 men who had committed deviant sexual acts, 75% admitted that their most common sexual fantasy pertained to their paraphilic behavior. Evans (1968) found that 50% of his sample of exhibitionists admitted to regularly masturbating to fantasies of exposure. Finally, 52% of 129 child molesters admitted having sexual fantasies about children (Marshall, Barbaree, & Eccles, 1991). Deviant sexual fantasies are not reserved for sex -offenders alone. Briere and Runtz (1989) surveyed college students and found that a total 21% of males reported having a sexual attraction to small children attracted them sexually. The survey revealed that 9% of the participants described having had sexual fantasies involving children, 5% reported having had masturbated to those fantasies, and 7% reported that they would likely have sex with a child if they were guaranteed not to get caught (Briere & Runtz, 1989). However, none of these individuals reported that that they had molested children. A study by Crepault and Couture (1980) found that 61% of adult males had experienced a sexual fantasy where they sexually

PAGE 35

35 initiated a young girl. Another study found that 17% of males admitted to recently experiencing sexual thoughts about having sex with girls under age 15, as well as 5% of males with girls under age 12 (Templeman & Stinnett, 1991). Rape submission fantasies have been found to be especially prevalent amongst woman. A study by Herold (1988) reported that 51 % of females had experienced fantasies that involved being forced to submit sexually. A study by Fisher (1973) found that 20% of the fantasies that woman reported having had experienced during intercourse were about either being raped or humiliated. Two other studies had found that 29% of the female in their samples had experienced rape submission fantasies ( Talbot, Beech, & Vaughn, 1980; Kanin, 1982). A rndt et al. (1985) found that 22% of their female sample had the fantasy I m made to suffer before a man will satisfy me sexually and 30% had the fantasy I m a slave who must obey a mans every wish. Additionally, Knafo and Jaffe (1984) found that, of the 21 fantasies that were listed in their study the fantasy of being overpowered was reported to be the one that occur red mos t frequently during intercourse. In contrast to the disproportionate amount of women that have rape submission fantasies, more men tend to have fantasies about forcing someone to have sex. For example, Hunt (1974) found that 13% of men compared to only 3% of women had the fantasy of forcing someone to have sex. Sue (1979) reported figures of 24% and 16% for men and women, respec tively; Miller and Simon (1980); 24% and 6%; Arndt et al. (1985), 39% and 25%; and Person et al. (1989), 31% vs. 5% (as cited in Leitenberg & Henning, 1995) Two other studies that had examined only men reported that 33% and 54% of men had fantasies of fo rcing sex on women (Crepault & Couture, 1980; Gredlinger & Byrne, 1987).

PAGE 36

36 Fantasies about group sex also tend to be relatively common amongst men and women. For example, Hunt (1974) reported that 33% of men had fantasies about having sex with multiple partners of the opposite sex, in comparison with 18% of women. Wilson (1987) found that 31 % of the men and 15% of the women reported that fantasizing about group sex was their favorite type of sexual fantasy Similar findings associated with group sex fanta sies were reported in a number of other studies. In each of the following studies men reported having had experienced more sexual fantasies associated with group sex: Davidson (1985), 42% and 17%; Hessellund (1976), 37% and 7%; Person et al. (1989), 52% and 27%; and Sue (1979), 19% and 14%. The results of these studies suggest that deviant sexual fantasies are prevalent, and in some cases, experiencing deviant sexual fantasies may be the norm. Although higher rates of deviant sexual fantasies may be reported amongst sexual offenders, it is clear that this experience is not reserved to that population (Alwyn, Reddin, & Burke, 2005). It is also important to note that although individuals report experiencing sexual fantasies associated with sexual offenses th at it does not necessarily lead to the commission of that behavior (Leitenberg & Henning, 1996). Pornography In recent years, the world has seen the proliferation of all types of pornography. The United States pornography industry continues to thrive with annual sales growing from eight billion dollars in 1996 to over twelve billion dollars in 2005. Over a ten year span, the annual production of adult videos in the United States has increased from 5000 to 13,000 (AVN, 2006). By comparison, the major Hollywood studios released 549 new titles in 2005 (MPAA, 2006). Although consumers may purchase pornographic materials through a multitude of media outlets (e.g., Internet, cable, video on demand (VOD), telephone, and magazines), it is the videotapes

PAGE 37

37 (VHS) an d digital video discs (DVD) that remain the most popular. In 2005, 950 million adult videos were rented annually (Stack, Wasserman, & Kern, 2004). In a 2002 Kinsey Institute Survey, 97% of individuals admitted to having viewed pornography having ever viewed pornography while 77% admitted to viewing pornography regularly. It is no coincidence that pornography has become exponentially more popular since the advent of the Internet. Coopers 2003 Online Sexual Activity Study revealed that 69% of participants rated online erotica as being their preferred sexual activity (as cited in Danebeck, Cooper, & Mansson, 2005). Even in 1998, Cooper seemed to recognize the Internets erotic potential as he described the Internet as possessing a Triple A engine of facto rs that would soon overwhelm human sexuality. He suggested that the high levels of Accessibility, Affordability, and Anonymity would drive more people to use pornography than any other time in history (Cooper, 1998). Perhaps another byproduct of the aforementioned Anonymity has been the concurrent rise in pornography associated with deviant sexual interests. A 2006 content analysis associated the previous years best selling pornography revealed a variety of hardcore scenes (Wosnitzer, Scharrer Bridges, Sun, & Liber man, 2006). The most frequent sexual act portrayed in the sample was femaleto -male oral sex (90.1%, n = 274). Vaginal intercourse was the second most frequent sexual act (86.2%, n = 262). Anal sex appeared in 55.9% ( n = 356) of the scenes, ass to mouth sequences in 41.1% ( n = 125), simultaneous vaginal and anal penetration in 18.1% ( n = 55), and double anal penetration in 2.0% ( n = 6). Portrayals of homosexuality occurred only for women, with femaleto female oral sex taking place in 22.7% ( n = 69) of the scenes. The researchers detected an unequivocal rise of aggression and female degradation in pornography in both verbal and physical forms, compared to previous studies.

PAGE 38

38 When pornographic material moves too far in the direction of deviance, obscenity laws are executed. Although the First Amendment protects most pornographic publications, there are certain images and representations that are said to offend the prevalent morality of the time and are consequently are deemed obscene. Once something has been determined to be obscene, the production, sale, or display of those depictions would, henceforth, be criminal. In the landmark case of Miller v. California (1976), the following three pronged approach was used to determine the ob scenity ruling: 1. Whether the average person, applying contemporary community standards, would find that the work, taken as a whole, appeals to the prurient interest, 2. Whether the work depicts/describes, in a patently offensive way, sexual conduct or excretory functions specifically defined by applicable state law, and 3. Whether the work, taken as a whole, lacks serious literary artistic political or scientific value (Miller v. California, 1976). If all three of the previous conditions are met the material is ruled to be obscene. An interesting characteristic of obscenity rulings are that, as described in Prong 1, they potentially differ based on the community that they are judged. For example, material that is ruled obscene in California may not be ruled obscene in Alabama. Another type of pornography that is prohibited is child pornography. Although most child pornography would also be considered obscene, the child pornography determinations are made using a different set of principles. The restrictions are actually less ambiguous than obscenity rulings as they have more to do with the laws associated with child labor laws than community standards (Wells, Finkelhor, Wolak, & Mitchell, 2007). Because of the way the way child pornography laws are framed a loophole exists. Simulated and virtual child pornography is currently legal under those child pornography laws because no children were used in the production of the images, (e.g., hentai). These renditions are currently being tested in the United

PAGE 39

39 Sta tes courts. The arguments against these depictions are that viewing the subject matter somehow primes individuals to sexually offend. There currently is some evidence that viewing renditions of child pornography makes it more likely for individuals to as sociate sex or sexuality with sexually neutral images of children (Paul, & Linsz, 2008). However, the authors caution that these cognitive effects do not imply an increased likelihood to offend or the increased acceptance of either actual child pornography or sexual interactions with children Masturbation Masturbation refers to sexual stimulation, especially of ones own genitals (self masturbation), often to the point of orgasm The stimulation can be performed manually, by other types of bodily contact (short of sexual intercourse), by use of objects or tools, or by some combination of these methods. The prevalence of masturbation appears to have stayed relatively constant over time. Kinseys 1948 and 1953 studies, based on detailed interviews with white American adults, found that 92 percent of men and 62 percent of women reported that they had masturbated at some point during their life. Gender discrepancies also exist in terms of frequency as 55 percent of men reported masturbating on a regular basis in comparison to 38 percent of women. Twenty-five (25) percent of males reported masturbating frequently in comparison to 10 percent of women (Janus, & Janus, 1993). As individuals masturbate they commonly engage in sexual fantasy and/or view pornographic images. Unfortunately, there have been few studies examining masturbation practices in the context of specific sexual fantasies and/or specific types of pornography. In general, fantasizing during masturbation is reported to be a fairly common. Acr oss 13 studies summarized by Leitenberg, & Henning (1995), an average of 85.9% of men and 68.8% of women stated that they had sexual fantasies while masturbating. Also, pornography exposure during masturbating is also quite common. Just accounting for Internet pornography, a 2000

PAGE 40

40 MSNBC poll reported that 75% have admitted to having masturbated while viewing materials online (as cited in Danebeck, Cooper, & Mansson 2005). In spite of the obvious connection between masturbation and pornography, there is st ill almost no literature examining the content of peoples masturbation sessions. One reason why this content data would be important is that sexual deviance, like other sexual preferences, can be influenced or conditioned at the hands of masturbation. D eviant scenes or fantasies that are repeatedly paired with pleasurable arousal and orgasm during masturbation will likely become masturbatory content preferences in the future (ODonohue & Plaud, 1994). Despite this notable gap in research, conclusions can be drawn by examining several pieces of information. The Internal Revenue Service gives a conservative account of U.S. pornography revenues in 2006 as 13.3 billion dollars. This figure is larger than the combined revenues of ABC, CBS, and NBC. As was predicted by Cooper (1998), the Triple As (Accessibility, Affordibility, and Anonymity) associated with todays pornography have created exponential growth in adult entertainment industry. Since 72% of individuals reported that the reason they accessed pornography was to masturbate (Cooper, 1998) and the most popular pornographic content exhibits extremely deviant scenarios (Wosnitzer and Bridges, 2006), it might be safe to conclude that most people are masturbating to deviant pornographic scenarios. Assessments Measures associated with sexual deviance are notorious for having reliability and validity issues (Becker & Levine, 1986; Laws & ODonohue, 1997). The tendency for individuals to supply misinformation may be partially due to the inherently private nature of sex and the fear of penalization. It is likely a person would be hesitant to forfeit sexual secrets even to their closest friends let alone a stranger that ask for it during a research study Participants involved in sexual deviance research may also feel that they will suffer deleterious consequences if they disclose the

PAGE 41

41 truth about past or present experiences. These cautions are not unfounded. For example, according to the American Psychological Associations ethical guidelines, if a re search participant were to disclos e a sexual offense that they had committed on a child, that information would need to be reported to the authorities. Also, a similar disclosure from an incarcerated sex offender could either provoke additional investigations or lengthen that offenders civil confinement at a sex offender treatment program. In regards to observational data, assessments experience a double edged sword. Collecting sexual behavior data as it naturally occurs is not ethically possible. The private nature of sexual behavior makes it so that most intimate sexual expressions occur behind closed doors. Because of this, researchers do not have the ability to observe sexual behaviors in their natural environments. Any attempt at collecting data from unknowing participants would be an extreme invasion of personal privacy and against the ethical principles described in the APA code of ethics (APA, 2002). Conversely, collecting data in a laboratory setting is likely to alter a research participants expression of sexual arousal. For example, the penile plethysmograph (PPG) is referred to by many as the gold standard to which all other sexual interest measures should be compared (Quinsey & Lalumiere, 2002). This being said, PPG procedures have been found to produce extreme anxiety in subjects which consequently affect the ability of those individuals to display sexual arousal. Feigning non-interest is also an issue for plethysmograph assessments as it is easy to inhibit sexual arousal through self -monitoring and mental imagery (Eccl es, Marshall, & Barbaree, 1994). According to Sti nson and Becker (2008), assessments of sexual interest can be grouped into three primary categories: a) historical, b), self -report and c) physiological. Descriptions of the most widely used assessments are provided in the following paragraphs.

PAGE 42

42 Historical Assessments Historical assessments focus on collecting information that has occurred in the past. Due to the private nature of sexual experiences, information must be gleaned from instances where the individual has been caught engaging in prohibited activities, individuals that have lived in close proximity to the person in question, or self reports. Although many established self report measures do focus on collecting historical information, the established measures will be summarized in the section covering assessment instruments. Self reports Self report measures collect information past sexual behaviors, the content of current and past sexual fantasies, sexual inter ests or preferences, and any other relevant information that might establish clear patterns of sexual interest. Criminal records When examining the sexual interest of sex offenders, it is considered prudent to gather historical data from a rrest s, correct ional reports, parole and probation officers, as well as court documents. This is because self -report information from sex offenders is often unreliable. Information from investigations, and/or convictions can offer support for approximating specific sex ual interests, progression of deviant sexual interests, range of diverse sexual interests, and sexual interest frequency. Unfortunately, this information may only represent the small sample of i ndividuals sexual behavior patterns that is known to already known by others. Collateral interviews Valuable information may also be attained from parents, spouses, and even ex-spouses of sexual offenders (McGrath & Hoke, 2001). Although these individuals may sometimes be privy to deviant sexual interests it is c ommon to keep these inappropriate desires covert and discrete. For example, people may limit expressing sexual interests to those behaviors they feel will not be

PAGE 43

43 detected by others. When conducting collateral interviews it is important to ask for cues th at may have been subtle in nature. For example, it may be beneficial to ask the interviewee if he or she had ever found sexual pornography links on the home computer. It may also be useful to inquire about any age inappropriate relationships that the per son may fostered in the past. Subtle clues like these may support other findings but should not be considered sufficient to determine specific sexual interests. Physiological Assessments These instruments focus on gathering physiological information that can be used to detect the human responses associated with sexual arousal, attention, and anxiety. Abel assessment for sexual interest (AASI) During the AASI the participant is shown slides of individuals of different ages and genders engaged in a vari ety of different sexual scenarios. As the participant observes each slide, he or she is asked to indicate his or her preference for the scenario. Although this may seem like a standard self report assessment, what is actually being recorded is the visual reaction time or the amount of time it takes the participant transition to the next slide. The longer the participant gazes at a slide the more he or she is assumed to be interested in that particular sexual scenario Although this measure has adequate sp ecificity (77%98%), sensitivity (76%-91%), and efficiency (77.5%-96.9%), it is relatively easy to dupe (Stinson & Becker, 2008). Vaginal and penile plethysmographs (PPG) The PPG is the most widely used physiological assessment associated with sexual devia nce. The assessment measures penile or vaginal tumescence associated with changes in sexual arousal. During an assessment the participant is exposed to a series of slides and recorded dialogue depicting both appropriate sexual scenarios and inappropriate sexual scenarios. The slides transition through different ages, levels of coercion, and types of paraphilias. If a person

PAGE 44

44 experiences physiologically arousal as a specific slide is shown then he or she is assumed to possess sexual interest toward that s exual category. One of the benefits associated with the plethysmograph is that there are standards associated with the assessment and it is codified into the guidelines of the Association for the Treatment of Sexual Abusers. The PPG also has several shortcomings. One of these is that the assessments are only found to be useful differentiating between positive responses. In other words, they are not able to differentiate between offenders and nonoffenders (Eccles, Marshall, & Barbaree, 1994). Another li mitation of the PPG is related to its validity and reliability. These issues make it difficult to use within forensic populations. Murphy and Peters (1992) conclude about the forensic use of the penile plethysmograph: The results of the studies using erection data suggest that, although group differences are reliably found, the ability to classify an individual would produce error rates that would not be appropriate for the trial situation. In addition, in cases of incest or when patients deny charges, one would even expect to find either no responding in the laboratory or a normal response pattern. Further, it is clear that individuals can fake their responses and the absence of significant responding is basically meaningless in terms of a clinical interpretation. Like the MMPI literature, we find the conditions under which the test has been validated do not meet legal requirements. (pp. 32-33) Polygraph Forensic populations, such as sex offenders, often fear legal recourse associated with personally damagi ng disclosures and because of this they have a vested interest in providing information that portrays them in a positive light. In these instances, when the veracity of the information may be called into question, a polygraph assessments can be administered to substantiate self -reports information. These polygraphs use heart rate, respiration rate, and galvanic skin response to determine the truthfulness of statements.

PAGE 45

45 Scales, Inventories and Questionnaires These instruments have been designed to assess for the presence of deviant sexual interests in sex offenders. The data collected through these measures includes historical information about behaviors, sexual preferences and sexual fantasies. In addition, some of the measures also focus on sex offender risk assessment information. Sexual i nterest c ard sort q uestionnaire r evised (SICSQ R) The SI CSQ-R is a 75 item, selfreport measure that examines sexual interest within 15 discrete categories of sexual interest. Categories measured are heterosexuality, homosexuality, extrafamilial molestation of girls, extrafamilial molestation of boys, intrafamilial molestation of girls, intrafamilial molestation of boys, voyeurism, frotteurism, exhibitionism, sadism, masochism, rape, transvestic fetishism, male ge nder identification, and female gender identification. Participants are instructed to rate their interest toward 75 sexual activities on a seven point Likert scale, from extreme sexual repulsion to extreme interest. Results from this data are used to evaluate interests and behaviors of participants and to recommend treatment alternatives (Holland et al., 2000). Multiphasic sex inventory second edition The Multiphasic Sex Inventory Second Edition is a self report assessment instrument with scales measuring a variety of constructs related to sex offending behavior, including cognitive distortions, sexually deviant behavior, substance use, antisocial attitudes, and psychological characteristics common to individuals who have committed sexual offenses (Nich ols & Molinder, 1984). The developers of this instrument have described a number of factors that support the content validity of the scales on this instrument, as well as their ability to discriminate between different types of sexual behavior and interest. The information obtained from the Multiphasic Sex Inventory Second Edition provides the current state of a participants

PAGE 46

46 sexual interest and has been sh own to be a good predictor of progress through therapy. It also has shown relatively moderate level s of internal consistency as its alphas range from .50 and .90 (Kalichman, Henderson, Shealy, & Dwyer, 1992). Unfortunately, the degree of face validity associated with the scale is concerning and some worry about contamination (Hudson & Ward, 1997). Clar ke sexual history scale for males The Clarke Sexual History Scale for Males examines deviant sexual behaviors and preferences in males. The scale consists of 189 questions that inquire about the frequency and age of occurrence for a wide of sexual behaviors including paraphilias. It is designed to evaluate a sex offender s sexual preference profile as well as evaluating their potential for sexually conventional behavior. It is the most researched instrument in the field and has been able to differentiat e sex offenders from non -sex offenders (Langevin et al., 1990). Wilson s exual fantasy q uestionnaire The Wilson Sexual Fantasy Questionnaire is a 40 item self report questionnaire that assesses 4 types of sexual fantasies: Exploratory, Intimate, Impersonal, and Sadomasochistic. The purpose of this measure is to assess the quantity of sexual desires, preferences, and sexual activities of participants in clinical research settings (Wilson, 1978). Psychopathy c heck l ist r evised Although it is not directly a measure of sexual interest or behavior, it is often correlated or related to other measures of sexual interest and recidivism. The Psychopathy Check List Revised is an assessment of psychopathy, which includes such features as antisocial behavior, deceit fulness, superficiality, impulsivity, irresponsibility, and lack of remorse. Extensive

PAGE 47

47 research has demonstrated the content and predictive validity and reliability of the items on this instrument.

PAGE 48

48 Ta ble 2 1. Theories of sexual deviance Learning theor y /conditioning Individuals learn to experience deviant sexual arousal through the association sexual responses with deviant stimuli. Maintenance of these preferences is reinforced through operant conditioning principles (Laws & Marshall, 1993). Habituat ion An individuals level of deviant sexual preference tends to escalate because of habituation effects associated with his or her sexual arousal (Palk & OGorman, 2004; Walker, 1998). Emotional incongruence The inability of adults to form emotionally f ulfilling relationships with adults leads them to embark on these endeavors with children. The sexual interest is believed to be a consequence of this emotional attachment (Ward, Hudson, & Marshall, 1996) Models of sexual abuse Those that come from fami lies exhibiting sexually assaultive or sexually deviant interests learn that these are appropriate forms of sexual expression (Prentky et al., 1989). Power & control Those that feel threatened will seek ways to exert power over the intimidating party. Males that feel threatened by females may choose to exert their structural power by inflicting violence and/or rape (OBrien, 1991). Deviant sexual scripts Assumed to result from early abuse or a sexualized childhood, individuals with deviant sexual scri pts will often confuse relationships with sexual closeness and, when possessed in conjunction with psychological flaws, often exhibit pronounced pedophilic tendencies (Ward & Hudson, 2001). Psychodynamic Freudian theory assumes that forces the primal de sires of the id and the moral authority of the superego are in balance a person expresses normal sexual urges. However, when the id overwhelms the superego a person will select a fetish based on the ids infantile sexual responses (Freud, 1962; Barnard, 19 89). Developmental impairment Those that are socially incompetent are unable to satisfy sexual and emotional needs with appropriate partners. Consequently, they turn to less threatening partners, (i.e., children ), and inappropriate methods, (i.e. rape, voyeurism, and frotteurism ) to satisfy these unmet needs (Katz, 1990). Natural selection/ Evolution Sexually coercive males ( e.g., rapists ) hold characteristics that are evolutionarily advantageous. These are fulfillment of a dominance role, adhering to a short term mating strategy, and possessing hostile masculinity (Malamuth, 1986). Imprinting Humans experience a critical period in their adult youth when sexual preferences may be imprinted. Exposure to deviant stimuli during this sensitive time period influences subsequent sexual arousal patterns (DUdine, 1990).

PAGE 49

49 Tab le 2 2. Comparison of R square for several studies examining the relationship between social learning theory and deviant behavior (Akers, 1998) Deviant Behavior Study R Square N Marijuana use Akers et al. (1979) .68 948 Alcohol use Akers et al. (1979) .54 1764 Smoking Spear and Akers (1988) .54 2156 Sexual Aggression Boeringer et al. (1991) .53 262 Table 2 -3. List of common sex crimes (Sex crimes, n.d.) Sexual assault Lust murder Child sexual abuse Statutory rape Spousal rape Obscenity Human trafficking Frotteurism rubbing one s self against a non -consenting stranger in public Exhibitionism and voyeurism, if deliberate and non consensual, called indec ent exposure and peeping tom respectively in this context. Incest between close relatives Telephone scatologia making obscene telephone calls Bestiality Sex with animals Sexual harassment Sexual acts by people in a position of trust (such a s teachers, doctors and police officers), towards any person they are involved with professionally. Extra marital relations (also Zina and Polygamy). Public order crimes are crimes that interrupt the flow of daily life and business according to local c ommunity standards. Public order crimes include paraphilia (deviancies). Homosexuality Various paraphilias and sexual fetishes such as transvestitism Pornography Prostitution and/or pimping Ownership of vibrators and other sex toys Public urination Stealing underwear, sometimes regarded as more serious when done in a sexual context.

PAGE 50

50 Table 2 4. List of paraphilias (DSM IV -TR, 2000) Pedophilia The interest in sexual activity with a prepubescent child or children (generally age 13 or younger ). Fetishism The interest in non human objects to enhance sexual arousal Transvestic Fetishism Associated with heterosexual males that dress in female clothes to enhance sexual arousal Exhibitionism The sexual interest in the exposure of ones genital s to an unsuspecting stranger Frotteurism The sexual interest in touching and rubbing against a non consenting person Sexual Sadism The sexual interest in acts (real, not simulated) in which the psychological or physical suffering (including humiliation) of the victim is sexually exciting to the person). Sexual Masochism The experience of being beaten, bound or tortured enhances sexual arousal Voyeurism The sexual interest in observing an unknowing and non consenting person, usually unclothed and/or engaged in sexual activity, to produce sexual excitement

PAGE 51

51 CHAPTER 3 METHODS AND MATERIAL S Overview This study can be divided into the scale construction phase and the survey administration phase. During the scale construction phase, a sexual deviance measure was created named the Sexual Deviance Inventory (SDI). The following six social learning measures were also developed to assess factors associated with social learning theory: Differential Association Scale (DAS), the Differential Reinforcement Pe er Scale (DRPe), the Differential Reinforcement Physiology Scale (DRPh), the Reinforcement Balance Scale (RBS) the Modeling Scale (MOD), and the Definitions Scale (DEF). These measures along with several others were administered through an online survey. Phase 1 Social Learning Theory Six scales were created to measure social learning theory in the context of sexual deviance. These measures were named D ifferential Association, Differential Reinforcement Peers, Differential Reinforcement Physiology, R einforcement Balance, Definition, and Modeling. Construct items Using measures from past studies as templates, the social learning theory measures were developed to assess social learning theory. To be effective, the items had to relate to both the essenc e of the social learning theory factor and the dependent variable of sexual deviance (Akers, 1985; Akers, 1998; and Akers & Cochran, 1985; Rossi, Waite, Bose, & Berk, 1974). These scale items were be reviewed by Dr. Ronald Akers to assess and confirm conc eptual accuracy.

PAGE 52

52 Eliminate items Items were eliminated from the scale based on their contribution to the internal consistency of the measure. Standard deviations were also examined and items with the highest standard deviations were reexamined for clarity and conceptual accuracy. Items that were not congruent with the social learning theory factors were also eliminated. Develop s coring method Referencing similar studies that had developed their own social learning theory measures, the scale scores wer e calculated by summing the scale values (Akers, 1998; Boeringer, 1992). The variances of those scale values were then examined to determine the necessity of any transformations to achieve normality. Sexual Deviance Inventory The purpose of building the SDI was to construct a scale able to measure sexual deviance within a non-sex offender population. Up until now, sex offenders have been the target of most sexual deviance studies and therefore these measures have been developed to assess the high levels of sexual deviance found in those populations. Because these past sexual deviance measures assessed only extreme levels of sexual deviance, (i.e., paraphilias and sex crimes), it is believed that they would not be sensitive enough to detect sexual deviance within a nonsex offender population. The design of the SDI needed to account for the increased variance in sexual deviance that was hypothesized to occur in non-sex offender populations. It was also hypothesized that, when compared to individuals with lower levels of sexual deviance, individuals with a higher levels of sexual deviance would a) endorse more sexually deviant items b) endorse greater frequencies of sexual interest for sexually deviant items, and c) endorse items with higher ratings of s exual deviance. Based on these assumptions, it was necessary for the SDI to include items

PAGE 53

53 with ratings of perceived sexual deviance, items that represented various levels of sexual deviance and the option to endorse a scaled response to account for frequency of sexual interest. Construct items The principle investigator and two teaching assistants held several meetings dedicated to developing a list of sexual scenarios to screen for the SDI. The goal was to create an extensive list of sexual scenarios w ith varying levels of cultural acceptability. These scenarios came from a variety of sources including, but not limited to, focus groups, journal articles, psychological scales, and internet searches. Collect community standards of deviance severity da ta Individuals were solicited from online courses as well as recruited from the Psychology Department Participant Pool. In most cases, participants received extra credit for their participation. The form requesting the community standards data was loaded onto the secure Psychology Department server. This was necessary so that information could be presented online to the participants and the data secured thereafter. During classes and through email communications, potential participants were provided the URL address for this internet survey. They were informed that the study was sexual in nature and that participation in this study was voluntary. In most cases, students were offered extra credit for attempting the survey. At the website, individuals we re presented a brief description of the study along with an informed consent form. Given the sexually explicit nature of the project they were reminded that they may withdraw from the study at any time without penalty. The participants were told that the se materials would not be connected to their names in any way and they were once again assured of their anonymity. For this purpose, the data from the survey and the informed consent were entered into two separated databases.

PAGE 54

54 After participants selected to agree to the conditions of the informed consent they were prompted to provide their electronic signature by typing their name into a designated field. Participants were then presented the list of sexual scenarios and asked to rate the degree of sexual deviance for each of the items. The instructions read as follows: Please rate the sexual deviance of each item on a scale from 1 to 11. Acts that are determined to be not at all sexually deviant are generally perceived as being sexually appropriate beha viors and acts that are determined to be extremely sexually deviant are generally perceived as being sexually inappropriate behaviors. When answering these questions please do NOT use your own opinion of the sexual scenarios. Rate the items from the vantage point of an average person applying contemporary community standards (Appendix B). Response choices were based on an 11 -point scale ranging from Not at All Sexually Deviant to Extremely Sexually Deviant. Many sexual scenarios found in Phase 1 also appeared in Phase 2. In Phase 1, participants were asked estimate their community standards ratings of sexual deviance for each item. In Phase 2 these items were present in the Sexual Deviance Inventory where participants were asked to report frequency of sexual interest for each item. However, only the most consistently rated items with the unique sexual deviance ratings were used to determine the SDI score. Eliminate items An examination of the sexual deviance ratings determined whether or not they would be eliminated from the survey and final sexual deviance scale. High standard deviations are indicative of inconsistent ratings among participants and are considered poor scale items for weighted scales. Therefore, items that had large standard deviations associated with their sexual deviance ratings were eliminated from the SDI survey. Additionally, when constructing a scale measures attitudes or proclivities across a continuum of a construct measures items with similar or duplicate ratings were eli minated from calculating the SDI score. In constructing scales with

PAGE 55

55 weighted items, it is desirable to have items that are equally distributed across the attitude or personality trait continuum (Trochim, 2006). Format questionnaire This SDI questionnaire was administered online along with numerous other measures during Phase 2 of the study. The instrument was formatted so participants could read one set of instructions for the entire list of sexual scenarios. The instructions were stated as follows: When people experience sexual arousal they can often identify the arousing image or thought that leads up to the arousal. Has visualizing, thinking about, or reflecting on any of the following scenarios ever caused you to become sexually aroused? (Appendix B) Response choices were based on a 5 point scale ranging from Never in Lifetime to Often. Participants were asked to select one of five radio buttons that corresponded with the appropriate frequency of interest. Develop scoring method In an attempt to create a more sensitive measure than those used in the past, the decision was made to develop a weighted scale to more accurately assess the nuances of this dimensional construct. According to past research, weighted scales were often more sensitive than their unweighted counterparts (Besette, et. Al, 1998). Referencing several Thurstone scaling techniques, the SDI was created and scored based loosely on its scaling principles. In place of the attitudinal ratings that are commonly used to determine Thurstone item weights the SDI used community standards of sexual deviance ratings to determine individual SDI items weights (Thurstone, 1928; Trochim, 2006). The SDI score was calculated based on a formula that accounted for the number of interests endorsed, the frequency of interest for those items endorsed, and the sexual deviance ratings of the items. The following formula was used to calculate the SDI score: Sum (frequency

PAGE 56

56 sexual deviance rating). A sample of completed item responses from the online survey can be found in Figure 3.1. Table 3.1 provides an example how SDI is scored using that sample data. Phase 2 Survey Administration Participants were solicited from online psychology courses and the Psychology Departments participant pool. In most ca ses, participants received extra credit for their participation. For convenience and to ensure the privacy of participants, the survey was administered online. All of the survey materials were loaded onto the Psychology Departments secure server. This was necessary so that information could be presented online to the participants and the data secured thereafter. During classes and through email communications, potential participants were provided the URL address for this internet survey. Students wer e informed that they would not be allowed to participate in the study if they had participated in Phase I of the study. They were informed that the study was sexual in nature and that participation in this study was entirely voluntary. However, in most c ases, students were offered extra credit for attempting the survey. At the website individuals were presented a brief description of the study, an informed consent, the survey materials, a demographic questionnaire, and a final debriefing form. Once th e participants read the informed consent they were required to designate that they understand what they have read before being granted access to the survey materials. They also were required to provide an electronic signature by typing their name into for m field. The participants were told that the survey data would not be connected to their names in any way and they could be assured of their anonymity. For this purpose, the informed consent form data and the survey data were entered into two separated d atabases. The survey was approximately 64 pages long

PAGE 57

57 and it was comprised of two sexual deviance measures, six social learning theory measures, and social desirability scale. Social learning theory scales Differential association scale (DAS): Using the principles of differential association, six questions were created that inquired about the number of friends a person has that commit, model, or support issues related to sexual deviance. Participants were instructed to answer the questions about their f riends to the best of their ability. An example of a differential association question is, To the best of your knowledge, how many of your friends would agree to send you hardcore pornography over the Internet? Responses the questions are based on a five-point scale ranging from None to A ll of T hem. Other questions inquired about knowledge of peer involvement with pornography, specific deviant sexual acts, sexual discussions, and email correspondences. Differential reinforcement peer scale (DRPe): Using the principles of differential reinforcement, nine questions were created that examined the degree that a person is differentially reinforced for sexual deviance by his or her peer group. Participants were instructed to predict their friends reactions to the specified scenarios to the best of their ability. An example of a DRPe question is, How approving do you think your friends would be if you were arrested for having sex in a public place? Responses the questions are based on a 5 -point scale ranging from Very Disapproving to V ery Approving. Other questions asked about peer responses to sexual promiscuity, criminal sexual activity, and personal sex videos. Differential reinforcement physiology scale (DRPh): Using the principles of differential reinforcement, five questions were created that examined the degree that a person is reinforced for sexual deviance through his or her physiological reactions. Participants were instructed to predict what their own physiological reactions would be to each specified scenario.

PAGE 58

58 An example of a DRPhS scenario is, Reading a story involving several incestuous scenes. Responses the questions are based on a 5 point scale ranging from Very Favorable to Very Aversive. Other questions inquired about personal physiological reactions to phone sex, viewing pornography, and promiscuity. Reinforcement balance scale (RBS): Using the principles of reinforcement balance, six questions were created that examined a persons overall reinforcement for sexual deviance. Par ticipants were instructed to report, on balance, what the outcome would be for engaging in the specified scenarios. An example of a RBS scenario is Fantasizing about forcing a person to engage in sexual activities. Responses to the questions are based on a five-point scale ranging from Most likely it would be positive to Most likely it would be negative. Other questions inquired about responses to homoerotic pornography, group sex, and pedophilia. Modeling scale (MOD): Using the principles of modeling, six questions were created that assessed the degree that a person has been exposed to models representing sexual deviance. Participants were instructed to report if they have been exposed to any of the following sexual scenarios. An example of a MOD item is, Images or videos that involve sexual bondage scenarios. Responses to the statements is based on a five -point scale ranging from Never it lifetime to Often Other items referenced being exposed to erotica, homoerotic pornography, and media illustrating unusual sexual scenarios. Definition scale (DEF): Using the principles of definition, 13 statements were created that represent personal definitions associated with sexual deviance. Participants were instructed to indicate whether they agree d or disagreed with the statements. An example of a DEF statement is, Couples that incorporate some kinkiness into the bedroom have more satisfactory sex lives. Responses to statements are based on a five item scale ranging from Strongly Agree

PAGE 59

59 to Stron gly D isagree. Other statements made reference to masturbation, sexual desirability, pornography, and extent of arousal. Sexual deviance measures Sexual deviance inventory (SDI): The SDI is a 40 item instrument that examines a persons degree of sexual deviance by assessing the frequency he or she recalls having been aroused by a variety of different sexual scenarios. The instructions for the inventory reads, When people experience sexual arousal they can often identify the arousing image or thought that leads up to the arousal. Has visualizing, thinking about, or reflecting on any of the following scenarios ever caused you to become sexually aroused? The sexual scenarios provide various levels of sexual deviance. An example of an SDI item is Sexu al behavior in public places. Item responses were based on five -point scale ranging from Never in Lifetime to O ften In general, higher scores on the SDI are associated with individuals that admit to possessing higher levels of sexual deviance. Sexua l interest card sort questionnaire revised (SICSQ R): The SICSQ R is a 39 item self report measure designed to assess sexual deviance in male sex offenders. The questionnaire examines the following ten categories of sexual deviance: a) Adult homosexuality, b) Adult heterosexuality, c) Voyeurism, d) Exhibitionism, e) Frotteurism, f) Extrafamilial molestation of girls, g) Intrafamilial molestation of girls, h) Extrafamilial molestation of boys, i) Intrafamilial molestation of boys, j) Rape of adult females, k) Sadism, l) Masochism, m) Female gender identity, n) Male gender identity, and o) Transvestic fetishism. The internal consistency for these subscale factors have been found to have alpha coefficients ranging between .78 and .97 (Holland et al., 2000). The SICSQ -R can also designate an overall sexual deviance score by combining paraphilia subscales. Questions related to three of the subscales were deleted from the survey (i.e., masochism, sadism, and rape). This decision was made based on the seve rity of

PAGE 60

60 the violence depicted in the scenarios. Item responses were rated on a seven -point scale ranging from e xtremely sexually repulsive to e xtremely sexually arousing. A sample question reads, Im wearing a matching bra, panties and a slip, all lacy. Im touching and feeling the underclothes against my body. High scores on the SICSQ-R tend to be associated with those that are sexually aroused by extreme sexual deviance. In contrast to the SDI, the SICSQ -R only assesses extreme forms of sexual deviance. Also, because of how the questions are phrased the SICSQ -R is only appropriate for males (Holland et al., 2000). Other Measures Social desirability scale: The survey materials included the following preestablished measure: the Crowne Marlowe Social Desirability Scale (CMSDS, Crowne & Marlowe, 1960). The CMSDS is a 33 item true/false scale that assesses a person s tendency to answer questions in a socially desirable manner (Crowne & Marlowe, 1960). Johnson and Fendrich (2002) calculated that t he CMSDS has a moderate degree of internal consistency with an alphacoefficient of .61. A sample question reads, I have never intensely disliked anyone. A high score on the CMSDS is associated with individuals who have a tendency to present themselves in a favorable light. Several additional measures were administered during Phase 2 that did not specifically relate to social learning theory or sexual deviance as defined by this study. The majority of these scales were included because they were present in the Pilot Study (Appendix A). Although these measures were not the focus of this study some data from these scales can be viewed in Appendix C.

PAGE 61

61 Data Analyses These data analyses were performed to test the research hypotheses of this study and to ex plore the nature and prevalence of this sexual deviance in a non -sex offender population as measured through the SDI. Pre Study Sample s ize 1. Conducted an apriori sample size calculation for multiple regression analyses associated with the SDI Phase 1 Devia nce severity data 1. Calculated standard deviations for community standards data and discarded inconsistently rated items 2. Examined the means of the community standards of sexual deviance items that did or did not appear at approximately equal intervals 3. Exami ned the demographic information for significant trends and differences Phase 2 SLT d ata 1. Determined Chronbachs Alpha scores for the six SLT measures and deleted items that detracted from each measures internal consistency 2. Examined the means and standard deviations for the individual items within the SLT measures 3. Ran oneway ANOVAs to calculate significant gender differences for the individual items within the SLT measures 4. Examined the means and standard deviations for the SLT scale scores 5. Examined the skewness and kurtosis of the SLT measures 6. Created a correlation matrix comprised of the SLT measures to detect significant relationships and the possibility of multicollinearity

PAGE 62

62 Sexual deviance data 1. Examined the means and standard deviations of the scaled SDI item responses 2. Calculated the prevalence of the endorsed dichotomous SDI items 3. Calculated the lifetime prevalence and average number of items endorsed for the categorical SDI items 4. Performed oneway ANOVAs and Chi squared analyses to detect gender differenc es on the scaled and dichotomous SDI responses 5. Performed oneway ANOVAs to detect differences amongst the categorical demographic variables for SDI items. 6. Examined correlations to detect significant trends between SDI items scores and continuous demographic variables. 7. Calculated the SDI scores 8. Examined the descriptive data for the SDI scores 9. Calculated the SICSQ R subscale scores 10. Examined the descriptive data for the SICSQR subscale scores 11. Calculated the SICSQ R combined scales scores 12. Examined descriptive data for the SICSQR combined scale scores 13. Performed two reliability analyses for the SDI scale scores, (i.e., Guttman split half coefficient and interitem reliability) 14. Estimated concurrent validity for the SDI by conducting a bivariate correlation analy sis with SICSQ -R 15. Created a correlation matrix comprised of the sexual measures to detect significant relationships and the possibility of multicollinearity Relationship between SLT and sexual deviance 1. Calculated partial correlations between the SLT measure s and both the SDI (all, females, males) and SICSQ R (males) with and without controlling for SDS 2. Created a partial correlation matrix between the 6 SLT measures and all 13 of the sexual deviance measures controlling for SDS (males only)

PAGE 63

63 3. Performed three li near regression analyses using the ENTER method with the SLT measures as the independent variables and the SDI as the dependent measure (one for all participants, one for males only, and one for females only). 4. Performed 13 linear regression models using ENTER method with the SLT measures as the independent variables and all the different measures of sexual deviance as the dependent variables (males only), (i.e., SDI, SICSQR subscales, and SICSQ -R combined scale score). 5. Performed a bivariate correlation b etween the community standards of sexual deviance item ratings and average endorsed sexual interest frequency 6. Performed a bivariate correlation between the community standards of sexual deviance item ratings and the percentage of participants that indicated that particular sexual interest

PAGE 64

64 24. Heterosexual intercourse Missionary style (Vaginal) Never in Lifetime Maybe Once Rarely Occasionally Often 25. Stating or hearing obscene language for sexual gratification Never in Lifetime Maybe Once Rarely Occasionally Often 26. Simultaneous oral sex Heterosexual 69 Never in Lifetime Maybe Once Rarely Occasionally Often Figure 3-1. Sample of three completed SDI items Table 3 -1. Calculating the SDI score M ean sexual deviance ratings for SDI items SDI 24 = 2.08 SDI 25 = 2.31 SDI 26 = 4.08 Response values Never in Lifetime Maybe On ce Occasionally Rarely Often 0 1 2 3 4 SDI score = sum ( deviance ratings frequency) = ( 2.08 2) + (7.27 0) + (9.27 5) = 59.59 Sample survey responses taken from Figure 3-1

PAGE 65

65 CHAPTER 4 RESULTS Overview Data for this study was collected at a large state university in the southeastern United States. The city the university resides has a population of 115,000 and is located at approximately the same latitude as Houston, Texas. The results of this study are divided into four sections. The first section describes the participants from Phase 1 and from Phase 2 of this study. The second section descri bes the data associated with the scale development, item responses, and scale responses for the SDI. The third section describes the data associated with the scale development, item responses, and scale responses for the six social learning theory measures. The fourth section describes the relationship between social learning theory measures and the two sexual deviance measures. For the purposes of this study, the social learning theory measures were the independent variables and the sexual deviance measures were the dependent variables and the social desirability scale acted as a control measure. Phase 1 Participants During this first phase, 235 participants completed an online questionnaire that took them approximately 15 minutes. The mean age of the participants was 20 years, ranging from 18 to 23. The gender of these participants was 27% males and 73% females. For 76% of participants, both parents were perceived as their primary care providers, 20% of participants perceived their mother to be their primary care provider, and 4% perceived the father as filling this role. Sixty five (65) percent of participants were in their freshman year of college whereas 18%, 12%, and 4% were in their sophomore, junior, and senior year, respectively. Ninetyseven (97) percent of participants reported that they were single at the time of the survey whereas 1% reported being

PAGE 66

66 married, 1% divorced, and 2% stated they were living with a significant other. Thirty-seven (37) percent of participants had voted in the past five years. Sixty -six (66) percent of respondents identified as Caucasian, 9% identified as African American, 14% as Hispanic, and 6% as Asian. The modal response for family of origin income was over $100,000 per year and it accounted for 32% of participa nts. SLT Measures Based on previous studies, SLT measures are most effective when created in the context of the dependent variable. For this study, the measures needed to assess differential association, differential reinforcement physiology, differentia l reinforcement peers, reinforcement balance, definitions, and modeling in the context of sexual deviance. Approximately 60 items were constructed and then pared down to 46 after Dr. Ron Akers selected only the items with the strongest relationship to SLT factors. Community Standards of Sexual Deviance During Phase 1, participants rated each of the 71 sexual scenarios according its level of sexual deviance as would be determined by contemporary community standards. These ratings were made on an l1 -point scale ranging Not at All Sexually Deviant to Extremely Sexually Deviant. The mean deviancy ratings for the items ranged from 1.44 for heterosexual open mouthed kissing to 10.82 for lust murder. Many of the scenarios associated with mental diagnoses (e.g., prepubescent pedophilia) rated 50th most deviant out of 50 ( n = 340, SD = .39), and criminal behaviors (e.g., rape) rated 45th out of 50 ( n = 343, SD = .85) were consistently rated among the most deviant items. The number of item responses, mean scores, and standard deviations can be found in Table 41. For ease of interpretation, the sexual scenarios have been arranged from least to most deviant.

PAGE 67

67 Standard deviations from the sexual deviance scores in Phase 1 were used to select the items best suited for the scale. Items with relatively small standard deviations were determined to be more consistently rated than other items and left on the survey to be administered in Phase 2 of the study. During Phase 2 participants were presented a questionnaire with 50 consistently rated sexual scenarios; however, only 40 of the items were used to score the SDI. The reason all 50 of the items were not used to score the SDI is that they did not represent unique levels of sexual deviance. Even though the 10 items we re not used to calculate SDI scores, the data from these items was still used to examine trends in sexual interest frequency, prevalence of sexual deviance, and gender differences and reported later in this chapter. The items eliminated from the survey as well as the items presented but not scored are also designated in Table 4 -1. Phase 2 Participants Sample size estimation revealed that 241 participants would be needed to conduct a multiple regression analysis with six predictor variables, alpha levels se t at .01, and power value of 1 A target of 300 subjects for the group was identified to accommodate for a 20% survey incompletion rate During this second phase, a total of 353 participants completed the survey that took approximately 1 hour to complete The mean age of the participants was 20 yearsold and the range was 17 to 28years -old. The gender split was 18% males and 82% females. For 69% of participants, both parents were perceived as their primary care providers, 22% of participants reported t heir mother to be their primary care provider and 6% perceived the father as filling this role. Thirty (30) percent of participants were in their freshman year of schooling whereas 16%, 26%, and 28% were in their sophomore, junior, and senior year, respectively. Ninety -four (94) percent of participants reported that they were single at the time of the survey whereas 2% were married, less than 1% was divorced, and 3% lived with a significant other.

PAGE 68

68 Sixty -four (64) percent of respondents identified as Caucasian, 11% identified as African American, 11% as Hispanic, and 9% as Asian. SLT Data and Scale Scores SLT scale construction These measures were presented in an online format during Phase 2. After participants filled out the surveys, the data from t he scale items was examined for inter item consistency. As a result, three more items were deleted that detracted from the reliability of the factors making the internal consistencies for all six scales moderately strong (Table 4.4). SLT item responses S cale scores for each of these measures were calculated by summing the value of the five item scale responses. Only scales with all items endorsed were determined to be valid scores. The number of valid scale scores along with their means, standard deviat ions, skewness, and kurtosis values was presented in Table 4 -6. SLT scale scores The data from the SLT measures provided normal distributions. To determine the relationship between these SLT measures, a bivariate correlation matrix was produced and examined. Based on the strength of correlations, the relationships between these scale scores were deemed significant but not so strong that multicollinearity was an issue. The correlations between these measures ranged from .41 to .65. The entire matrix can be found in Table 4-7. Sexual Deviance Data and Measures Both the SDI and the Sexual Interest Card Sort Questionnaire Revised (SICSQ R) were administered to assess sexual deviance. The SDI was a new scale designed to measure sexual deviance in nonsex offender populations whereas the SICSQ-R was a previously established measure used to measure sexual deviance in sex offender populations.

PAGE 69

69 Sexual Deviance Inventory The SDI was designed to assess the non -extreme levels of sexual deviance hypothesized to exist in a non-sex offender population. The measure accounted for the number, frequency, and deviance severity of the participants sexual interests. S caled item responses (frequency): Participants were asked to self report report the sexual interest frequency for a variety of sexual scenarios. Responses were based on a five -point scale that ranged from Never in Lifetime to Often Never in Lifetime scores received a value of 0 whereas Often scores received a value of 4. This scaled item response data can be found in Table 4.2 and, for ease of interpretation, has been arranged in order from low to high levels of sexual deviance. Dichotomous item responses (prevalence): Another question one might ask in regards to sexual interest is whether or not a person has ever experienced sexual interest toward a specified sexual scenario. The scaled response data was t ransformed into dichotomous data by designating the Never in Lifetime endorsements with a value of 0 and designating any Maybe Once Rarely Occasionally, and Often responses with a value of 1. This data transformation provided prevalence data for specific types of sexual deviance. In this study, the sexual interest prevalence ranged from 1% for those endorsing sexual interests related to Vomit to 94% for those endorsing sexual interests related to the Heterosexual French Kiss. To review the entire list of dichotomized response data please see Table 4.3. Categorically deviant items: Another data transformation allowed for the conceptual placement of sexual behaviors into nondeviant and deviant categories. This was accomplished by designating the midpoint on the 11-point community standards of sexual deviance scale from Phase as the value that determined whether or not behavior was considered s exually deviant. Any behavior attaining a mean community standard of sexual deviance rating below a 6 was

PAGE 70

70 considered nondeviant. Therefore, the sixteen (16) sexual scenarios presented in Phase 2 with scores equal to or lower than Foot Fetish were consid ered nondeviant; whereas, the 34 sexual scenarios with scores equal to or greater than Males French Kissing were considered deviant. On Tables 1, 2, and 3, there is a bold line that divides these conceptual categories. According to these non-deviant and deviant categories, 90.4% of the participants endorsed being aroused by at least 1 of the 34 deviant sexual scenarios. Furthermore, participants endorsed being aroused by an average of 6.97 ( SD = 6.13) of the 34 deviant sexual scenarios. Gender differences for individual items: One way independent ANOVAs revealed gender differences on numerous scaled SDI item responses. These gender differences were detected in 25 of the 50 items for the scaled responses. An examination of the means revealed that, with out exception, males disclosed a higher frequency of interest than did females. Chisquared analyses also revealed significant gender differences for the dichotomous SDI item transformations. When asked about specific sexual scenarios, proportionally mor e males recalled having ever been aroused by 20 of the 50 sexual scenarios. A summary of the percentage of participants that expressed sexual interest by gender can be found in Table 4.3. Significant differences between genders are also noted within the table. Community standards of deviance severity and deviant sexual interest When examining the sexual scenarios listed in the survey, a significant relationship was detected between ratings of community standards of sexual deviance and selfreported sexua l interest. Interest was measured through both incident and frequency. The graph in Figure 4.4 shows a negative relationship between community standards of sexual deviance ratings and the likelihood that a person will admit to having ever been aroused by deviant scenarios, r ( n = 50) = -.87, p < .01. It appears that individuals are less likely to experience an incident of sexual

PAGE 71

71 interest toward behaviors with high levels of perceived sexual deviance. Figure 4.5 shows a similar relationship existing betw een deviant sexual behavior and the frequency of sexual interest, r ( n = 50) = .81, p < .01. In other words, people report less frequent interest for behaviors with high levels of perceived sexual deviance. SDI scores Normality: The SDI scores were norm ally distributed (skewness = .80, kurtosis = .35) across a wide range of sexual deviance ( M = 169.51, SD = 110.11, See Figure 4-1). Distributions were also examined for the SDI and the SICQS -R selecting males only. The SDI ( M = 362.17, SD = 124.75) again produced normal distributions (skewness = -.17, kurtosis, -.40) whereas the SICQS R ( M = 77.63, SD = 28.22) responses scores did not produce a normal distribution of SICQS R measure scores (skewness = 1.260, kurtosis = 2.75). Demographic differences: Du ring Phase 2, were provided the SDI as part of an online survey. The score was calculated by using the following formula: sum (frequency weight). This produced a mean SDI score of 449.72 ( SD = 118.72, n = 286). Several demographic differences were d etected within the SDI scores. Regarding gender, males tended to score significantly higher on the SDI than did females, F (1, 284) = 32.41, p < .0005, as the mean score for males was 527.44 ( SD = 124.75, M = 527.44) and the mean score for females was 431.20 (SD = 109.64, N = 231). Religiousness and political orientation also had a significant influence on SDI scores with negative correlations of .26, p < .01, and .19, p <.01, respectively. In other words, people that identified with being religious or conservative tended to score lower on the SDI. However, no significant SDI differences were detected between religious groups, racial categories, ages, income classes, or marital status. R eliability: Reliability analyses were performed to determine split -half and interitem correlations. Splithalf reliability analysis was found to be sufficient as the analysis yielded a

PAGE 72

72 Guttman split-half coefficient of .84. The average interitem correlation was determined to be high as that reliability analysis produc ed a Chronbachs alpha of .93. Validity: In order to assess for concurrent validity, a similar sexual deviance scale c oncurrent validity, a correlation analysis between the SDI and the SICSQ -R was conducted using only the data from the male participants. Because the SICSQ -R is only intended for use with males the analysis was conducted using only male participant data. The SDI possess moderate levels of concurrent validity base on the degree of correlation with the SICSQR, r ( n = 46) = .41, p < .01. Sexual Deviance and Social Learning Theory Correlations The relationship between SLT and sexual deviance was examined through several correlational analyses. All of the correlations between the SDI and the SLT measures were significantly strong (Table 4 -8). The strongest relationship existed between SDI and Reinforcement Balance, r (287) = .701 whereas the weakest relationship existed between SDI and Differential Reinforcement Peer r (287) = .452. When examining the relationship between the SICSQR and the SLT measures the results the relationships were also strong but not as robust as the SDI. Every relationships except for the one between DRP and the SICSQR, r ( n = 52) = .500, p < .01 was weaker and the significance of the MOD correlation dropped from .01 to .05. Because of the gender discrepancy in the sample and the fact that the SICSQ R can only be used for males, the correlations were run a second time using only males for the SDI. However, limiting the analysis to males did not change the pat tern of differences.

PAGE 73

73 Controlling for the SDS Partial correlations were calculated between the SLT variables and the SDI while controlling for the SDS. When controlling for SDS, the strength of the relationships diminished slightly but remained significant, p < .01, and the strength of correlations stayed between .433 and .691. The largest transition happening for DRP as it shifted from .452 to .433. However, the strength of the relationships between the SLT measures and SICSQ -R diminished significantly when controlling the SDS. Both the relationships between MOD and DEF became insignificant while the significance of the relationships to DA and DRP dropped to .05. A significant difference in SDS scores was detected between males and females. The relationship between the SDI and the SDS was relatively small, r ( n = 287) = .168, p <.01, whereas relationship between the SICSQ R was moderately strong, r ( n = 52) = -.443, p < .01. That same pattern was present when only males were selected and the relatio nship between the SDI and the SDS examined, r ( n = 55) = -.425. Multiple Regressions Next, a multiple regression analysis using the Enter method to ender variables into the regression was performed to determine how well a model comprised of the SLT measur es could explain the variance in the SDI scores. The analysis revealed that the SLT model accounted for 53% of the variance in sexual deviance scores. The most significant beta weight was associated with RB ( = -.474). A direct comparison using all the participants could not be made between the SDI and the SICSQR because the SICSQ Rs questions have specifically been designed for males. However, multiple regression analysis using the same predictor variables was also performed on the SICSQ-R. In this case, the social learning theory model only 39% of the variance in that particular sexual deviance scale. Limiting the regression analysis to only males revealed that the SLT model accounted 11% more (R-Square = .49 versus R-Square = .38) of the

PAGE 74

74 variance in the SDI scores, (i.e., F (6, 35) = 5.63, p < .0005 versus the SICSQR, F (6, 33) = 3.33, p = .011).

PAGE 75

75 Figure 41. SDI score distribution (all participants).

PAGE 76

76 Figure 42. SDI score distribution (males).

PAGE 77

77 Figure 43. SICSQ R score distribution (males).

PAGE 78

78 Figure 4-4. The relationship between deviance severity and frequency of sexual interest Lifetime prevalence (%) Deviance severity for SDI items

PAGE 79

79 Figure 4-5. The relationship between deviance severity and lifetime prevalence for sexual interest Lifetime prevalence (%) Deviance severity for SDI items

PAGE 80

80 Table 4 1. Deviance se verity ratings from Phase 1 Rank Sexual Scenario Mean SD N 1 + Holding Hands 0.42 1.69 236 2 + Hugging 0.44 1.82 237 3 + Heterosexual Kissing Mouth Closed 0.48 1.69 237 4 + Massage 0.56 1.70 236 5 +++ Heterosexual Kissing Mouth Open (4) 0.63 1.74 2 35 6 +++ Heterosexual Intercourse Missionary (24) 0.92 2.08 236 7 +++ Maintaining Shaven Pubic Region (3) 1.14 2.00 236 8 +++ Male Fondling Female Breast (19) 1.15 2.07 235 9 ++ Heterosexual Intercourse Rear Entry Vaginal (15) 1.49 2.22 237 10 + Male Fondling Females Genitalia 1.55 2.38 237 11 + Miscegenation 1.81 2.65 237 12 + Body Modification 2.09 2.63 235 13 +++ Hand Job (40) 2.11 2.63 234 14 +++ Heterosexual Fellatio (28) 2.12 2.55 236 15 +++ Heterosexual Cunnilingus (18) 2.19 2.46 236 16 + Male Masturbating 2.19 2.89 237 17 + Female Fondling Own Breast 2.19 2.64 235 18 +++ Heterosexual 69 (26) 3.08 2.91 236 19 +++ Female Masturbating (44) 3.09 3.01 235 20 + Viewing Pornography 3.22 2.82 237 21 +++ Food for Sexual Purposes (5) 3.30 2. 83 236 22 +++ Erotic Performance (14) 3.41 2.94 237 23 + Sex with Disabled 3.44 3.27 237 24 +++ Narratophilia (25) 3.48 2.60 236 25 + Fetishism 3.67 3.38 235 26 + Pregnancy 3.68 3.09 236 27 +++ Female Homo Mouth Closed Kiss (43) 3.70 2.82 236 28 + Sex Toys 3.77 2.98 235 29 + Body Parts 3.77 3.25 237 30 +++ Male Homo Kiss Mouth Closed (49) 4.69 3.34 237 31 +++ Female Homo French Kiss (45) 4.89 3.06 237 32 +++ Foot Fetish (39) 4.95 3.00 235 33 +++ Male Homo French Kiss (47) 5.42 3.47 234 34 +++ Heterosexual Anal Sex (29) 5.65 3.02 236 35 +++ Sex with a Stranger (37) 5.90 2.87 236 36 +++ Bondage (16) 6.03 2.83 237 37 +++ Public Places (38) 6.07 2.32 235 38 ++ Homosexual Cunnilingus (46) 6.14 3.16 236 39 ++ Exhibitionism (8) 6.27 2.92 235 40 ++ Infantilism (34) 6.31 2.93 237 41 +++ Cross Generational (22) 6.31 2.78 236 42 +++ Group Sex 1 Male & 2 Female (33) 6.50 2.86 237 43 ++ Homosexual Anal Sex (48) 6.56 3.27 237 44 + Observing Sex 6.58 2.73 236 45 ++ Homosexual Fellatio (50) 6.64 3.31 235 46 +++ Voyeurism (13) 6.81 2.99 236 47 +++ Transvestism (41) 6.89 2.87 236 48 ++ Group Sex 1 Female & 2 Male (31) 6.98 2.75 235 49 +++ Transexualism (27) 7.03 3.20 237 50 +++ Sadism and Masochism (42) 7.17 2.73 237 + Represented in Phase 1 surve y but eliminated from Phase 2 survey ++ Represented in Phase 1 survey and Phase 2 survey but not used to calculate the SDI score +++ Represented throughout all surveys and used to calculate the SDI scores

PAGE 81

81 Table 4 1. Continued Rank Sexual Scenario Mean SD N 51 + Somnophilia 7.37 2.72 237 52 + Prostitution 7.58 2.76 235 53 +++ Sex with Partner as Audience (23) 7.58 2.69 234 54 +++ Enema (32) 7.59 2.62 236 55 +++ Adultery (9) 8.08 2.49 236 56 +++ Post Pubescent Pedophilia (6) 8.27 2.28 237 57 + Sex Between Children 8.35 2.48 235 58 + Obscene Phone Calls 8.49 2.36 236 59 +++ Defecation (10) 8.56 2.42 235 60 +++ Bribery (7) 8.61 2.34 237 61 +++ Frotteurism (1) 8.68 2.26 235 62 +++ Asphyxiation (17) 8.81 2.34 234 63 +++ Urophilia (21) 8.91 2.22 236 64 ++ Vomit (11) 9.14 2.21 236 65 ++ Rape (35) 9.15 2.05 237 66 +++ Bestiality (12) 9.43 1.80 235 67 + Sex Between Siblings 9.44 1.93 236 68 +++ Necrophilia (2) 9.46 2.02 235 69 ++ Lust Murder (36) 9.50 2.06 237 70 +++ Parent Child Incest (30) 9 .50 1.90 237 71 +++ Prepubescent Pedophilia (20) 9.55 1.24 237 + Represented in Phase 1 survey but eliminated from Phase 2 survey ++ Represented in Phase 1 survey and Phase 2 survey but not used to calculate the SDI score +++ Represented throughout all surveys and used to calculate the SDI scores

PAGE 82

82 Table 4 2. Differential Association Please answer the following questions about your friends to the best of your ability. (None = 0 to All of them = 4) Gender Difference Mean Scores SD Male Fem All All 1. To the best of your knowledge, how many of your friends would find anal sex arousing? ** 1.25 0.78 0.87 .92 2. To the best of your knowledge, how many of your friends engage in promiscuous sexual behavior? 1.52 1.38 1.42 1.05 3. To the bes t of your knowledge, how many of your friends have pornographic videos depicting risqu sexual practices? ** 2.06 1.00 1.20 1.09 4. To the best of your knowledge, how many of your friends do you think have sought out hardcore sexual material over the in ternet? ** 2.71 1.12 1.43 1.21 5. About how many of your friends would tell a dirty joke referencing risqu sexual practices? ** 2.81 2.25 2.35 1.18 6. Item deleted 7. To the best of your knowledge, how many of your friends would agree to send you hardcore pornography over the Internet? ** 1.90 0.61 0.86 1.13 p < .05 ** p < .01

PAGE 83

83 Table 4 3. Differential Reinforcement Peer To the best of your knowledge, what do you think your friends reaction would be to the following scenarios? (V ery disapproving = 0 to Very approving = 4) Gender Difference Mean Scores SD Male Fem All All 1. If they found out you were arrested for having sex in a public place. ** 2.16 1.36 1.53 1.21 2. If they knew that you owned a collection of hardcore pornographic ** 2.13 1.17 1.34 1.08 3. If they found out you had engaged in anal sex with a member of the opposite sex. ** 1.78 1.29 1.37 1.00 4. Item deleted 5. If they knew you forced someone to pleasure you sexually. ** 0.48 0.98 0.90 .97 6 If they found out you had sexual relations with a stranger. ** 1.95 0.85 1.08 .99 7. If they knew you had engaged in sexual relations with someone that was 16. ** 1.08 0.40 0.53 .77 8. If they saw a video of you having group sex on the internet. 1.30 0.16 0.38 .83 9. If they knew you engaged in deviant sexual behaviors frequently ** 1.44 0.72 0.85 .92 p < .05 ** p < .01 Table 4 4 Differential Reinforcement Physiology To the best of your knowledge, what do you think your physiologica l reaction be as you engaged in the following activities? (Very favorable = 0 to Very aversive = 4) Gender Difference Mean Scores SD Male Fem All All 1. Viewing hardcore pornographic videos ** 1.41 2.33 2.16 1.24 2. Reading a story involving sev eral incestuous scenes. 2.62 2.91 2.86 1.21 3. Having sexual intercourse with someone you barely knew. ** 2.03 2.96 2.76 1.20 4. Reading a story involving several incestuous scenes. 2.87 2.98 2.97 1.14 5. Discussing sexually explicit topics w ith a telephone sex operator. ** 2.57 3.19 3.07 1.03 p < .05 ** p < .01

PAGE 84

84 Table 4 5 Reinforcement Balance Whether or not you would ever actually engage in the following, on balance what do you think is most likely to be the outcome? (Most likely w ould be positive = 0 to Most likely would be negative = 4) Gender Difference Mean Scores SD Male Fem All All 1. Masturbating to hardcore sexual pornography ** 1.10 2.21 2.00 1.39 2. Having sexual relations with a minor ** 2.90 3.65 3.51 .86 3. Having sexual relations with more than one partner ** 1.59 2.88 2.64 1.19 4. Fantasizing about forcing a person to engage in sexual activities ** 3.08 3.44 3.36 1.02 5. Masturbating to homoerotic pornography ** 3.60 3.08 3.17 1.14 6. Participating in hardcore sexual behaviors with another person ** 1.03 1.89 1.73 1.34 p < .05 ** p < .01 Table 4 6 Modeling Have you ever been exposed to any of the following? (Never in lifetime = 0 to Often = 4) Gender Difference Mean Scores SD Mal e Fem All All 1. Images or videos you would categorize as hardcore pornography? ** 3.21 1.80 2.07 1.17 2. Images or videos that involve sexual bondage scenarios? ** 1.92 1.22 1.35 1.03 3. Images or videos that involve homoerotic scenarios? 1.36 1.4 6 1.45 1.04 4. Images or videos that portray individuals being forced into sexual situations? ** 1.48 0.96 1.05 1.02 5. Images or videos that involve individuals engaging in unusual sexual activities? ** 2.11 1.44 1.56 1.00 6. Literature or erotica d escribing hardcore sexual activities? 1.92 1.59 1.64 1.11 p < .05 ** p < .01

PAGE 85

85 Table 4 7 Definitions We all have opinions on pornography and sexual behavior. Please indicate your agreement or disagreement with the following statements about se xual arousal and sexual behavior. (Strongly agree = 0 to Disagree strongly = 4) Gender Difference Mean Scores SD Male Fem All All 1. If everyone involved is a consenting adult it is acceptable to look at some hardcore pornography. ** 0.83 1.38 1. 29 1.30 2. If everyone involved is a consenting adult it is acceptable to participate in some hardcore sexual behaviors 0.95 1.23 1.20 1.21 3. Couples that incorporate some kinkiness into the bedroom have more satisfactory sex lives. 1.13 1.25 1.21 1.02 4. Individuals that are more willing to experiment with risqu sexual behaviors are sexier. ** 1.34 2.05 1.91 1.07 5. Individuals that feel risqu sexual behaviors are sick are more likely to have routine sex lives. 1.46 1.73 1.67 1.01 6. Fantasizing about risqu sexual scenarios is harmless. 0.95 1.12 1.08 1.09 7. Masturbating to hardcore pornography is harmless. ** 0.92 1.57 1.45 1.27 8. Participating in risqu sexual scenarios is harmless. ** 1.40 1.87 1.79 1.14 9. Item deleted 10. Men willing to experiment with hardcore sexual scenarios are sought after as sexual partners. ** 2.10 2.56 2.47 .94 11. Women willing to experiment with hardcore sexual scenarios are sought after as sexual partners. 1.19 1.54 1.47 1.11 12. It is necessary to engage in some risqu sexual practices in order to keep ones sex life interesting. 1.65 2.01 1.93 1.06 13. Risqu sexual scenarios are more sexually arousing than most normal sexual scenarios. ** 1.60 2.17 2.06 1.10 p < .05 ** p < .01 Table 4 8 Descriptive statistics for social learning theory variables Social learning theory Chronbachs alpha Number of items Skew Kurt N M SD Differential Association .83 6 .637 .103 342 8.21 4.73 Diff Reinforcement Peers .81 8 .484 .055 329 7.75 4.97 Diff Reinforcement Phys .71 5 .534 .250 349 13.83 3.96 Reinforcement Balance .74 6 .281 .535 345 16.46 4.62 Modeling .81 6 .139 .203 346 9.12 4.54 Definitions .89 12 .670 .684 335 19.43 9.02 Table 4 9. Correlation matrix for social learning theory variables DA DRPe DRPh RB MOD DEF Differential Association (DA) .596 .431 .588 .473 .538 Diff Reinforcement Peer (DRPe) .517 .593 .407 .548 Diff. Reinforcement Phys (DRPh) .648 408 .512 Reinforcement Balance (RB) .596 .644 Modeling (MOD) .475 Definitions (DEF) All correlations are significant at the .01 level

PAGE 86

86 Table 4 10 SDI frequency from Phase 2 Mean Scores by Gender Rank Diff Sexual Scenario Male (63) Female (279) Diff. All (353) SD 1 +++ Heterosexual Kissing Mouth Open (4) 3.03 3.05 3.05 1.23 2 +++ Heterosexual Intercourse Missionary (24) 3.14 2.99 3.02 1.25 3 +++ Maintaining Shaven Pubic Region (3) 1.51 1.22 1.27 1.44 4 +++ Male Fon dling Female Breast (19) 3.16 2.89 2.94 1.20 5 ++ Hetero Intercourse Rear Entry Vaginal (15) 3.03 2.46 ** 2.56 1.46 6 +++ Hand Job (40) 2.71 2.30 2.37 1.32 7 +++ Heterosexual Fellatio (28) 2.87 2.18 ** 2.30 1.42 8 +++ Heterosexual Cunnilingus (18) 2.60 2.66 2.65 1.40 9 +++ Heterosexual 69 (26) 2.63 2.04 ** 2.15 1.42 10 +++ Female Masturbating (44) 2.59 1.47 ** 1.67 1.45 11 +++ Food for Sexual Purposes (5) 1.70 1.41 1.46 1.27 12 +++ Erotic Performance (14) 1.16 1.21 1.20 1.21 13 +++ Narratoph ilia (25) 1.49 1.25 1.30 1.29 14 +++ Female Homo Mouth Closed Kiss (43) 2.14 .77 ** 1.03 1.35 15 +++ Male Homo Kiss Mouth Closed (49) .19 .23 .23 .63 16 +++ Female Homo French Kiss (45) 2.44 .87 ** 1.16 1.42 17 +++ Foot Fetish (39) .46 .22 .26 .70 18 +++ Male Homo French Kiss (47) .24 .29 .28 .73 19 +++ Heterosexual Anal Sex (29) 1.52 .71 ** .86 1.17 20 +++ Sex with a Stranger (37) 1.71 .71 ** .89 1.07 21 +++ Bondage (16) 1.35 1.29 1.30 1.33 22 +++ Public Places (38) 1.81 1.60 1.64 1.19 23 ++ Homosexual Cunnilingus (46) 2.49 .66 ** 100 1.41 24 ++ Exhibitionism (8) .49 .36 .38 .84 25 ++ Infantilism (34) ,27 ,11 ,14 ,48 26 +++ Cross Generational (22) .52 .14 ** .21 .61 27 +++ Group Sex 1 Male & 2 Fem (33) 2.41 .73 ** 1.04 1.41 28 ++ Ho mosexual Anal Sex (48) .16 .18 .17 .58 29 ++ Homosexual Fellatio (50) .24 .19 .20 .63 30 +++ Voyeurism (13) 1.06 .55 ** .64 ,94 31 +++ Transvestism (41) .03 .09 .08 .43 32 ++ Group Sex 1 Female & 2 Male (31) .81 .66 .69 1.09 33 +++ Transexualism ( 27) .19 .10 .12 .51 34 +++ Sadism and Masochism (42) .44 .47 .47 .90 35 +++ Sex with Partner as Audience (23) 1.17 .47 ** .60 1.00 36 +++ Enema (32) .10 .05 .06 .30 37 +++ Adultery (9) .86 .35 ** .44 .84 38 +++ Post Pubescent Pedophilia (6) .62 .09 ** .18 .57 39 +++ Defecation (10) .06 .03 .04 .27 40 +++ Bribery (7) .51 .14 ** .21 .59 41 +++ Frotteurism (1) 1.62 1.60 1.61` .88 42 +++ Asphyxiation (17) .33 .14 .17 .61 43 +++ Urophilia (21) .17 .06 .08 .42 44 ++ Vomit (11) .05 .02 .02 ,20 45 ++ Rape (35) .38 .49 .47 .85 + Represented in Phase 1 survey but eliminated from Phase 2 survey ++ Represented in Phase 1 survey and Phase 2 survey but not used to calculate the SDI scores +++ Represented throughout all surveys and used to calculate the SDI scores p < .05 ** p < .01

PAGE 87

87 Table 4 10 Continued Mean Scores by Gender Rank Diff Sexual Scenario Male (63) Female (279) Diff. All (353) SD 46 +++ Bestiality (12) .08 .05 .06 .29 47 +++ Necrophilia (2) .22 .03 ** .06 .38 48 ++ Lust Murder (36) .11 .03 .04 .30 49 +++ Parent Child Incest (30) .17 .04 ** .07 .31 50 +++ Prepubescent Pedophilia (20) .19 .04 ** .07 .39 + Represented in Phase 1 survey but eliminated from Phase 2 survey ++ Represented in Phase 1 survey and Phase 2 survey but not used to calculate the SDI scores +++ Represented throughout all surveys and used to calculate the SDI scores p < .05 ** p < .01

PAGE 88

88 Table 4 11 SDI l ifetime prevalence from Phase 2 % Prevalence by Gender Rank Sexual scenario Diff Male (63) Female (279) All (353) 1 +++ Heterosexual Kissing Mouth Open (4) 95.2 93.9 94.1 2 +++ Heterosexual Intercourse Missionary (24) 93.7 91.7 92.1 3 +++ Maintaining Shaven Pubic Region (3) 67.7 52.2 55.0 4 +++ Male Fondling Female Breast (19) 95. 2 92.8 93.2 5 ++ Hetero Intercourse Rear Entry Vaginal (15) 90.3 83.5 84.7 6 +++ Hand Job (40) 90.5 87.0 87.6 7 +++ Heterosexual Fellatio (28) 85.5 83.0 83.4 8 +++ Heterosexual Cunnilingus (18) 90.3 86.3 87.1 9 +++ Heterosexual 69 (26) 87.3 78.0 79.7 10 +++ Female Masturbating (44) ** 85.7 63.9 67.9 11 +++ Food for Sexual Purposes (5) 77.4 68.4 70.0 12 +++ Erotic Performance (14) 64.5 60.0 60.7 13 +++ Narratophilia (25) 71.0 59.9 61.9 14 +++ Female Homo Mouth Closed Kiss (43) ** 77.4 38.0 45.2 15 +++ Male Homo Kiss Mouth Closed (49) 11.3 15.3 14.5 16 +++ Female Homo French Kiss (45) ** 81.0 41.5 48.8 17 +++ Foot Fetish (39) 64.5 59.8 60.7 18 +++ Male Homo French Kiss (47) 14.3 16.2 15.9 19 +++ Heterosexual Anal Sex (29) ** 69.8 38 .3 44.1 20 +++ Sex with a Stranger (37) ** 73.0 46.9 51.8 21 +++ Bondage (16) 58.7 60.9 60.5 22 +++ Public Places (38) 81.0 77.3 77.9 23 ++ Homosexual Cunnilingus (46) ** 77.8 33.2 41.5 24 ++ Exhibitionism (8) 30.2 20.1 22.0 25 ++ Infantilism (34) 15.9 7.9 9.5 26 +++ Cross Generational (22) ** 30.2 10.5 14.2 27 +++ Group Sex 1 Male & 2 Fem (33) ** 77.8 34.3 42.4 28 ++ Homosexual Anal Sex (48) 7.9 10.9 10.3 29 ++ Homosexual Fellatio (50) 11.7 11.9 11.9 30 +++ Voyeurism (13) ** 66.7 36.1 41.5 31 +++ Transvestism (41) 3.2 5.0 4.7 32 ++ Group Sex 1 Female & 2 Male (31) 47.6 32.7 35.5 33 +++ Transexualism (27) 11.3 5.5 6.5 34 +++ Sadism and Masochism (42) 30.6 26.1 26.9 35 +++ Sex with Partner as Audience (23) ** 54.0 29.2 33.8 36 +++ Enema (32) 6.3 3.6 4.1 37 +++ Adultery (9) ** 46.0 22.4 26.8 38 +++ Post Pubescent Pedophilia (6) ** 34.9 6.1 11.4 39 +++ Defecation (10) 4.8 1.8 2.3 40 +++ Bribery (7) ** 31.7 9.7 13.8 41 +++ Frotteurism (1) ** 46.8 18.0 23.2 42 +++ Asphyxiation ( 17) 15.9 8.3 9.7 43 +++ Urophilia (21) 8.3 4.0 4.8 44 ++ Vomit (11) 3.2 1.1 1.5 45 ++ Rape (35) 31.7 28.1 28.7 46 +++ Bestiality (12) 4.8 4.3 4.4 47 +++ Necrophilia (2) ** 9.5 2.2 3.5 ++ Represented in Phase 1 survey and Phase 2 survey but not u sed to calculate the SDI scores +++ Represented throughout all surveys and used to calculate the SDI scores p < .05 ** p < .01

PAGE 89

89 Table 4 11 Continued % Prevalence by Gender Rank Sexual scenario Diff Male (63) Female (279) All (353) 4 8 ++ Lust Murder (36) 4.8 1.8 2.3 49 +++ Parent Child Incest (30) ** 12.7 3.2 5.0 50 +++ Prepubescent Pedophilia (20) ** 12.7 2.2 4.2 + Represented in Phase 1 survey but eliminated from Phase 2 survey ++ Represented in Phase 1 survey and Phase 2 surve y but not used to calculate the SDI scores +++ Represented throughout all surveys and used to calculate the SDI scores p < .05 ** p < .01 Table 4 12. Descriptive statistics for sexual deviance variables Chronbachs alpha Number of items Skewness Kurtosis n M SD SDI (all) Scale score .91 40 .80 .35 287 169.51 110.11 SDI (males) Scale score .90 40 .23 .38 54 251.23 114.41 SICSQ R (males) .89 32 .65 .77 53 71.63 20.76 SICSQ R (males) Subscale Scores Adult hom osexuality .62 3 1.53 1.71 62 4.90 2.92 Adult heterosexuality .90 3 1.54 1.17 62 17.18 5.60 Voyeurism .93 3 .98 .25 63 10.92 3.61 Exhibitionism .79 3 1.40 1.75 54 6.00 3.65 Frotteurism .91 3 .99 .34 63 7.19 4.79 Extrafamilial molestation of girls .72 3 1.94 3.12 62 4.58 2.96 Intrafamilial molestation of girls .75 3 3.55 12.96 62 3.56 1.67 Extrafamilial molestation of boys .59 3 3.67 13.94 63 3.44 1.46 Intrafamilial molestation of boys .76 3 3.97 16.59 63 3.52 1.66 Rape of adult femal es .89 3 2.30 4.56 63 4.68 3.28 Transvestic fetishism .66 2 1.41 .99 63 3.40 2.07

PAGE 90

90 Table 4 13. Correlation matrix for sexual deviance variables 1 2 3 4 5 6 7 8 9 10 11 12 13 1. SDI .43** .35** .28* .42** .43** .56** .42** .16 .23 .01 .36** .0 2 2. SICSQ -R .47** .38** .60** .84** .79** .55** .65** .20 .43** .68** .61** 3. Adult homosexuality .27* .09 .46** .28* .31* .34** .35** .23 .46** .38** 4. Adult heterosexuality .83** .13 .30** .05 .04 .40 ** .23 .07 .10 5. Voyeuri sm .34* .54** .25* .23 .27* .05 .29* .31* 6. Exhibitionism .74** .64** .57** .38** .40** .73** .55** 7. Frotteurism .63** .53** .05 .20 .67** .46** 8. Extrafamilial molestation girls .42** .36** .28* .55** .29* 9. In trafamilial molestation girls .43** .72** .71** .52** 10. Extrafamilial molestation boys .57** .31* .24 11. Intrafamilial molestation boys .53** .42** 12. Rape of adult females .52** 13. Transvestic fet ishism Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level Table 4 -14. Correlation for the SLT variables for the sexual deviance variables controlling and not controlling for social desirabilit y (across gender) SDI (all participants) SDI (females only) SDI (males only) SICSQ R (males only) Exclude pairwise Range N (265 to 287) Range N (218 to 232) Range N (40 to 54) Range N (40 to 52) No Control Control for SDS No Control Control for SDS No Control Control for SDS No Control Control for SDS SLT Differential Association .54** .53** .45** .44** .50** .40** .44** .33* Differential Reinforce Peer 46** .46** .39** .38** .36* .35* .50** .50** Differential Reinforce Phys .51** .50** .43** .41** .57** .51** .39** .30* Reinforce Balance .70** .69** .68** .67** .61** .55** .58** .53** Modeling .52** .50** .46** .44** .51** .43** .36* .25 Definition .50** .48** .46** .43** .51** .41** .39** .28 SDS .17** .17* .44** .44** p < .05 ** p < .01

PAGE 91

91 Table 4 15. Correlation across sexual deviance variables for the SLT variables controlling for social desirability (males only) Social learning theory variables Sexual deviance variables DA DRPe DRPh RB MOD DEF SDI .40 ** .35* .51** .55** .43** .41** SICSQ R .33* .50** .30* .53** .25 .27 SICSQ R subscales Adult homosexuality .11 .12 .30* .33* .28 .09 Adult heterosexuality .38** .33* .10 .18 .06 .23 Voyeurism .40** .39* .27 .30* .08 .38** Exhibitionism 19 .26* .20 .28 .20 .17 Frotteurism .06 .32* .32* .34* .05 .25 Extrafamilial molestation girls .05 .17 .32* .22 .03 .11 Intrafamilial molestation girls .00 .23 .11 .23 .07 .14 Extrafamilial molestation boys .27 .12 .20 .12 .06 .01 Intrafamilial molestation boys .09 .17 .03 .02 .01 .02 Rape of adult females .01 .14 .22 .21 .06 .15 Transvestic fetishism .09 .19 .20 .18 .11 .09 Exclude pairwise, range of participants from 40 to 47 p < .05 ** p < .01 Table 4-16. Multiple regression analy ses testing the utility of the SLT model in expaining sexual deviance in the generalized sexual deviance variables (across gender) SDI (all participants) N = 246 SDI (males only) N = 41 SICSQ R (males only) N = 39 SLT Differential Association .0 89 .146 .086 Differential Reinforcement Peers .012 .230 .086 Differential Reinforcement Phys .068 .269 .274 Reinforcement Balance .474** .236 .506 Modeling .154** .301* .135 Definitions .047 .132 .298 R Square .53** .49** .38* F /Probability F 44.201/.0005 5.625/.0005 3.338/.011 p < .05 ** p < .01

PAGE 92

92 Table 4 17. Multiple regression analyses testing the utility of the SLT model in explaining sexual deviance in the sexual deviance variables (males only) Measures of sexual devia nce R Square F/Probability F SDI .49** 5.53/.0005 SICSQ R .38* 3.33/.011 SICSQ R subscales Adult homosexuality .32* 3.29/.010 Adult heterosexuality .29* 2.82/.022 Voyeurism .34** 3.60/.006 Exhibitionism .23 1.72/.147 Frotteurism .31* 3.01/.0 16 Extrafamilial molestation girls .35** 3.52/.007 Intrafamilial molestation girls .12 0 .92/.489 Extrafamilial molestation boys .13 1.02/.429 Intrafamilial molestation boys .05 0 .37/.891 Rape of adult females .17 1.36/.254 Transvestic fetishism .19 1.57/.180 Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level

PAGE 93

93 CHAPTER 5 DISCUSSION Overview and Hypothesis As was described in the introduction, most theories attempting to explai n sexual deviance lack empirical support. Repeated insignificant findings from sex offender studies have caused researchers to ponder the deleterious effects of small sample sizes, research methodology, and funding limitations (Laws & ODonohue, 1997). Results from this study suggest that nonsex offenders may be a more appropriate target for sexual deviance research. When examining deviant sexual interests in non sex offenders the phenomenon can be accounted for using social learning theory. Four hypotheses were tested through this study. The first was that non sex offender populations experience deviant sexual interests. The second was that sexual deviance, as defined by and measured as a dimensional construct, could be explained by social learning theory. The third hypothesis was that sexual deviance measured by combination of deviant sexual interests could be better accounted for than a sexual deviance measured by isolated deviant sexual interests. The fourth hypothesis was that sexual deviance m easured by a combination of various levels of sexual deviance could be better accounted for than sexual deviance measured by a combination of only the most extreme forms of sexual deviance. The results of this study are discussed followed by an account of various research choices, methodological limitations, and policy implications Conceptual Shift The research and findings within this paper represent a shift in the conceptualization of sexual deviance. This shift motivated the development of a new survey instruments and changed the target population of sexual deviance studies.

PAGE 94

94 Behavior versus Interest The majority of research examining sexual deviance has been performed on sex offenders. One of the benefits of targeting this population is that these individuals usually have a confirmed deviant sexual interest associated with their conviction. In contrast to only examining behavioral manifestations of deviant sexual interest, this research uses a more liberal definition of interest that includes sexual fantasy, masturbation, and pornography. Pathological versus Normal The presence of deviant sexual interests, such as pedophilia and exhibitionism, in the DSM IV -TR has encouraged the conceptualization of sexual deviance as pathology. Although this conceptualization has benefited society in terms of controlling potentially harmful behaviors it has complicated the study of sexual deviance. The data associated with this study and others suggest that the possession of deviant sexual interests is a normal experience and therefore is not pathological. Specific versus Generalized The vast majority of research examining sexual deviance has been conducted on isolated sexual interests. Although some studies do expand the notion of sexual deviance to categories of interest, none so far acknowledge the relationship between all deviant sexual interests. In contrast, this study examines a broad array of deviant sexual interests and understands sexual deviance to be an overall proclivity instead of an isolated or specific category of deviant sexual interest. Dichotomou s versus Dimensional Prescribing to sexual deviance is a medical construct encourages individuals to conceptualize sexual deviance in dichotomous terms. Either a person fits the criteria for a paraphilia and is considered sexually deviant or he or she does not fit the criteria and is not

PAGE 95

95 considered sexually deviant. However, acknowledging that the proclivity toward deviant sexual interests is a normal phenomenon allows for this construct, like other normal traits, to be conceptualized along a continuum. Through the assessment of a broad array deviant sexual interests scores are determined that range from low to high levels of sexual deviance. What is Sexual Deviance? Sexual deviance is a persons ove rall proclivity toward deviant sexual interest. Forty (40) sexual scenarios were examined and the following formula was applied to determine an individuals sexual deviance score: SDI = Sum (rating of deviance severity for sexual scenario frequency of i nterest). The terms interest and deviance severity need to be clarified to fully understand this construct and its formula. Devian ce Severity The deviance severity for a sexual scenario is the community standard the acceptability of specific sexual behaviors. Some argue that this type of determination of deviance is too subjective. However, similar ratings have been used to determine obscenity standards in court proceedings (Miller v. California, 1974). Individuals in this study were able to assign ratings of deviance severity with moderate levels of consistency as determined by the standard deviations of their ratings. Prevailing social discourse is thought to provide individuals with this deviance severity information. Many factors are though t to influence how sexual behaviors are conceived within our society such as their potential for harm, risk to health, religious ideals, political views, and legal ramifications. Sexual Interest This study expands the definition of sexual interest to include experiences associated with fantasy, masturbation, pornography, as well as behavior. However, the study does limit sexual

PAGE 96

96 interest to only those events where sexual arousal takes place. The specific wording of the SDI is as follows: When people experience sexual arousal they can often identify the arousing image or thought that leads up to the arousal. Has visualizing, thinking about, or reflecting on any of the following scenarios ever caused you to become sexually aroused? Prevalence of deviant sex ual interest Similar to the results from the Pilot study, a large percentage of individuals reported having experienced deviant sexual interests.4 For instance, 90% of participants reported that they had been aroused by at least 1 of the 24 deviant sexua l scenario s listed in the SDI. These were similar findings to those that were revealed in a previous study by Bowman and Schneekloth (1999). The study followed a similar format and found that 97% of college students had experienced arousal toward at least one of the 37 deviant sexual scenarios listed. Other studies have also detected deviant sexual interests in non -forensic populations. For example, Briere (1989) found that 21% of males from a university sample reported having a sexual attraction to sma ll children. Crepault and Crepault (1980) found that 61% of males reported that had experienced sexual fantasies about initiating a young girl. Another study revealed that 17% of males had experienced having had sexual thoughts about girls under the age of 15. Those experiencing sexual thoughts dropped to five percent for girls under the age of 12 (Templeman & Stinett, 1991). That interest was similar to the percentages of males claiming the interest toward underage females in this study. Of the males surveyed in this study, 34% endorsed having had experienced arousal to a girl that had experienced puberty but that was 4 The cutoff for what is considered deviant versus nondeviant was determined in the pilot study and can be found in Appendix A.

PAGE 97

97 below the age of consent. Thirteen percent (12.7%) endorsed having had experienced arousal toward girls below the age of puberty. Gender differences in deviant sexual interests Throughout this study, males reported higher amounts of sexual interest than females. This was true for both nondeviant and deviant sexual interests. A higher percentage of males endorsed being aroused toward 17 of the 50 sexual scenarios listed on the SDI, p < .05 (Table 4.3. Males also endorsed higher frequencies of arousal associated with 25 of the 50 sexual scenarios listed on the SDI, p < .05 (Table 4.2). Out of the 14 deviant sexual scenarios that had 20% or more participants reporting interest, 9 of these deviant sexual interests had higher percentage of males reporting than females (Table 4.3). There were no exceptions where women reported a higher percentage of interest. The prevalence for female h omoerotic scenarios was high relative to analogous male homoerotic scenarios. Both women and men reported significant amounts of interest toward these scenarios. The percentage of females and males reporting having been aroused by two females performing cunnilingus on one another was 33.2% and 77.8% respectively. The percentage of females and males reporting having been aroused by two females French kissing was 41.5% and 81% respectively. It seems that even though females are aroused by these scenarios, the male interest in these scenarios may be a greater sexual influence in our society. In contrast to the data from female homoerotic scenarios, there was very little difference for the prevalence of interest between genders for the sexual interests toward male homoerotic scenarios. The percentage of male and female interest toward two males engaged in anal sex was 7.9% and 10.9%, respectively. The percentage of male and female interest toward two males engaged in fellatio was 11.7% and 11.9% respective ly. The percentage of male and female interest toward two males French kissing was 14.3% and 16.2% respectively.

PAGE 98

98 It is also interesting to note that, although not statistically significant, a higher percentage of female sexual interest was only reported for 4 out of 50 of the SDIs sexual scenarios and that three of these were associated with male homoerotic scenarios. One possible explanation for this trend is that male sexual interest could be depressed as a reaction to the homophobic attitudes that e xist within our society (Kite & Whitley, 2003). Another possibility is that females could be experiencing a similar phenomenon that is occurring for males toward female homoerotic scenarios. This type voyeuristic arousal has not been documented in research studies but it has been documented in a content analysis of best selling pornography videos for the year 2006. Whereas 22.6% of these videos displayed two females engaging in cunnilingus, none of the most popular videos showed any males engaging in any homoerotic sexual acts. The interest toward females engaging in homoerotic activity has also become wildly popular in todays fringe media. For example, a large portion of the scenes depicted on the Girls Gone Wild infomercials and videos depict this brand of girl on girl bisexual behavior. The popularity of these videos has allowed the production company to monopolize many of pay per view programming slots available through cable advertising. The Los Angeles times reported Mantra Films, the production company that owns the rights to Girls Gone Wild, boasts as much as $40 million in sales each year (Hayes, 2006). In contrast to the previously afore mentioned SDI data, there may have been one exception where females would have endorsed a higher frequen cy of sexual interest that this study failed to detect. This would have been for the female interest toward submissive rape scenarios. Whereas males endorsed a higher frequency of interest for 24 of the 50 sexual scenarios and more individuals reported the lifetime prevalence of interest for 20 of the 50 scenarios; females did not endorse a higher degree of sexual interest for any of the sexual

PAGE 99

99 scenarios. However, this study did not dissect the sexual interest toward rape scenarios into submissive and aggressive behavior as many studies had done in the past. Five studies that examined the sexually interest toward dominance rape fantasies showed that, on average, 26.2% of males expressed this interest versus 11% of females (Hunt, 1974; Sue, 1979; Miller & Simon, 1980; Arndt et al., 1985; Person, 1989). Whereas men report higher rate of sexual interest toward the dominance role associated with rape scenario, women report higher rates of sexual interest toward the submissive role associated with rape scenarios. A review of rape literature summarized that between 31% and 57% of women have fantasies in which they are forced into sex against their will, and for 9% to 17% of women these are a frequent or favorite fantasy experience (Critelli & Bivona, 2008). Although this study did not separate the sexual interest toward rape into dominant and submissive roles, these findings support the prevalence of these interests as 31.7% of males and 28.2% of females reported a sexual interest in rape scenarios. If this study had separated interest into submissive and aggressive categories, it is assumed that the findings would have been more congruent those previous studies with woman tending to have more submission fantasies and men tending to have more rape perpetration fantasies. Instead, both tendencies were combined into one category with the greater degree of female submissive interest and the greater degree of aggressive interest neutralizing each others statistical significance. Moderate prevalence rates have al so been found for the sexual interest toward group sexual experiences. Six studies examined the sexual interest toward having sex with multiple partners and found that more males consistently reported being interested in those experiences. The average percentage of participants reporting that they had a sexual interest in group sexual experiences was 35.6% males and 16.3% for females (Hunt, 1974; Wilson, 1987; Davidson,

PAGE 100

100 1985; Hellesund, 1976; Person, 1989, Sue, 1979). Slightly higher prevalence rates were found in this study for both as 77.8% of males and 34.3% of females indicated that they had sexual interests in a group sexual experience with one male and more than one female. The sexual interest toward a group sexual experience with one female and mor e than one male was reported by 47.6% of males and 32.7% of females. Other deviant sexual scenarios where a considerable percentage (over 20%) of participants reported having arousal toward were a) heterosexual anal sex 44.1%, b) sexual relations with a stranger 51.8%, c) bondage 60.5%, d) sex in public places 77.9% f) homosexual cunnilingus 41.5%, e) exhibitionism 22%, h) voyeurism 41.5%, i) sadism and/or masochism 26.9%, j) sexual relations with partner as audience 33.8%, k) adultery 26.8%, and l) frotteurism 23.2%. These findings support the notion that sexual deviance does, in fact, exist in non-sex offender populations. Co occurrence of deviant sexual interests Individuals also reported the cooccurrence of several deviant sexual interests. On average, they endorsed having been aroused by 5 (or 22%) of those 24 deviant sexual scenarios listed. This data was similar to the results of the pilot study where on average participants endorsed having been aroused by 13 of those 37 deviant sexual scenarios. Past research associated with both sex offenders reported similar findings. Abel, Becker, CunninghamRathner, Mittelman, & Rouleau (1988) found that 61.1% of pedophiles with female victims had three or more additional paraphilias, while 54.2% of those with male victims had three or more additional paraphilias. Among the rapists in Abel et al. s study, 55.5% had three or more additional paraphilias. Deviance Severity and Sexual Interest The perception of deviance severity seems t o influence the likelihood that participants experience or are willing to endorse sexual interests. In regards to frequency of interest, there was a negative relationship between deviance severity and sexual interest, r = -.87, p < .01

PAGE 101

101 (Figure 4.1). In r egards to lifetime prevalence of interest, there was a negative relationship between deviance severity and sexual interest, r = -.84, p < .01 In other words, the more a scenario is perceived as deviant the less likely people are going to endorse having e xperienced that sexual interest. One can only speculate that if there were no reinforcements for possessing deviant sexual interests, most would be interested in only the most non-deviant sexual scenarios. This would be true because there are harsh and s wift punishments associated with many deviant sexual interests. However, because reinforcers related to sexual deviance do exist they inspire many to become interested in extremely deviant sexual scenarios. This presence of both reinforcers and punishments for sexual deviance probably forces individuals to perpetually conduct cost/reward analyses associated with sexual interests. Sexual Deviance Inventory The data is congruent with these findings. These differences in interest have likely influenced by sexual scripts and sex role stereotypes that are taught to men and women (Gagnon & Simon, 1973). These finding are congruent research indicates that males have shown more interest in deviant sexual scenarios and sex in general when compared to women (Janus & Janus, 1993; Kinsey, 1948; Masters & Johnson, 1966). Sociobiological theory suggests that hormonal differences may be the physical mechanism through which evolutionary forces govern the difference in frequency of sexual fantasies between men and women (Ellis & Symons, 1990; Kinsey et al.,1953). Studies have shown that a large difference in androgen (testosterone) levels exists between men and women when they reach sexual maturity and that testosterone can influence frequency of sexual fantasy. Before puberty male and female testosterone levels are similar, but afterward there is about a 10 -fold to 20fold increase in male levels and only a doubling of female levels (Udry, Talbert, & Morris, 1986).

PAGE 102

102 Biology aside, women and men in Western cultures are also socialized differently about sex (Gagnon & Simon, 1973). The traditional message transmitted to women has been to be wary of sexual overtures by men and to inhibit sexual responsivity unless it is part of a committed relationship. Women are encouraged to be cautious to avoid unwanted pregnancy and to avoid acquiring a loose reputation. Part of being masculine is sexual success ( scoring), and part of being feminine is to limit sexual accessibility to the most desirable partner (make a good catch ). Everyone knows what is meant by a good girl in regard to sex, but what is the parallel for boys? Men who are sexually active with many women might be considered studs, whereas women in the same situation would more likely be viewed as sluts. If given the chance to have sex with a physically attractive stranger, men traditionally are much more likely to see this as an opportunity and women as a danger. There was a mild correlation between participants extent of conservativeness and the SDO. Political attitude can be conceptualized as a factor that influences a persons personal definition. For example, if a person adheres to conservative values then there is a good chance that person will uphold conservative sexual values as well. Politically conservative values frequently admonish sexual behaviors and attitudes that are more permissive of sexual deviances. The data supports this notion as participants that rated themselves as being more conservative were likely to have scores on the SDI. The degree to which a person identifies as being religious was also related to the SDI, (i.e. the more religious a person reports to be the less likely they are to report sexual deviance) Interestingly, there were no differences in SDI scores between the religious categories measured. It seems that religiosity restricts a persons sexual deviance regardless of the specific denomination.

PAGE 103

103 Social Learning Theory and Sexual Deviance Based on its success in explaining other forms of cultural deviance, social learning t heory was used to determine the individual proclivity toward sexual deviance. The theory was an especially good fit for this study because it built on previous works that used considered the role learning mechanism, such as reinforcement, punishment, and modeling, play in determining sexual behavior (Laws & Marshall, 1991; ODonahue & Plaud, 1994). At first the idea that society rewarded deviant sexual interests above and beyond other nondeviant sexual interests was difficult to fathom. Criminal and soc ial penalties associated sex crimes, paraphilias, and sex offender registration had been well documented throughout the sexual deviance literature (Fortney, Levenson, Bannon, & Baker 2007). Negative consequences for these offenses include, but are not limited to, criminal prosecution, mandated treatment, community notification, and social condemnation. Other negative experiences, more subtle in nature, include anxiety, harsh interpersonal interactions as well as public and interpersonal embarrassment. In contrast, any reinforcement for the experience of sexual deviance has not been well documented. A close examination of t he social discourse associated with sexual deviance revealed the presence of reinforcers. Attitudes and reactions to sexual deviance are communicated through conversations, news broadcasts, movie scripts, music lyrics, dance moves, literature, and television commercials. It is believed that these communications have dramatically shaped sexuality in our culture. Or as Gagnon and Simon s tate, communicating the sexual does not simply shape sexuality but actually creates it (1973, p. 37). Ordinary sex is mocked as being routine and mundane. Tell friends that your favorite sexual position is missionary style and watch their reaction. Individuals that are known to be sexually adventurous are thought of as exciting and reveredup to a point. Taken too far and a person will suffer

PAGE 104

104 social condemnation. Those labeled as perverted or promiscuous are not seen as desirable mates and risk public condemnation. Based on the results from this study, social learning theory can be extended to account for sexual deviance. The social learning theory measures accounted for a large percentage of the variance in sexual deviance scores as determined by the multiple regression analyses. Based on these findings, significant relationships exist between sexual deviance and all of the SLT variables. Differential Association There was a significant positive correlation between the SDI and the DAS, r = .53, p < .01. In other words, people are more likely to have deviant sexual interests if they have naughty friends. It is assumed that many people may have to draw their own conclusions about the sexual deviance of their associations. Because much of sexual cond uct occurs in privacy behind closed doors it is impossible to verify others proclivities for deviant sexual interests. It is also common for individuals not to disclose sexual information. In this absence of information, people are likely to rely on more subtle cues, (e.g., number of partners, vocation, age, dress, physique, etc). The Internet has provided a way for people to associate themselves with like minded individuals. Cyber communities range from conservative groups looking for relationships within the confines to marriage to sexual predators seeking out underage victims. Taking into account the strength of the relationship between sexual deviance and differential association, the grouping of sexual predators within sexual offender treatment pr ograms also becomes a concern. Differential Reinforcement Peers There was a significant positive correlation between the SDI and the DRPe, r = .46, p < .01. In other words, people are more likely to have deviant sexual interests if their friends

PAGE 105

105 reward them for acting naughty. Since most sexual behaviors occur in private, the conveyance of deviant sexual interests will likely occur through the recall of past experiences or personal disclosures about interest. The personal reinforcements for sexual deviance can be subtle rewards such as personal attention or laughter. Other reinforcements can be more socially beneficial if peers are willing communicate approval through the expression of high regard and introductions. Individuals may also express disapproval for individuals that express their proclivity for deviant sexual interests. This may take the form of awkward silences, public condemnation, or utter avoidance. Individuals must also pay attention to the context that they communicate their level of sexual deviance. Since society is extremely sensitive to sexual discourse, it is likely that a personal disclosure may cause anxiety. For example, attempting to communicate sexual information to one person may come across quite different when it is communicated to a crowd. Differential Reinforcement Physiology There was a significant positive correlation between the SDI and the DRPh, r = .50, p < .01. In other words, people are more likely to have deviant sexual interests if they are experience physiological rewards for acting naughty. These reinforcers may include but not be limited to, the endorphin response associated with sensation seeking and the opponent process reaction, strength and frequency of orgasms. Conversely, aversive experiences related to embarrassment and anxiety may reduce a persons proclivity for deviant sexual interests. Reinforcement Balance There was a significant positive correlation between the SDI and RB, r = .69, p < .01. In other words, people are more likely to have deviant sexual interests if, on the whole, acting naughty equates to a positive experience. Sexual deviance was more influenced by RB than any of the other SLT variables. The relative strength of RB amongst the SLT variables has been

PAGE 106

106 found several other studies examining matters of social deviance. It seems that the proclivity toward deviance is highly influenced by the overall consequence of having that proclivity. People are thought to balance out the positive and negative consequences and make decisions based on the outcome. For sexual deviance, legal consequences, social condemnation, and guilt may be among the strongest deterrent; whereas, sexual pleasure, excitement, and being regarded as sexy or experienced may be among the strongest incentive. Mo deling There was a significant positive correlation between the SDI and MOD, r = .50, p < .01. In other words, people are more likely to have deviant sexual interests if they have been exposed to people or depictions of people acting kinky or naughty. Def ini tion There was a significant positive correlation between the SDI and DEF, r = .48, p < .01. People are more likely to have deviant sexual interests if define naughtiness in more positive terms. When referring to a persons sexuality this naughtine ss is generally conceived as positive attributes. It appears that our society may revere those that are sexually adventurous similar to how thrillseekers or risk takers are admired. Limitations This study was performed on a college sample so the ability to generalize from this population should be called into question. Also, a lthough the overall sample size was sufficient for both phases of this study, there were not a substantial number of male participants. This negatively affected some of the statistics conducted that were specific to the male gender. One reason for the restricted male sample was that the study utilized the Psychology Participant Pool to obtain a majority of its participants. At the University of Florida, psychology majors are predominantly female and this consequently reflected in our sampling. Another reason for lack

PAGE 107

107 of males is that a substantial portion of our participants for the second phase of our study was recruited from a large online course that focused on eating disorders. After examining the class roster it was evident that an overwhelming majority of the class roster was female. This unequal gender split may have caused some of our statistics to be less confident, especially when they involved examining potential gender differences. As has been mentioned throughout the paper, it is currently not the norm to examine sexual deviance as a generalized construct. The other obstacle is that almost all of the scales designed to assess sexual deviance have been made specif ically for sex offenders. Although there were several issues that detracted from the value of the Sexual Interest Card Sort Questionnaire Revised (SICSQ -R), in the end, the choice to administer the scale. On the positive side, the SICSQ -R did give individuals a global score related to sexual deviance. However, the questionnaire was designed to be used specifically for sex offenders and it examines only a restricted range of interests confined to the most extreme forms of sexual deviance. Several scale items were so extreme that the investigators opted to exclude them from the scale in anticipation of IRB reactions Furthermore, because of how the questions are phrased, the SICSQ -R is only appropriate for males.

PAGE 108

108 CHAPTER 6 CONCLUSION In contrast to the prevailing conceptualization that suggests sexual deviance is pathological and confined to specific interests, this study examined sexual deviance as the overall proclivity toward deviant sexual interests that occurs in nearly everyone. A new scale designed to capture this construct, the Sexual Deviance Inventory, revealed highly significant findings in a realm that has been notoriously void of empirical data. It appears that sexual interests, as defined by behavior, fantasy, pornography and masturbation, are all influenced by our social learning processes. In other words, it is our peers associations, anticipated experiences, and permissive attitudes toward naughty sexual experiences that determine our degree of sexual deviance. It is the intention of this research to not only explain innocuous forms of sexual deviance, such as fantasizing about a threesome, but to also explain more consequential forms of sexual deviance, such as the commission of pedophilia and rape.

PAGE 109

109 APPEND IX A PILOT STUDY A pilot study for this dissertation was conducted by Bowman and Schneekloths (1999) study at a school in southeastern Texas. The study involved (A) the construction of a scale that was similar in nature to the Thurstone scale constructed in this investigation, (B) an independent (i.e., dichotomous) evaluation of the surveys sexual scenarios that determined which scenarios were considered sexually non -deviant and which scenarios were considered sexually deviant, (C) and the administration of the Thurstonelike scale along with several variables hypothesized to be related to the construct. Participants The three distinct samples of participants for this pilot study were recruited voluntarily from a school in southeastern Texas. The first sample consisted of 50 college students whose ages ranged from 18 to 29 years ( M = 21). The gender split for this sample was 34% males and 66% females. Research assistants distributed the participants a deck of 62 index cards with various sexual scenarios listed on them and then requested the participants to rank the cards in order from the least to the most deviant sexual scenarios. They were asked to not rank the cards according to their own beliefs but to rank them on behalf of an average citizen in their local community. These rankings were used to determine the scale weight of each sexual scenario. This task took the participants approximately 45 minutes to accomplish. The second sample of participants consisted of 64 college students whose ages ran ged from 18 to 23 years ( M = 19). The gender split for this sample was 24% males and 76% females. The participants for this portion of the study received the list of 62 behaviors ranked from least to most deviant. Participants were requested to determin e a cutoff between the non-deviant and deviant and sexual scenarios. For this task, they were asked not to establish the cutoff based on

PAGE 110

110 their own beliefs but instead to act on behalf of an average citizen in their local community. The averaged responses were used to establish this communitys deviancy standard. This task took the participants approximately five minutes to accomplish. The third sample of participants consisted of 231 college students whose ages ranged from 17 to 44 years ( M = 21). The gender split for this sample was 30% males and 70% females. The questionnaire packet included the Generalized Sexual Deviance Scale Beta (SDI ), the Arnett Inventory for Sensation Seeking (AISS), the Sexual Boredom Scale (SBS), the Sexual Situations Inventory (SSI), and the Sexual Opinion Survey (SOS), the Social Desirability Survey (SDS) and a short demographic questionnaire. Materials The SDI was an alphabetical list of 62 sexual behaviors. For each behavior, participants are asked to endorse the rat e that they have fantasized about that specific behavior ranging from never to one or more times per day. The Arnett Inventory for Sensation Seeking is a 20item scale that requires participants to respond in a Likert-type format ranging from describes me very well to does not describe me at all. A sample question reads, I like the feeling of standing next to the edge on a high place and looking down. The Sexual Boredom Scale is an 18 item scale that requires participants to respond in a Likert -type format ranging from I strongly agree to I strongly disagree. A sample question reads, Sex frequently becomes unexciting in a longterm relationship. The Sexual Situations Inventory is a 31item scale assessing sexual embarrassment. The scale also requires participants to respond in a Likert-type format ranging from I would not feel

PAGE 111

111 the least embarrassed to I would feel strongly embarrassed. A sample question reads, Suppose your lover asked you to masturbate in their presence. The Sexual Opinion Survey is a 21item scale assessing a persons sexual attitude. It has been noted that this scale is related to whether or not a person was raised in a sexually liberal environment (Fisher, Byrne, White, & Kelley, 1988). The scale requires partici pants to respond in a Likert-type format ranging from I strongly agree to I strongly disagree. A sample question reads, I think it would be very entertaining to look at erotica (sexually explicit books, movies, etc.). The Social Desirability Scale i s a 32 item true/false scale assessing a persons tendency to answer questions in a socially desirable manner. A sample question reads, I have never intensely disliked anyone. Procedures In order to alleviate possible discomfort surrounding the explicitness of the questionnaires: (a) we spent excessive time reviewing the terms confidentiality and informed consent, (b) we ensured participant that participant numbers would not assigned until data entry, (c) we isolated participants in private cubicles, (d) we ensured participants that their questionnaire responses would never be viewed while in their possession, (e) we had participants to seal their questionnaire packets in an unmarked envelope, (f) and we had participants mix their envelopes into a large box when leaving the study. After the research assistant assured the confidentiality of the participants, they lectured about the terms sexual arousal and sexual fantasy. They also explained the directions for each of the scales and encouraged the particip ants to ask questions at any time during the experiment. The task of completing the 13-page questionnaire packet lasted approximately one hour.

PAGE 112

112 Results Sexual Deviance Inventory Beta (SDI ) Scoring The list of the 62 sexual behaviors ranked by order of perceived deviancy is shown in Figure 1. The average rankings presented in Table-A.1 were used to develop a scoring criterion for the SDI However, during the study, the list of sexual scenarios in the SDI were presented in alphabetical order. The rates of peoples fantasies were not used in determining their SDI score although these values may be used in future analyses. The score for the SDI was calculated by summing the rankings of the endorsed items (if they have ever fantasized about the behavior). For example, if a participant endorsed having been aroused by 2 of the 62 sexual behaviors and those two behaviors having an weighted scores of 10.2 and 11.8 (as determined in Phase 1) the participants SDI score would be 22. Non deviant/devi ant c utoff In Phase 2, the average non deviant/deviant cutoff score for the ranked sexual scenarios rested between the scenarios Narratophilia and Maieusophilia ( M = 24.39, SD = 10). In other words, the sexual scenarios with an average transgression ranking greater than or equal to Maieusophilia were rated as deviant ( Table A 1.). Deviant sexual interest Using the deviancy cutoff established in Study 2, participants endorsed having experienced arousal toward an average of 13 of the 38 deviant sexual scenarios. Also, 97% of the participants endorsed having experienced arousal toward at least one of the deviant sexual scenarios listed in the SDI Table 1 lists the percentage of males and females that endorsed having fantasized about each of the sexual scenarios listed. In addition, there was a strong

PAGE 113

113 correlation between peoples most deviant fantasy and their overall SDI score ( r = .686, p < .0005).5 In other words, a person that aroused by an extremely deviant sexual scenario is more likely to be aroused by multiple deviant sexual scenarios. Personality c orrelates There most significant correlations were between SDI and the following scales: the AISS ( r = .442, p < .0005), the SBS ( r = .362, p < .0005), the SSI ( r = -.251, p < .0005), the SOS (r = .584, p < .0005), and the SDS ( r = -.341, p < .0005, see Figure 1). The correlations between the SDI and these scales remained robust even when controlling for SDS. Conclusions These results support to the notion that those people who experience a spe cific sexually deviant fantasy will likely possess a more generalized interest in deviant sexual scenarios. If sexually deviant interest were specific in nature then there should be no relationship between a persons most deviant fantasy and his or her SDI score. However, as we hypothesized, people who become aroused by extremely deviant fantasies are more likely to have become aroused by deviant fantasies in general. An analysis of the item responses also reveals that normal (non sex offender) populations may experience more sexually deviant fantasies than previously expected. An examination of the personality correlates revealed a positive relationship between the SDI and both the AISS and the SBS. In other words, people that have a tendency to sensation seek or be sexually bored are more likely to fantasize about sexually deviant scenarios. In addition, there was a negative relationship between the SDI and the SII. Apparently, people who have a tendency to be sexually embarrassed are not likely to fantasize about sexually deviant scenarios. The strongest relationship was between a peoples socialization experience 5 The SDI was adjusted by removing each participants most severe transgression from the SDI score.

PAGE 114

114 and their tendency to be aroused by deviant sexual scenarios. Although there was a positive relationship between a persons score on the SDI and the SDS all of the hypothesized relationships remained robust even when controlling for social desirability.

PAGE 115

115 Table A 1. SDI lifetime prevalence from Pilot Study Rank Sexual Scenarios M 1. Touching hands 1.3 86.8 91.4 90.0 2. Hugging prol onged sexual embrace 2.2 91.2 94.4 93.5 3. Massaging bare shoulder region of the recipient 3.1 91.2 94.4 93.5 4. Kissing heterosexual couple/mouth open 5.7 95.6 96.3 96.1 5. Kissing heterosexual couple/mouth closed 6.9 95.6 93.8 94.4 6. Sexual in tercourse heterosexual missionary 11.2 95.6 95.7 95.7 7. Male fondling a females breast 11.3 95.6 92.6 93.5 8. Maintaining a shaven pubic region 11.9 75.0 74.7 74.9 9. Female fondling a males genitalia 12.2 92.6 90.1 90.9 10. Female fondling own br east 12.6 91.2 74.1 79.2 11. Masturbation male fondling own genitalia for sexual arousal 14.7 63.2 74.7 71.4 12. Male fondling a females genitalia 15.3 95.6 92.0 93.1 13. Masturbation female fondling own genitalia for sexual arousal 15.4 95.6 6 7.3 75.8 14. Sexual intercourse heterosexual rear entry (vaginal) 15.5 95.6 90.1 91.8 15. Fellatio female providing oral stimulation to a males genitalia 19.5 95.6 88.9 90.9 16. Miscegination sexual relations between different racial groups 21.0 79.4 56.8 63.6 **17. Sitophilia using food for sexual purposes 21.0 85.3 79.0 81.0 18. Cunnilingus male providing oral stimulation to a females genitalia 21.5 95.6 93.8 94.4 **19. Podophilia attending to a persons feet for sexual pleasure 21.9 45.6 32.1 35.9 20. Simultaneous oral sex between a heterosexual couple 69 position 22.4 94.1 93.2 93.5 21. Maieusophilia sexual relations with or as a pregnant woman 24.3 52.9 38.3 42.9 22. Narratophilia stating or hearing obscene language for sexual gratification 24.6 75.0 54.3 60.6 23. Obscene phone calls vocalizing obscenities toward an unconsenting recipient 25.6 13.2 11.7 12.1 24. Using objects for sexual gratification, (i.e., dildos, vibrators, foreign objects, etc.) 26.1 75.0 68.5 70.6 DEVIANCY DETERMINATION 25. Anal sex heterosexual 28.0 76.5 38.3 49.4 **26. Bondage restricting a partners movement for sexual gratification 29.0 79.4 78.4 78.8 27. Xenophilia sexual relations with a stranger 29.5 83.8 58.0 65.8 28. Kis sing female homosexual couple/mouth closed 30.6 80.9 45.7 56.3 **29. Exhibitionism exposing self for sexual gratification 33.0 44.1 46.3 45.9 30. Sexual relations with the physically abnormal, (i.e., obese, midgets, amputees, etc.) 31.0 14.7 8.6 10 .4 31. Group sex one male and more than one female 31.7 85.3 52.5 62.3 32. Kissing female homosexual couple/mouth open 32.3 83.8 44.7 56.5 33. Agorophilia sexual behaviors in public places 33.8 91.2 83.3 85.7 34. Scopophilia observing a peo ple engage in sexual acts watched are aware of observed 34.3 79.4 62.3 67.5 35. Kissing male homosexual couple kissing/mouth closed 35.0 17.6 13.6 14.7 **36. Voyeurism observing people for sexual gratification the watched unaware of obs. 36.2 83.8 60.5 67.5 37. Group sex one female and more than one male 36.3 60.3 58.0 58.9 38. Kissing male homosexual couple kissing/mouth open 37.3 16.2 13.6 14.3 39. Juvenilism treating or being treated as infant, child, or adolescent for sexual grati fication 38.0 32.4 24.7 27.3

PAGE 116

116 Table A 1. Continued Rank Sexual Scenarios M 40. Cross generational sexual relations sexual relations between the elderly and young 38.1 29.4 16.7 20.3 41. Somnophilia sexual relations with a sleeping person 38 .4 54.4 42.0 45.9 42. Triolism having sexual relations with another while your partner is present 39.5 58.8 40.1 45.9 43. Cunnilingus female providing oral stimulation to another females genitalia 40.1 77.9 46.9 56.3 **44. Sadism/Masochism in flicting or receiving pain for sexual gratification 40.5 61.8 39.5 46.3 45. Adultery sexual relations when at least one of participants is married to someone else 40.5 73.5 40.7 50.6 46. Non physically coercive sex forcing relations through bribe ry, threats, harass, etc. 40.7 41.2 22.8 28.6 47. Fellatio male providing oral stimulation to another males genitalia 42.8 19.1 19.8 19.9 48. Klismaphilia administering or receiving an enema for sexual gratification 43.1 14.7 9.9 11.3 **59. Post pubescent pedophilia relations with child has reached puberty but under age consent 43.5 50.0 15.4 25.5 **60. Transvestisism sexual relations with or as a cross dresser 46.3 7.4 8.0 7.8 61. Anal sex male homosexual 47.0 20.6 16.0 17.3 62. Sexual relations between two children 47.2 10.3 6.2 7.4 **63. Urophilia urinating or being urinated on for sexual pleasure 47.5 10.3 4.9 6.5 64. Mid operative transexualism sexual relations with a person who possesses both a penis and breasts 48.0 7.4 6.2 6.5 **65. Coprophilia defecating or being defecated on for sexual pleasure 51.5 14.7 10.5 11.7 66. Incest between two siblings 51.9 25.0 10.5 14.7 **67. Pre pubescent pedophilia sexual relations with a child that has not yet reached pub 53.2 8.8 2. 5 4.3 **68. Zoophilia sexual relations with an animal 54.5 8.8 6.2 6.9 69. Physically coercive sex the use of physical force to induce sexual relations 56.5 44.1 29.6 34.2 **70. Parent/child incest sexual relations between a parent and his or he r child 57.2 11.8 4.3 6.5 **71. Necrophilia sexual relations with the deceased 58.8 2.9 1.2 1.7 M Mean ranks for community standards of sexual deviance rankings Significant differences between males and females calculated by independent t tests, p<.005 **Acting out these behaviors is diagnosable as a mental disorder according to the DSM IV TR

PAGE 117

117 APPENDIX B SURVEY MATERIALS : PHASE I

PAGE 118

118

PAGE 119

119

PAGE 120

120

PAGE 121

121

PAGE 122

122

PAGE 123

123

PAGE 124

124

PAGE 125

125

PAGE 126

126

PAGE 127

127

PAGE 128

128

PAGE 129

129

PAGE 130

130 APPENDIX C SURVEY MATERIALS: PH ASE II

PAGE 131

131

PAGE 132

132

PAGE 133

133

PAGE 134

134

PAGE 135

135

PAGE 136

136

PAGE 137

137

PAGE 138

138

PAGE 139

139

PAGE 140

140

PAGE 141

141

PAGE 142

142

PAGE 143

143

PAGE 144

144

PAGE 145

145

PAGE 146

146

PAGE 147

147

PAGE 148

148

PAGE 149

149

PAGE 150

150

PAGE 151

151

PAGE 152

152

PAGE 153

153

PAGE 154

154

PAGE 155

155

PAGE 156

156

PAGE 157

157

PAGE 158

158

PAGE 159

159

PAGE 160

160

PAGE 161

161

PAGE 162

162

PAGE 163

163

PAGE 164

164

PAGE 165

165

PAGE 166

166

PAGE 167

167

PAGE 168

168

PAGE 169

169

PAGE 170

170

PAGE 171

171

PAGE 172

172

PAGE 173

173

PAGE 174

174

PAGE 175

175

PAGE 176

176

PAGE 177

177

PAGE 178

178

PAGE 179

179

PAGE 180

180

PAGE 181

181

PAGE 182

182

PAGE 183

183

PAGE 184

184

PAGE 185

185

PAGE 186

186 RE FERENCES AVN. (2006). State of the U.S. adult industry. Adult Video News, 22, 30-31. Abel, G. G, Becker, J., Mittelman M., CunninghamRathner, J., Rouleau, J., & Murphy, W. (1987). Self reported sex crimes on nonincarcerated paraphiliacs. Journal of I nterpersonal Violence, 2(1) 3-25. Abel, G. G. & Becker, J. V. (1985). Sexual interest card sort. Unpublished. Abel, G. G., Becker, J. V., Cunningham-Rathner, J., Mittelman, M, & Rouleau, J. L, (1988). Multiple paraphilic diagnoses among sex offenders. The Bulletin of the American Academy of Psychiatry and the Law, 16, 153-168. Abel, G. G., Jordan, A., Hand, C. G., Holland, L. A, & Phipps, A. (2001). Classification Models of Child Molesters Utilizing the Abel Assessment for Sexual Interest. Child Abuse & Neglect, 25, 703 -718. ACE Study Prevalence Adverse Childhood Experiences (2005). Retrieved June 10, 2009, from http://www.cdc.gov/nccdphp/ace/prevalence.htm Advisory Commission on Sentencing (2002). 2002 Annual Report of the District of Columbia Advisory Commission on Sentencing. Retrieved July 10, 2009, from http://acs.dc.gov/acs/lib/acs/pdf/2002_Forward.pdf Akers, R. L. & Cochran, J. K. (1985). Adolescent marijuana use: A test of three theories of deviant behavior, Deviant Behavior 6:323-346 Akers, R. L. (1985). Deviant Behavior: A Social Learning Approach. (3rd Ed). Belmont, CA: Wadsworth. Akers, R. L. (1998). Social Learning and Social Structure: A General Theory of Crime and Deviance. Boston, MA: Northeastern University Press. Akers, R. L., Krohn, M. D., Lanza-Kaduce, L., & Radosevich, M. (1979). Social learning and deviant behavior: a specific test of a general theor y, American Sociological Review, 44, 635-655. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders: DSMIV TR. Washington, DC: Author. American Psychological Association. (2002). American Psychological Associati on ethical principles of psychologists and code of conduct. Retrieved February 9, 2009, from http://www.apa.org/ethics/code2002.html Arndt, W. B., Foehl, J. C., & Good, F. E. (1985). Specific sexual fa ntasy themes: A multidimensional study. Journal of Personality and Social Psychology, 48, 472-480.

PAGE 187

187 Arnett, J. (1994). Sensation seeking: a new conceptualization and a new scale. Personality of Individual Differences, 16, 289. Attorney Generals Commission on Pornography (1986) Retrieved April 29, 2004, from Web site: http://www.afa.net/ pornography/attygenrpt.html AVN. (2006, January). State of the U.S. adult industry. Adult Video News, 22, 30-31. Aylwin, A. S., Clelland, S. R., Kikby, L., Re ddon, J. R., Studer, L. H., & Johnston, J. (2000). Sexual offense severity and victim gender pr eference: A comparison of adolescent and adult sex offenders. International Journal of Law and Psychiatry 23, 113 Aylwin, A. S., Reddon, J. R., & Burke, A. R. ( 2005). Sexual fantasies of adolescent male sex offenders in residential treatment: A descriptive study. Archives of Sexual Behavior, 34 (2), 231-239. Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall. Barbaree, H. E., & Seto, M. C. (1997). Pedophilia: Assessment and treatment. In D. R. Laws & W. O'Donohue (Eds.), Sexual deviance: Theory, assessment, and treatment (pp. 175193). New York: Guilford Press. Barbaree, Laws, and Marshall (1990) Handbook of sexual assault: I ssues, theories and treatment of the offender. New York: Plenum Press. Bessette, L., Sangha, O., Kuntz, K. M., Keller, R. B., Lew, R. A., Fossel, A. H. & Katz J. A. (1998). Comparative responsiveness of generi c versus disease-specific and weighted versus unweighted health status measures in Carpal Tunnel Syndrome. Medical Care 36, 491-502 Boeringer, S. (1992). Sexual coercion among college males: Assessing three theoretical models of coercive sexual behavior. Ph.D. Dissertation. Univ ersity of Florida. Boeringer, S., Shehan, C.L., & Akers, R. L. (1991) Social contexts and so cial learning in sexual coercion and aggression: Assessing the c ontribution of fraternity membership, Family Relations, 40, 558-564. Bowman, J. Z., & Schneekloth, P. (1999). Sexual transgression inventory: Measuring a generalized arousal toward sexual transgressions. Poster presented at the 2000 American Psychological Society Confer ence, Miami, Florida. Briere, J. (1989) University males sexual intere st in children: Predicti ng potential indices of Pedophilia in a nonforensic sample. Child Abuse & Neglect, 13, 65-75.

PAGE 188

188 Briere, J., & Runtz, M. (1989). The Trauma Symptom Checklist (TSC-33): Early data on a new scale. Journal of Interpersonal Violence, 4, 151-163. Bureau of Justice Statistics (2007). Criminal Justice Statistics. Retrieved July 9, 2009, from U.S. Department of Justice Web site: http://www.ojp.usdoj.gov/bjs /crimoff.htm#sex Burgess, R. L., & Akers, R. L. (1966). A differential associationreinforcement theory of criminal behavior. Social Problems, 14, 128-147. Burt, M. (1980). Cultural myths and supports for rape. Journal of Personality and Social Psycholo gy, 38(2), 217-230. Camara, J. C. & Echeternacht, G. (2000). The SAT I and high school grades: Utility in predicting success in college. The College Board, 10(7), 1-12. Coates, V. E. (1995). Measuring constructs accurately: a prerequisite to theory testing. Journal of Psychiatric and Mental Health Nursing, 2(5), 287-293. Cooper, A. (1998). Sexuality and the Internet: Surfing into the new millennium. CyberPsychology and Behavior, 1 (2), 187-194. Crepault, C., & Couture, M. (1980). Men's erotic fantasie s. Archives of Sexual Behavior, 9 565 581. Critelli, Joseph W. & Bivona, Jenny M. (2008). Women's erotic rape fantasies: An evaluation of theory and research. Journal of Sex Research, 45 (1), 57-70. Crowne, D. P. and Marlowe, D. (1960), A new scale of s ocial desirability independent of psychopathology, Journal of Consulting Psychology, 24, 349-354. Danebeck, K., Cooper, A., Mansson, S. (2005). An internet study of cybersex participants. Archives of Sexual Behavior, 34 (3), 321-328, Darkness to Light (n.d.). Statistics, prevalence and consequences of child sexual abuse. Retrieved July 10, 2009, from http://www.darkness2light.org/KnowAbout/statistics_2.asp Davidson, J. K. (1985). The utilization of sexual fantasies by sexually experienced university students. Journal of American College Health, 34, 24-32. Dorr, D. (1998). Psychopathy in pedophiles. In T. Millon, E. Simonsn, M. Birket-Smith, & R. D. Davis (Eds.), Psychopathy: Antisoc ial, criminal and violent behavior (pp. 304-320). New York: Guilford Press. DUdine, B. (1990). The modification of sexual behavior through imprinting: A rodent model. In J. R. Feierman (Ed.), Pedophilia: Biosocial dimensions. New York: SpringerVerlag.

PAGE 189

189 Eccles, A., Marshall, W. L, & Barbaree, H. E. (1994). Differentiating rapists and non-offenders using the rape index. Behaviour Research and Therapy, 32, 539-546. Eskridge, W. N., & Hunter, N. D. (1997). Sexuality, Gender and the Law. New York: The Fou ndation Press. Evans, D. R. (1968). Masturbatory fantasy and sexual deviation. Behaviour Research and Therapy 6, 17. Felitti, V.J., Anda, R.F., Nordenberg, D., Williamson, D.F., Spitz, A.M., Edwards, V., Koss, M. P., & Marks, J.S., 1998. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The adverse childhood experiences (ACE) study. American Journal of Preventative Medicine, 14, 245. Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: Free Press Finkelhor, D., Hotaling, G., Lewis, I. A. & Smith, C. (1990). Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk factors. Child Abuse & Neglect, 14, 19-28. Fisher, W. A., Byrne, D. White, L. A., Kelley, K. (1988). Erotophobia-erotophilia as a dimension of personality. The Journal of Sex Research, 25, 123-151. Flora, R. (2001). How to work with sex offenders: A handbook for criminal justice, human service, and mental health professionals. Binghamton, NY: The Hawthorne Clinical Practice Press. Fortney, T., Levenson, J., Brannon, Y., and Baker, J. (2007). Myths and facts about sexual offenders: Implications for treatment and public policy. Sex Offender Treatment, 2, 1-17. Freud, S. (1962). Fetishism. In J. Strachey (Ed. And Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 3), London: Hogarth Press. (Original work published 1927) Greene, W. (2006). Censored data and truncated distributions. Retrieved July 10, 2009. http://archive.nyu.edu/bitstream/2451/26101/3/5-8.pdf.txt Grissom, R. J., and Kim, J. J. (2005). Effect sizes for research: A broad practical approach. Lawrence Erlbaum Associates: New Jersey. Groseclose, T. (2006). Extreme sample selection bias: Conditions that cause the correlation between two variables to switch signs Retrieved, September 11, 2009 from Web site: http://www.sscnet.ucla.edu/polisci/faculty/groseclose/Working.Papers/ext.samp.pdf Hanson, R. K., Gordon, A., Harris, A. J. R., Marques, J. K., Murphy, W., Quinsey, V. L., & Seto, M. C. (2002). First report of the collaborative outcome data project on the Sex Offender

PAGE 190

190 Treatment and Evaluation Project (SOTEP). Sexual Abuse: A Journal of Research & Treatment, 17(1), 79-107. Hayes, M. (2006) Girls Gone Wild Pleads Guilty to 2257 Violations. Retrieved April 29, 2007, from http://www.xbiz.com/news_piece.php?id17123. Hessellund, H. (1976). Masturbation and sexual fantasies in married couples. Archives of Sexual Behavior, 5, 133-147. Hickey, E. W. (2006). Sex crimes and paraphilia Upper Saddle River, New Jersey: Pearson Education. Holland, L., Zolondek, S. C., Abel, G. G., Jordan, A. D. & Becker J. V. (2000). Psychometric analysis of the sexual interest cardsort questionnaire. Sexual Abuse: A Journal of Research and Treatment, 12, 107-122. Holmes, R. M. & Holmes, S. T. (2001). Sex crimes: Patterns and behavior (Second Ed.). New York: Sage. Hooper, P. (2008). ACLU of NV wins permanent injunction against retroactive enforcement of new sex offender la ws. Retrieved July 9, 2009, from ACLU: American Civil Liberty Union of New Jersey Web site: http://www.aclunv.org/aclunv wins perma nentinjunction-againstretroactive -enforcement-newsex -offenderlaws Hudson, Z. (2009) Sample size, power, and effect size What all researchers need to know Physical Therapy in Sport, 10, 4344. Hudson, S. M., & Ward, T. (1997). Intimacy, loneliness, and attachment style in sexual offenders. Journal of Interpersonal Violence, 12, 323-339. Hunt, M. (1974). Sexual behavior in the 70's. Chicago: Playboy Press. Jackson, C. (2009). Truncated normal distribution. Retrieved July 10, 2009 from Web site: http://rss.acs.unt.edu/Rdoc/library/msm/html/tnorm.html Jacobs, D. (2009). Why sex offender laws do more harm than good. Retrieved July 9, 2009, from ACLU: American Civil Liberty Union of New Jersey Web site: http://www.aclunj.org/issues/criminaljustice/whysexoffenderlawsdomoreha.htm Janus, S. S., & Janus, C. L. (1993). The Janus report on sexual behavior. New York: Wiley. Johnson, T. P., Fendrich, M. & Hubbell, A. (2002). A validation of the Crowne-Marlowe Social Desirability Scale. Paper presented at the 57th Annual Meeting of the American Association for Public Opinion Research, St. Pete Beach, FL.

PAGE 191

191 Joyal, C. C., Black, D. N., & Dassylva, B. (2007). The neuropsychology and neurology of sexual deviance: a review and pilot study. Sex Abuse, 19(2), 155-173. Kalichman, S. C., Henderson, M. C., Shealy, L. S, & Dwyer, M. S. (1992). Psychometric propertie s of the Multiphasic Sex Inventory in assessing sex offenders, Criminal Justice and Behavior, 19, 384-396. Kalmbach, K. C., & Lyons, P. M. (2002). Ethical and legal standards for research in prisons. Behavioral Sciences and the Law, 21, 671-686. Katz, R. C. (1990). Psychosocial adjustment of child molesters. Child Abuse and Neglect, 14, 567-575. Kessler, R. C. (2002). The categorical versus dimensional assessment controversy in the sociology of mental illness. Journal of Health and Social Behavior, 43, 171 -178. Kettner, C. (2007). Penthouse purchases Adult Friend Finder Retrieved July 10, 2009, from, Domain Name News Web site: http://www.domainnamenews.com/domain sales/penthousepurchases-adultfriend -finder/1340 Kinsey, A., Pomeroy, W., & Martin, C. (1948). Sexual behavior in the human male. Philadelphia, PA: W. B. Saunders. Kite, M. E. & Whitley, B. E. (2003). Do heterosexual women and men differ in their attitudes toward homosexuality? A conceptual and methodological analysis. In L. D. Garnets and D. C. Kimmel (Eds.) Psychological Perspectives on Lesbian, Gay, and Bisexual Experiences (2nd ed., pp. 165-187) New York: Columbia University Pres. Kru eger, R. F., Watson, D., & Barlow, D. H. (2005). Introduction to the Special Section: Toward a Dimensionally Based Taxonomy of Psychopathology. Journal of Abnormal Psychology, 114(4), 491. Lane, D. (2009). Hyperstat online statistics text book. [Elec tronic version]. Retrieved January 27, 2009, from http://davidmlane.com/hyperstat/index.html Langevin, R., Paitich, D., Russon, A., Handy, L., & Langevin, A. ( 1990). Clarke Sex History Questionnaire for Males manual. Etobicke, Canada: Juniper Press Laws, D. R., & Marshall, W. L. (1991). Masturbatory reconditioning with sexual deviates: An evaluative review. Advances in Behavior Research and Therapy, 13, 1325. Laws, D. R., & ODonohue, W. (1997). Sexual deviance: Theory, assessment, and treatment. New York: Guilford Press. Levenson, J. S., & Cotter, L. P. (2005). The effect of Megans Law on sex offender reintegration. Journal of Contemporary Criminal Justice, 21(1), 49-66.

PAGE 192

192 Levesque, J. R. (2000). Adolescents, sex, and the law: Preparing adolescents for responsible citizenship. American Psychological Association: Washington, DC. Lonsway, L. A., & Fitzgerald, L. F. (1994). Rape myths: In review. Psychology of Women Quarterly, 18, 133-164. Malamuth N. M. (1986). Predictors of naturalistic sexual aggression. Journal of Personality and Social Psychology, 50, 953-962. Marques, J. K., Wiederanders, M., Day, D. M., Nelson, C., & Van Ommeren, A. (2005). Effects of a relapse prevention program on sexual recidivism: Final results from California's effectiveness of treatment for sex offenders. Sexual Abuse: A Journal of Research and Treatment, 14(2), 169-194. Marshall, W. L. (1997). Introduction: Fundamental issues in sexual deviance. In D. R. Laws, & W. O'Donohue (Eds.), Sexual deviance: Theory, assessment, and treatment. (pp. 152174). New York: The Guilford Press. Marshall, W. L., Barbaree, H. E., & Eccles, A. (1991). Early onset and deviant sexuality in child molesters. Journal of Interpersonal Violence, 6, 323-336. Martin, J. L. (2003). What is field theory? American Journal of Sociology, 109, 1. McGrath, R. J. & Hoke, S. E. (2001). Vermont assessment of sex offender risk manual. Retrieved July 1, 2009 from http://www.csom.org/pubs/VASOR.pdf McIver, J., & Carmines, E. G. (1981). Unidimensional Scaling. Newberry Park, CA: Sage Publications. Masters, W. H., & Johnson, V. E. (1966). Human Sexual Response New York: Bantam Books. McGrath, R. J., Cumming, G. F., & Bouchard, B. L. (2003). Current practices and trends in sexual abuser management: The Safer Society 2002 Nationwide Survey. Brandon, VT: Safer Society Press. McGuire, R. J., Carlisle, J. M., & Young, B. G. (1965). Sexual deviations as conditioned behaviour: A hypothesis. Behaviour Research and Therapy 3, 185. Med India Health Network (2009) Sexual deviance. Retrieved July 10, 2009, from Med India: Networking for Health Web site: http://www.medindia.net/patients/patientinfo/sexualdeviance.htm Miller v. California 413 U.S. 15, 24 (1972).

PAGE 193

193 Miller, P. Y, & Simon, W. (1980). The development of sexuality in adolescence. In J. Abelson (Ed.), Adolescent psychology (pp. 383407). New York: Wiley. Miller, R (1996). Embarrassment: Poise and peril in everyday life. New York: Guilford Press. Monahan, J., & Walker, L. (1998). Social science in law. New York: The Foundation Press. Moser, C. & Kleinplatz, P.J. (2005) DSM IV -TR and the Paraphilias: An argument for removal. Journal of Psychology and Human Sexuality, 17, 91-109. MPAA. (2006). Research & statistics. Retrieved September 28, 2006 from http://www.mp aa.org/researchStatistics.asp Netemeyeer, R., Beardon, W., & Sharma, S. (2003). Scaling procedures: Issues and applications Thousand Oaks, CA: Sage. Nicholaichuk, T., Gordon, A., Gu, D., & Wong, S. (2000). Outcome of an institutional sexual offender treatment program: A comparison between treated and matched untreated offenders. Sexual Abuse: A Journal of Research & Treatment, 12(2), 139-153. Nichols, H. R., & Molinder, I. (1984) Multiphasic Sex Inventory. Fircrest, WA: Nichols & Molinder Assessments. Noll, J. G., Trickett, P. K., & Putnam, F. W. (2003). A prospective investigation of the impact of childhood sexual abuse on the development of sexuality. Journal of Consulting and Clinical Psychology 71, 575-586. OBrien, R. M. (1991). Sex ratios and r ape rates: A power control theory. Criminology, 29, 99114. ODonohue, W., & Plaud, J. J. (1994). The conditioning of human sexual arousal. Archives of Sexual Behavior, 3, 385-393. O'Sullivan, L. F., Byers, E. F., & Finkelman, L. (1998). A comparison of male and female college students' experiences of sexual coercion. Psychology of Women Quarterly, 22, 177-195. Palk, G. R. & O'Gorman, J. (2004) Habituation of deviant sexual arousal in sex offenders: The role of cognitive processes. International Journal of Forensic Psychology 1(2), 58-70. Paolucci, E.O, Genuis, M.L, & Violato, C. (2001). A meta-analysis of the published research on the effects of child sexual abuse. Journal of Psychology 135 17-36. Paul, B., & Linz, D. (2008). The effects of exposure to virtual child pornography on viewer cognitions and attitudes toward deviant sexual behavior, Communication Research, 35(1), 3-38

PAGE 194

194 Person, E., Terestman, N., Myers, W, Goldberg, E., & Salvadori, C. (1989). Gender differences in sexual behaviors and fa ntasies in a college population. Journal of Sex and Marital Therapy, 15, 187-198. Prentky, R., Knight, R, Straus, H., Rokous, F., Cerce, D., & Sims-Knight, J. (1989). Developmental antecedents of sexual aggression. Development and Psychopathology, 1, 153-169. Quinsey, V. L. & Lalumiere, M. (2002). Assessment of sexual offenders against children Thousand Oaks: CA: Sage Publications. Quinsey, V. L., & Marshall, W. L. (1983). Procedures for reducing inappropriate sexual arousal: An evaluation review. In J. G. Greer, & I. R. Stuart (Eds.), The Sexual Aggressor: Current Perspectives on Treatment (pp. 257-289). New York: Van Nostrand Reinhold Company. Raymond, N. C., Coleman, E., Ohlerking, F., Christenson, G. A., & Miner, M. (1999). Psychiatric comorbidity in pedophilic sex offenders. American Journal of Psychiatry, 156, 786-788. Roberts, C. F., Doren, D. M., & Thornton, D. (2002). Dimensions associated with assessments of sex offender recidivism risk. Criminal Justice and Behavior, 29, 569-589. Rossi, P. H., Waite, E., Bose, C. E., & Berk, R. E. (1974). The seriousness of crimes: normative structure and individual differences. American Sociological Review, 39, 224-237. Seto, M. C., Harris, G. T., Rice, M. E., & Barbaree, H. E. (2004). The Screening Scale for Pedophilic Interests predicts recidivism among adult sex offenders with child victims. Archives of Sexual Behavior, 22, (5), 455 -466. Sex crimes (n.d). Retrieved July 10, 2009, from Wikipedia Web site: http://en.wikipedia.org/wiki/Category:Sex_crimes Sexual deviance (n.d.). Retrieved July 10, 2009, from Definition Of Web site: http://www.definition -of.com/sexual+deviance Soper, D.S. (2009). The free statistics calculators website. Retrieved July 10, 2009, from Web site: http://www.danielsoper.com/statcalc/ Spear, S., Akers, R.L. (1988). Social learning variables and the risk of habitual smoking among adolescents: the Muscatine Study, American Journal of Preventive Medicine, 4, 336-348. Sutherland, E. (1947). Principles of Criminology. (4th Ed.) Philadelphia: J. B. Lippincott. Stack, S., Wasserman, I., & Kern, R. (2004). Adult social bonds and the use of Internet pornography. Social Science Quarterly, 85(1), 75.

PAGE 195

195 Stein, J. (2007). The accidental friend finder. Retrieved July 10, 2009, from CNN Web site: http://money.cnn.com/magazines/business2/business2_archive/2007/04/01/8403370/inde x.htm?section=money_technology Stinson, J. D., & Becker, J. V. (2008). Assessing sexual deviance: A comparison of physiological, historical, and self report measures Journal of Psychiatric Practice, 14, (6), 385-388. Sue, D. (1979). Erotic fantasies of college students during coitus. Journal of Sex Research, 15. 299-305. Szasz, T. (1974). The Myth of Mental Illness: F oundations of a Theory of Personal Conduct (2nd ed.). New York: Harper and Row. Templeman, T. L., & Stinnett, R. D. (1991). Patterns of sexual arousal and history in a normal sample of young men. Archives of Sexual Behavior, 20, 137-150. Tewksbury, R., & Lees, M. ( 2006). Consequences of sex offender registration: Collateral consequences and community experiences. Sociological Spectrum, 26 (3) 309-334. Thurstone, L. L. (1928). Scale construction with weighted observations. The Journal of Educational Psychology, 19, 441-453. Trochim (2006). Thurstone scaling [Electronic version]. Retrieved June 2007, 2009, from Social Research Methods Web site: http://www.socialresearchmethods.net/kb/scalthur.php Walker, W. D. (1998). Escalation in Sexual Offending. Presented at The Association for the Treatment of Sexual Abusers: 17th Annual Research and Treatment Conference, Vancouver, Canada. Wang, P. (2009). Financial Econometrics (2nd ed.). New York: Taylor & Francis. Ward, T., Hudson, S. M., & Marshall, W. L. (1996). Attachment style in sex offenders: A preliminary study. Journal of Sex Research, 33, 17. Ward, T., & Beech, A. R. (2006). An integrated theory of sexual offending. Aggression and Violent Behavior 11, 44. Ward, T., Laws, D. R., & Hudson (2003). Sexual Deviance: Issues and Controversies. Thousand Oaks, CA: Sage Publications. Watt, J. D., & Ewing, J. E. (1996). Toward the development and validation of a measure of sexual boredom. The Journal of Sex Research, 33, 57 66. Wells, M., Finkelhor, D., Wolak, J., & Mitchell, K. (2007). Defining Child Pornography: Law Enforcement Dilemmas in Investigations of Internet Child Pornography. Police Practice and Research 8 269 282.

PAGE 196

196 Whitley, B. E. (1996). Principles of research in behavioral science. Mountain View, CA: Mayfield Publishing Company. Wiederman, M.W (2003). Paraphilias and fetishism. The Family Journal, 11, 315-321. Wilson, G. D. (1978). The secrets of sexual fantasy. London: Dent. Wilson, G. D. (1987). Malefemale differences in sexual activity, enjoyment and fantasies. Personality and Individual Differences, 8, 125127 Wosnitzer, R., Scharrer, E., Bridges, A., Sun, C., & Liberman, R. (2006). Aggression and sexual behavior in best -selling pornography: A content analysis update. Retrieved July 10, 2009 Web site: http://www.dshs.state.tx.us/hivstd/conference/2008/handouts/Pre_Sexuality.pdf Young, K. (1999). Internet addiction: Symptoms, evaluation, and treatment innovations in clinical practice. In L. VandeCreek & T. L. Jackson (Eds.), Sarasota, FL: Professional Resource Press.

PAGE 197

197 BIOGRAPHICAL SKETCH Jason Bowman was born in Fairfax, Virginia in 1970. During his undergraduate years he attended the University of Florida and earned a Bachelor of Science degree in health science education in 1994. He then attended the University of Colorado and earned a masters degree in clinical psychology in 1998. He then began his doctorate work in Forensic Clinical Psychology at Sam Houston State University only to return to his alma mater in order to finish his Ph.D. in counseling psychology at the University of Florida. Over his extended graduate career, Bowman has had the chance to take coursework in several academic areas, participate in numerous research projects, and work at several respectable professional agencies. His undergraduate premedical track and graduate coursework have pr ovided him extensive training in forensic psychology, organizational psychology, and health psychology. Project involvement has proven beneficial, as his credentials include initiating a hospital-based, congestive heart failure study, coordinating a NSF study examining jury decision making, conducting program evaluations on both correctional institutions and counseling psychology graduate schools, and developing an assessment instrument that examines deviant sexual interest. The agencies he has worked for include ComCor Inc., the Colorado Department of Corrections, the National Development and Research Institute, the Texas Department of Corrections, the University of Florida Counseling Center, the Bijou [Sex Offender] Treatment Center, the National Rural Be havioral Health Center, and the Alachua County Crisis Center. Bowman began working with clients experiencing sexual issues at ComCor, Inc., a Coloradobased community correctional facility. Initially, he worked with clients in the general population but eventually devoted all his time to the agencys sex offender treatment program (SOTP). His SOTP experiences included intake assessments, penile plethysmograph

PAGE 198

198 assessments, co -facilitating SOTP interpersonal/psycho educational groups, and covert desensitization treatments. So effected by these experiences he matriculated at a forensic clinical psychology Ph.D. program in Huntsville, Texas. There he conducted research on sexual deviance with sex offenders. Eventually he opted to transfer back to the University of Florida where he could earn a more general doctorate education in psychology. His interest in sexual deviance research remained but he expanded the focus of his examination to non-sex offender populations. He began postulating the larger questions, What is sexual deviance and why does it occur? In the spring of 2010, Bowman completed his doctorate education in the University of Floridas counseling psychology program. He is currently doing his APPIC approved internship at a consortium in Nashville Tennessee where he does rotations at the Steele Neuropsychology Clinic, Meharry Medical College, and the Kelly Miller Smith Center for Domestic Violence. In the future, he hopes to earn a job placement at a correctional facility where once again he can work with sex offenders and extend his current research on sexual deviance to a correctional population.