Beauty and the Eye of the Beholder: Gender and Attractiveness Affect Judgments in Teacher Sex Offense Cases

Abstract

The present study investigated the effects of gender and attractiveness on judgments of bail requirements, incarceration, and sex offender registration lengths, and attitudes toward offenders and victims in a teacher–student sexual perpetration scenario. Researchers presented 432 undergraduate students at a large southwestern university with one of four vignettes detailing a sexual relationship between a 35-year-old teacher and a 14-year-old student. Vignettes varied by both attractiveness and gender of the offender (using heterosexual offender–victim dyads). Results indicate that both gender and attractiveness affect judgments of sex offenders; specifically, female sexual offenders were viewed more leniently and judged less punitively than male sexual offenders. Although attractive female sexual offenders were given particularly lenient treatment, attractiveness did not affect judgments toward male sex offenders. In addition, although male and female participants tended to rate male offenders similarly, male participants were more lenient toward female offenders than were female participants. Finally, post hoc analyses revealed that, for many variables, unattractive female sexual offenders may not be viewed differently from male sexual offenders. These results have serious implications for the legal system, sex offender management, and societal views regarding male and female sexual offenders and their victims.

Keywords

female sex offenders sexual offending statutory rape attractiveness gender

Introduction

Just a few short decades ago, the prevailing view of female sexual offending could be summarily expressed by Mathis’s (1972) question, “[ . . .] what harm can be done without a penis?” (cited in Deering & Mellor, 2007). Females, by virtue of their sex, were viewed as harmless and subservient, incapable of committing such acts. The reality of female sexual offending is far more sobering. While research into sex offenses committed by males has proliferated over the last 30 years (Robertiello & Terry, 2007), female sexual offenders (FSOs) have generally taken the proverbial “back seat” to their male counterparts. Encouragingly, this trend appears to be changing somewhat, with increased interest in identifying traits of FSOs, their crimes, their victims, and their rates of recidivism (Gannon & Rose, 2008; Grayston & De Luca, 1999; Johansson-Love & Fremouw, 2009). Still, there is much to be learned.

Research in the area of female sexual offending has historically been plagued by methodological complications, including small (frequently less than 15) sample sizes, clinical samples, gaps in available archival data, over-reliance on descriptive statistics, and differing definitions as to what constitutes a female sexual offender and/or sexual offense (Harris, 2010; Johansson-Love & Fremouw, 2006). Recent studies have begun to address some of these issues. In 2010, for example, Cortoni, Hanson, and Coache’s meta-analysis examined recidivism rates of 2,490 FSOs. With this large sample size, the authors were able to more accurately measure the overall rate of sexual recidivism in females, which they found is considerably lower than in males (less than 3% vs. 13.5%). Statistical methods have become increasingly more sophisticated, with researchers taking advantage of latent class analysis (Wijkman, Bijeveld, & Hendriks, 2011), latent profile analysis (Miller, Turner, & Henderson, 2009), and hierarchical loglinear modeling (Vandiver & Kercher, 2004), among others.

As research has advanced, a number of generally accepted characteristics of female sexual offending have emerged. Female perpetration against another adult is extremely rare, with the vast majority of FSOs molesting children related to or at least known to them (Harris, 2010; O’Connor, 1987; Vandiver & Kercher, 2004). In addition, while it is likely that a significant number of sexual offenses committed by females go unreported (Becker, Hall, & Stinson, 2001), it is nevertheless evident that females commit a minority of sexual offenses, with estimates typically ranging between about 1% and 8% (Center for Sex Offender Management [CSOM], 2008; O’Connor, 1987). FSOs are more likely to offend with an accomplice, frequently a male romantic or marital partner (although there is disagreement as to the exact prevalence of this co-offending; Gannon, Rose, & Ward, 2008; Harris, 2010; Mathews, Matthews, & Speltz, 1989). Some have distinguished a “male-accompanied” offense, in which the female is not under duress and willingly participates in the perpetration, from a “male-coerced” offense, in which the female commits the offense solely for the satisfaction of her partner, often under threat of violence. These male-coerced women are frequently passive and dependent on their partners, reducing their likelihood of resisting facilitation of or engagement in the abuse (Harris, 2010).

The absence of theories of female sexual offending has historically resulted in the application of male sexual offending theories to FSOs, without evidence that this was either valid or useful (Harris, 2010). More recently, researchers have sought to fill the gap in theories of female sexual offending. As with male sexual offenders (MSOs), FSOs are a heterogeneous group, making it difficult to build an over-arching framework for their criminal behaviors. The construction of descriptive typologies has served as a vital starting point for more detailed theories (e.g., Mathews et al., 1989; Nathan & Ward, 2002; Vandiver & Kercher, 2004). In 2010, Gannon, Rose, and Ward published their study of FSOs in which they found three primary pathways to female sexual offending: (a) Explicit Approach, (b) Directed Avoidant, and (c) Implicit Disorganized. Although their sample was small (n = 22) and 18% (n = 4) of the women were not able to be classified in one of the three pathways, the model represents an important advance in the push for a greater theoretical understanding of FSOs.

A substantial body of literature has examined male and female criminals within a legal context, with many studies focused on sentencing differences between the sexes. In their 1994 meta-analysis, Mazzella and Feingold found that mock jurors were more likely to find an attractive defendant not guilty than they were an unattractive defendant. In addition, attractive defendants were given more lenient sentences, particularly for the crimes of robbery, rape, and cheating (p. 1325). They also found that, while perpetrator gender did not make a difference in determination of guilt, jurors did tend to assign lesser punishments to female defendants (although the effect size was small). However, the authors did not conduct any analyses to investigate a possible interaction effect of attractiveness and gender. In examining data for the state of Florida from 1994 to 2004, Van Slyke and Bales (2013) found that female offenders were treated more leniently than males, particularly when the judge in the female’s case was male (the gender of the judge did not have an effect for male offenders). Rodriguez, Curry, and Lee’s (2006) investigation into sentencing outcomes also found evidence that females received greater leniency in the criminal justice system. In addition, research examining the combined role of gender and attractiveness in sentencing has found that although attractiveness was associated with reduced sentences for females (e.g., DeSantis & Kayson, 1997), this was not true for males (Beckham, Spray, & Pietz, 2007). Theories as to why these differences exist vary, but most focus on the role played by gender stereotypes. The focal concerns perspective posits that when rendering sentencing decisions, judges take into account “blameworthiness, dangerousness (community protection), and practical constraints” (Van Slyke & Bales, 2013, p. 169). Because judges may tend to view females as less blameworthy, less dangerous to the community, and more adversely affected in their parenting role by a lengthy incarceration, they may give women reduced sentences (Van Slyke & Bales, 2013). Other theories examine the role of chivalry and paternalism in a male-dominated judicial system, suggesting that a judge may feel more “protective” toward a female defendant than a male (Curry, Lee, & Rodriguez, 2004) and focus more on helping rather than punishing her.

Although numerous studies have addressed the effects of gender on sentencing for non-sexual crimes, there is limited research comparing legal outcomes specifically for MSOs versus FSOs. What little research has been done indicates that MSOs are likely judged more harshly. In his study, Steffensmeier (1977) asked participants to impose sentences on male or female adults charged with seduction of a minor. Participants assigned significantly more jail time to the males than to the females. In another study, Broussard, Wagner, and Kazelskis (1991) found that, given identical crimes involving sexual activity with a minor, participants were more likely to label a male perpetrator as having committed “sexual abuse” than a female perpetrator. In 2012, Embry and Lyons conducted a study examining sentencing disparities between MSOs and FSOs. Utilizing the “evil woman hypothesis” as a framework, the authors postulated that women would be sentenced more harshly for a significant violation of social norms (i.e., committing a sexual offense), and that the more violent and norm-breaking the offense, the more severe the sentencing. However, on examining 10 years’ worth of data from the National Corrections Reporting Program, the authors discovered that the opposite was true: Across all sexual offenses, women received significantly shorter sentences than men. This was true even for jurisdictions in which sentencing guidelines had been implemented.

For many individuals, knowledge of female sexual offending may be limited to sensationalized media accounts detailing the “illicit affairs” of attractive teachers having sex with their male students (CSOM, 2008; Deering & Mellor, 2007; Frei, 2008; Saletan, 2006). These teachers are frequently portrayed as “emotionally fragile,” “misguided” women who simply made “a mistake” (Frei, 2008, p. 495). The same cannot generally be said of male teachers having sex with young female students. In fact, there is evidence that FSOs are viewed quite differently from MSOs; specifically, previous research has indicated that females accused of an identical sexual crime may be perceived as less abusive, less damaging to their victim, and less responsible for their actions (Becker et al., 2001; Broussard et al., 1991; Bunting, 2007; Fromuth & Holt, 2008). If true, the implications of this differential perception are alarming; failure to recognize the seriousness of female sexual offending ultimately puts children at risk by delegitimizing these crimes for prosecutors, lawmakers, law enforcement, and media personnel (Frei, 2008). In addition, these views may translate into the legal setting, wherein FSOs may be treated differently with regard to sentencing, bail payments, and sex offender registration based on nothing more than their gender. Vandiver and Walker (2002) address the need for research and greater recognition of FSOs, stating,

Law enforcement officials need to be aware of the existence of [female] offenders in order to reduce victimization and increase intervention; attorneys, prosecutors, judges, and legal advisors also need to know how to best handle female sex offenders in order to protect the community. (p. 285)

To these authors’ knowledge, there have been no studies examining the interaction of offender gender and attractiveness specifically in the context of sexual offending. Sexual offending is a unique crime in that it involves a particularly heinous act (sexually violating another person, frequently a child), and it is viewed largely through a gender-biased lens. That is, the commission of a sexual offense by a woman constitutes an especially contemptible violation of social gender norms, in a way that crimes such as burglary or drug use by a female do not (Embry & Lyons, 2012). It is entirely possible that this particular violation would affect sentencing differently than would other, non-sexual crimes committed by a female. Previous research has already indicated that defendant attractiveness influences judgments within a legal context (DeSantis & Kayson, 1997; Mazzella & Feingold, 1994), and media attention surrounding FSOs often focuses on those cases involving an attractive female offender (e.g., Mary Kay Letourneau, Debra LeFave, etc.). A quick Google search reveals numerous Top 10 lists of “hottest” and “sexiest” female sex offenders. Such lists do not appear to exist for males.

The present study set out to measure the effect of gender (of both offender and study participant) and attractiveness (of the offender only) on assignments of punishment for and attitudes about a sexual offending scenario. Specifically, it was hypothesized that (a) participants would rate FSOs more leniently than MSOs, both in punitive measures and in overall attitudes about the sexual relationship; (b) offender attractiveness would affect judgments of FSOs but not MSOs (i.e., participants will view attractive females even more leniently than unattractive females, but attractiveness will not make a difference in ratings of male offenders); and (c) male participants would be more lenient toward FSOs than would female participants.

Participants

Original data included 445 undergraduate participants, of which 13 cases were removed for not completing the online survey (e.g., answering only the first few questions), careless responding (e.g., providing the same response for every question), and/or being highly significant univariate outliers (e.g., writing that the defendant should have to pay US$9 million in bail when the next highest amount was US$500,000). This left an n of 432 for the main analyses, divided approximately equally among the four cells (range = 106-112). Of these participants, 149 were males between the ages of 18 and 42 (M = 19.54, SD = 2.7), and 280 were females between the ages of 18 and 55 (M = 18.97, SD = 2.68; 3 participants chose not to disclose their sex). There were no significant differences in average participant age across the four conditions. Participants were 55% Caucasian (range = 52%-57% across the four cells), 20% Latino (range = 13%-24%), 13% Asian (range = 9%-18%), and 3% African American (range = 2%-4%). The remaining participants were either of other races or did not provide racial information. Participants were recruited primarily from undergraduate psychology courses through a departmental experiments website. In exchange for their participation, they received course credit.

Method

This study was submitted to and approved by a university Institutional Review Board. An important first step involved obtaining objective measures of offender attractiveness for use in the vignettes. In an effort to increase ecological validity, 21 images each of male and female sex offenders and other convicted felons of similar ages were gathered from public Internet sex offender registries and online news media sources. These pictures were presented in random order to 30 undergraduate students, who rated each person’s attractiveness and estimated their age. Although the names and locations of these offenders were not included in study materials, participants were informed in the debriefing process that all pictures were of individuals residing outside of the geographical area from which participants were recruited. Male and female pictures were then matched according to these responses and narrowed down to the final four images of sex offenders (attractive male, unattractive male, attractive female, unattractive female). To avoid any confounding racial bias, all pictures were of Caucasian individuals.

On enrolling in the main study, participants were randomly assigned and emailed a link to a survey with one of four vignettes (each of the four cells had an n between 106 and 112). Each vignette described increasingly illicit sexual interactions between a 35-year-old male or female teacher and 14-year-old female or male student. The relationship begins with kissing, progresses to oral sex, and eventually results in multiple instances of sexual intercourse. The victim’s mother then finds sexual text messages from the teacher on the victim’s cell phone. Both the victim and perpetrator are subsequently interviewed by staff at a child advocacy center (a common procedure for underage victims of sexual perpetration) and the police, respectively (see Appendix A for the complete text of the vignette). While reading the vignette, participants were presented with a picture of an individual, labeled with the perpetrator’s name (Mr. or Ms. Anderson). In the debriefing session, participants were informed that, although the pictures they had viewed were of actual offenders, the story they had read was fictional. With the exception of the gender dyad of teacher–student and the associated picture of the teacher (attractive or unattractive), vignettes were identical. Participants were asked questions about punitive actions, including the amount of bail the offender should have to pay (Bail), the amount of time the offender should be incarcerated (Incarceration), and the length of time the person should be required to register as a sex offender (Registration). In addition, five scales were constructed to measure participant perceptions and attitudes about the relationship. Each scale was composed of five statements for which participants indicated their level of agreement or disagreement using a 7-point Likert-type scale. These 25 total questions were randomly presented. The scales constructed were designed to measure perceptions related to (a) Teacher Responsibility, (b) Student Responsibility, (c) Harm to Teacher, (d) Harm to Student, and (e) General Attitudes about similar statutory sexual relationships.¹ Responses for the five items were summed, with higher scores indicated a belief in greater responsibility, more harm done, and more negative attitudes toward statutory teacher–student relationships.

Analysis

Substantial ranges in participant responses to questions resulted in significant skewness and kurtosis of some variables. This was ameliorated by utilizing square root or log transformations on the three punitive variables (Bail, Incarceration, and Registration). All reported analyses were conducted using these transformations, and resultant tables and graphs for these three variables reflect these transformed values. Analyses consisted of two-way independent MANOVAs with planned contrasts. The independent variables were the version of the vignette the participant read (e.g., attractive male, unattractive male, attractive female, unattractive female) and the sex of the participant. In the first MANOVA, the dependent variables were Bail, Incarceration, and Registration. Levene’s test was non-significant, indicating that assumption of homogeneity of variance was met.

In the second MANOVA, the dependent variables were total scores on each of the five attitudinal scales (Teacher Responsibility, Student Responsibility, Harm to Teacher, Harm to Student, and General Attitudes). Box’s M was non-significant. Levene’s test was significant only for the Teacher Responsibility Scale (p = .011). Because of this, a more conservative alpha criterion of .025 was used for this variable (Tabachnick & Fidell, 2007).

In each MANOVA, planned contrasts were carried out comparing male offenders with female offenders (Contrast 1), attractive female offenders with unattractive female offenders (Contrast 2), and attractive male offenders with unattractive male offenders (Contrast 3).

Results

Punitive Variables

In the first MANOVA, omnibus test results indicated a significant multivariate main effect of Version, F(9, 1263) = 2.99, p < .01 (one-tailed), $η_{p}^{2} = . 021,$ and the interaction effect of Version × Sex, F(9, 1263) = 1.79, p < .05 (one-tailed), $η_{p}^{2} = . 013 .$ There was no significant main effect for Sex.

Follow-up one-tailed univariate tests revealed a significant effect of Version on Bail, F(3, 421) = 4.47, p < .01, $η_{p}^{2} = . 031,$ Incarceration, F(3, 421) = 5.66, p < .001, $η_{p}^{2} = . 039,$ and Registration, F(3, 421) = 4.53, p < .01, $η_{p}^{2} = . 031 .$ The Sex of participant had a significant effect on both Incarceration, F(1, 421) = 2.88, p < .05, $η_{p}^{2} = . 007,$ and Registration, F(1, 421) = 3.91, p < .05, $η_{p}^{2} = . 009,$ but not on Bail. Finally, the interaction of Version × Sex was significant for Bail (Figure 1), F(3, 421) = 3.93, p < .01, $η_{p}^{2} = . 027,$ and Incarceration (Figure 2), F(3, 421) = 2.60, p < .05, $η_{p}^{2} = . 018,$ but not for Registration.

Figure 1.

Bail^a by version and participant gender.

Figure 2.

Incarceration^a by version and participant gender.

The first planned contrast (comparing all male offenders with all female offenders) showed significant effects on all dependent variables (Bail, p < .01; Incarceration, p < .01; Registration, p < .05). The second planned contrast (comparing attractive females with unattractive females) found a significant effect for both Incarceration (p < .05) and Registration (p < .01) but not for Bail. The third planned contrast (comparing attractive males to unattractive males) showed no significant differences. Thus, the hypothesis that male and female offenders would be viewed differently was supported. The hypotheses that participants would view attractive females more leniently and that this would be particularly true for male participants were partially supported.

Scale Variables

The second MANOVA indicated significant multivariate main effects (one-tailed) for Version, F(15, 1257) = 4.88, p < .001, $η_{p}^{2} = . 055,$ and Sex, F(5, 417) = 3.75, p < .01, $η_{p}^{2} = . 043,$ and a marginally significant effect of the interaction Version × Sex, F(15, 1257) = 1.48, p = .053, $η_{p}^{2} = . 017 .$

Follow-up one-tailed univariate tests revealed a significant effect of Version on Teacher Responsibility, F(3, 421) = 9.89, p < .001, $η_{p}^{2} = . 066,$ Harm to Student, F(3, 421) = 22.78, p < .001, $η_{p}^{2} = . 140,$ Harm to Teacher, F(3, 421) = 2.19, p < .05, $η_{p}^{2} = . 150,$ and General Attitudes, F(3, 421) = 4.65, p < .01, $η_{p}^{2} = . 032$ (see Figure 3). There was no effect of Version on Student Responsibility (p > .05). Sex had a significant effect on Teacher Responsibility, F(1, 421) = 11.50, p < .001, $η_{p}^{2} = . 027,$ Harm to Student, F(1, 421) = 10.79, p < .001, $η_{p}^{2} = . 025,$ and General Attitudes, F(1, 421) = 15.92, p < .001, $η_{p}^{2} = . 036,$ but not on Student Responsibility or Harm to Teacher (p > .05).

Figure 3.

Average scales score by version.

Univariate tests of Version × Sex mirrored the effects of Sex, again revealing significant effects on Teacher Responsibility (Figure 4), F(3, 421) = 2.96, p < .025, $η_{p}^{2} = . 021,$ Harm to Student (Figure 5), F(3, 421) = 2.35, p < .05, $η_{p}^{2} = . 016,$ and General Attitudes (Figure 6), F(3, 421) = 2.30, p < .05, $η_{p}^{2} = . 016,$ but not on Student Responsibility or Harm to Teacher (p > .05).

Figure 4.

Teacher responsibility scale by version and participant gender.

Figure 5.

Harm to student scale by version and participant gender.

Figure 6.

General attitudes by version and participant gender.

The first planned contrast (comparing all males with all females) found a significant effect on the Teacher Responsibility (p < .001), Harm to Student (p < .001), and Harm to Teacher (p < .05) scales, but not for the Student Responsibility or General Attitudes scales (see Tables 1 and 2). The second contrast (comparing attractive and unattractive female offenders) found a significant effect for Teacher Responsibility (p < .01), Harm to Student (p < .001), and General Attitudes (p < .01), but not for Student Responsibility or Harm to Teacher. As expected, Contrast 3 showed no significant effect on any of the four variables for attractive versus unattractive male offenders.

Table 1.

Significance Levels and Confidence Intervals for Punitive Variable Planned Contrasts.

Variable	p	95% CI
Variable	p	LL	UL
Contrast 1 (Males vs. Females)
Bail	.001	−146.81	−36.27
Incarceration	.001	−0.74	−0.19
Registration	.022	−0.72	−0.01
Contrast 2 (Attractive Females vs. Unattractive Females)
Bail	.093	−33.44	6.51
Incarceration	.047	−0.18	0.01
Registration	.002	−0.32	−0.06
Contrast 3 (Attractive Males vs. Unattractive Males)
Bail	.273	−24.99	13.21
Incarceration	.062	−0.17	0.02
Registration	.452	0.90	0.13

Note. All p values are one-tailed. CI = confidence interval; LL = lower limit; UL = upper limit.

Table 2.

Significance Levels and Confidence Intervals for Scale Variable Planned Contrasts.

Variable	p	95% CI
Variable	p	LL	UL
Contrast 1 (Males vs. Females)
Teacher responsibility	.000	−11.08	−3.41
Student responsibility	.136	−6.77	1.91
Harm to student	.000	−12.53	−5.19
Harm to teacher	.011	0.64	8.25
General attitudes	.083	−6.60	1.13
Contrast 2 (Attractive Females vs. Unattractive Females)
Teacher responsibility	.001	−3.81	−1.04
Student responsibility	.275	−2.05	1.09
Harm to student	.000	−5.56	−2.90
Harm to teacher	.199	−0.78	1.97
General attitudes	.001	−3.72	−0.93
Contrast 3 (Attractive Males vs. Unattractive Males)
Teacher responsibility	.076	−2.29	0.36
Student responsibility	.155	−2.28	0.72
Harm to student	.095	−2.12	1.10
Harm to teacher	.375	−1.53	1.53
General attitudes	.243	−1.81	0.86

Note. All p values are one-tailed. CI = confidence interval; LL = lower limit; UL = upper limit.

Post Hoc Analyses

Visual inspection of a number of graphs appeared to indicate the possible presence of additional differences in the response patterns of male and female participants: Specifically, many scores for the unattractive female offender appeared to be equal to or, in some cases, greater than those for the male offenders. This appeared particularly true for the ratings of female participants. As such, a number of post hoc analyses were undertaken to further investigate this phenomenon.

A new planned contrast was implemented within the original MANOVA model, this time comparing the unattractive female with the attractive and unattractive males combined (leaving attractive females out of the analyses). This new contrast tested whether unattractive females were viewed or rated significantly differently from males. Surprisingly, for nearly all variables, unattractive females were not viewed significantly differently from the male offenders. There were two exceptions to this trend:

For male participants only, there were significant differences in ratings of unattractive females with regard to Bail and Incarceration (both with p values < .05, two-tailed). This indicates that, particularly when judged by female participants, unattractive FSOs may not be advantaged over MSOs, while attractive FSOs are especially advantaged.

Discussion

The three hypotheses were generally supported by the data. There were clearly significant differences in punitive judgments and attitudes toward males versus females accused of identical sexual crimes. Overall, participants believed in higher bail, lengthier incarceration, and greater time on the sex offender registry for male sex offenders. They viewed male offenders as being more responsible for their actions and more harmful to their victims, while they viewed female offenders as being more “harmed” by an illegal relationship the female herself had initiated. In addition, participants expressed more negative general attitudes about these statutory sexual relationships if they had read vignettes with male offenders. Encouragingly, there were no significant differences in ratings of Student Responsibility, indicating that, regardless of perpetrator gender, the victim in each scenario was not viewed as responsible for what happened.

The hypothesis that attractiveness would make a difference in judgments toward female sex offenders, but not male sex offenders, was partially supported. Whereas attractive female offenders were viewed as less responsible for their actions and less harmful to their victim, the attractiveness of the male offender made no difference. In addition, participants who had read a vignette with an attractive female were more lenient in their responses to the General Attitudes scale questions than were participants who read a vignette with an unattractive female. For participants who read vignettes with male offenders, attractiveness did not affect scores on the General Attitudes scale. Attractive females were rated as deserving lighter prison sentences and less time on the sex offender registry than were unattractive females (although female attractiveness did not affect Incarceration or Bail). Attractiveness of male offenders had no effect on any of the punitive variables.

In addition, the hypothesis that male participants would be more lenient toward female offenders than female participants was partially supported. Male participants who read vignettes about female offenders were more lenient in their responses to questions on the General Attitudes scale than were female participants. In addition, male participants rated female offenders as being less responsible for their actions, less harmful to the victim, deserving of lower bail payments, and deserving of lesser incarceration times.

Finally, post hoc analyses revealed that unattractive females may actually be judged similarly to both attractive and unattractive male offenders, while attractive females are viewed especially leniently. This appeared to be especially true when the offenders were rated by female participants.

Findings from this study may be relevant to professionals in the legal system; a defense attorney with an attractive female client accused of a sex offense may best serve their client by selecting as many male jurors as possible, while a prosecutor may want to put more females on the jury. A defense attorney may best serve a female client by making her as attractive as possible. Mental health professionals treating victims of female sexual abuse should be aware of their own potential biases, and adjust their view accordingly to maximize their effectiveness in helping their clients. If these findings reflect real-world biases in the criminal justice system, the consequences could be severe. For example, teacher background checks are generally done through the Federal Bureau of Investigation (FBI); however, these checks only look for felonies. An attractive female teacher charged with a felony sexual offense may be allowed to plead down to a lower misdemeanor charge in part because of biases like those evidenced in this study; a subsequent background check would then be unlikely to detect this previous sexual offense, thereby allowing the teacher to continue working with children (Knoll, 2010). Subsequently, these teachers may be employed in schools in which no one is the wiser as to their sexually inappropriate past behavior.

When female sex offenders are treated more leniently than male perpetrators, this devalues victims of female-perpetrated sexual abuse. One particularly disturbing finding in this study is that the largest effect size (η_p² = .140) was that of Version on perceptions of student harm. Male victims of FSOs were viewed as being significantly less harmed by the acts perpetrated against them. However, available evidence indicates that sexual perpetration committed by females causes significant harm to male victims, with many reporting long-lasting damage (Saradjian, 2010). One study (Sgroi & Sargent, 1993; cited in Saradjian, 2010) found that victims of sexual abuse committed by females reported being more harmed than victims of a male. It has long been hypothesized that rates of female sexual offending are significantly underestimated and/or underreported (Finkelhor & Russell, 1984); if society sends the message to victims of female perpetration that they are not as “victimized” as victims of male perpetrators, they will likely be reticent to report or seek treatment for abuse. It is essential that society takes note of these differential views and develop methods by which they may be combatted, perhaps by intervening within the legal process itself. Previous research suggests that altering jury instructions may result in reduced racial bias in capital cases (Lynch & Haney, 2000; Shaked-Schroer, Costanzo, & Marcus-Newhall, 2008). Others have suggested training for both prosecutors and defense attorneys to combat common misconceptions concerning the reliability and appropriate use of eyewitness testimony (Wise, Pawlenko, Safer, & Meyer, 2009). The implementation of attorney trainings and jury instructions specific to female sexual offending may be a good starting point in reducing these biases.

Limitations

While the results of this study are intriguing, there are some limitations of note. First, the convenience sample of primarily Caucasian undergraduates, although relatively large, may not generalize to other populations; while many undergraduates are registered voters and therefore subject to jury duty, this population does not reflect the full diversity of a typical jury pool. Furthermore, although there is evidence from the literature that gender and attractiveness affect judges’ sentencing decisions in criminal cases (DeSantis & Kayson, 1997; Rodriguez, Curry, & Lee, 2006; Van Slyke & Bales, 2013), results from this specific study may not broadly apply to judges. Although the most noteworthy aspect of the post hoc analyses conducted was the lack of significant differences between the unattractive female offender and the male offenders, the two significant findings (for Bail and Incarceration) should be interpreted with caution, as they were significant only at p < .05 (as opposed to a lower, corrected alpha level) and the likelihood of a Type I error increases with additional statistical tests. In addition, the vignettes participants read were limited to a single type of sexual offense and may not represent attitudes toward and judgments assigned in other sex offense scenarios (incest, perpetration against young children, etc.). Any sexual offense scenario involves dozens of potentially influencing factors that interact with the numerous components and processes of the criminal justice system to ultimately affect the outcome of a given case. Addressing the individual effects of these influencing factors is beyond the scope of any single research study. Finally, the attitudinal variables were measured using scores on five scales constructed specifically for this project. Although the scales were identical across the four conditions (with the exception of names and pronouns) and are therefore able to be compared directly, they have not been subjected to extensive evaluation of reliability and validity.

Future Directions

There are numerous other offender and participant characteristics that may affect judgments about a sexual offense. Investigating the role racial bias may play in punishing sex offenders could be an important next step. It is also possible that scenarios involving different sexual perpetration scenarios may affect the ways in which offenders are judged, including criminological characteristics such as younger victims, the level of force used by the perpetrator, incestuous abuse, and homosexual dyads. These different scenarios should be investigated.

To increase ecological validity of this particular line of inquiry (i.e., gender and attractiveness in sexual offense cases), future studies could utilize archival information for offenders of whom researchers are able to access a photograph (to assess attractiveness). Alternatively, research could survey those individuals directly involved in the criminal justice process, including law enforcement officers, jury pools, judges, and attorneys.

Finally, further investigation into the role of attractiveness in the judgment of FSOs is warranted, with added emphasis on the gender of the person making these judgments.

Conclusion

The results of this study indicate a bias in punitive judgments and attitudes toward MSOs and FSOs guilty of identical crimes. Females overall were viewed more leniently than males, although post hoc analyses indicate that this difference may have more to do with the influence of the attractive female offender, and that unattractive females are often judged similarly to males. While attractive females were given particularly lenient treatment, attractiveness did not affect judgments toward male sex offenders. In addition, while male and female participants tended to rate male offenders similarly, male participants were even more lenient toward female offenders than were female participants. Female sexual perpetration remains an understudied, under-recognized problem in society. Although recognizing how and under what circumstances biases toward female offenders occur is an important first step, preventing these biases from affecting the management of female sex offenders and from devaluing victims of female sexual abuse will be essential in future research.

Footnotes

Appendix A 2

Michelle Anderson is a 35-year-old junior high school teacher who was recently arrested and charged with Sexual Conduct with a Minor. The minor, Christopher Simmons, is a 14-year-old boy who was a student in her English class this past year.

Christopher’s mother discovered the relationship when she saw a text message on his cell phone from Ms. Anderson saying “I need u inside me now.” His mother proceeded to look through her son’s text history and found several such text messages, including “I love you” and “tonite @ my place.”

Christopher’s mother immediately called the school to report what she had found. The school then called the police. Christopher was picked up from school and taken to the Child Advocacy Center to be interviewed. Although he initially denied having a sexual relationship with Ms. Anderson, when he was confronted with the text messages his mother had found, he told interviewers that he and Ms. Anderson began having a sexual relationship approximately 6 months ago. He said that it began when Ms. Anderson asked him to come to her classroom after school for extra tutoring. She suggested that he come for tutoring after school 3 times a week. Christopher stated that Ms. Anderson sometimes rubbed his shoulders while he was working on his homework. Christopher was uncomfortable with this physical contact, but he told interviewers that he knew it “must be ok” because Ms. Anderson was a teacher. After several weeks of this behavior, Ms. Anderson kissed Christopher one evening in the classroom after school. Again, Christopher states that he was uncomfortable, but he liked the way the kissing made him feel. After kissing for a few weeks, he stated that Ms. Anderson and he began having oral sex (both giving and receiving). He stated that they had oral sex approximately 3 times before having sexual intercourse for the first time. He estimated that they had had sexual intercourse at least a dozen times, usually in her locked classroom after school. Christopher told interviewers that he had initially been uncomfortable with the relationship, but that because Ms. Anderson was a teacher, he did not think anyone would believe him if he were to tell them about the sex. He stated that he did not like having sex with Ms. Anderson at first, but that he now is in love with her. He tells interviewers that he and Ms. Anderson plan to marry when he is old enough.

Christopher has always been an A and B student, but his grades slipped the second half of the school year, and he now has three Cs. He has also started cutting classes and sneaking out of the house at night. Christopher has never had a girlfriend, but he attended two school dances last year with female friends. He says that he loves Ms. Anderson and does not want to get her in trouble.

Christopher’s parents are incredibly angry. They say that Ms. Anderson raped their son “over and over again” and that they want her to “rot in jail” for what she did. They are very worried about their son.

After Christopher’s interview, police brought Ms. Anderson to their precinct for questioning. Ms. Anderson originally denied having a sexual relationship with Christopher, saying that someone else must have stolen her phone and sent the sexual text messages to him. However, after police questioned her further, she began to cry and admitted to “making love” to Christopher “maybe six or seven times,” saying, “I know I shouldn’t have done it, but I really love him. He wants to marry me.” When asked if she thought it was wrong for a 35-year-old teacher to have sex with a 14-year-old boy, she stated, “You have to believe me—I just couldn’t help myself.”

Appendix B 3

1	2	3	4	5	6	7
Strongly disagree	Disagree	Disagree somewhat	Neither agree nor disagree	Agree somewhat	Agree	Strongly agree

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Becker

J. V.

Hall

S. R.

Stinson

J. D.

(2001). Female sexual offenders: Clinical, legal, and policy issues. Journal of Forensic Psychology Practice, 1(3), 31-53.

Beckham

C. M.

Spray

B. J.

Pietz

C. A.

(2007). Jurors locus of control and defendants’ attractiveness in death penalty sentencing. The Journal of Social Psychology, 147, 285-298.

Broussard

Wagner

W. G.

Kazelskis

(1991). Undergraduate students’ perceptions of child sexual abuse: The impact of victim sex, perpetrator sex, respondent sex, and victim response. Journal of Family Violence, 6, 267-278.

Bunting

(2007). Dealing with a problem that doesn’t exist? Professional responses to female perpetrated child sexual abuse. Child Abuse Review, 16, 252-267.

Center for Sex Offender Management. (2008). Female sex offenders. Silver Spring, MD: Author.

Cortoni

F. R.

Hanson

Coache

(2010). The recidivism rates of female sexual offenders are low: A meta-analysis. Sexual Abuse: A Journal of Research and Treatment, 22, 387-401.

Curry

T. R.

Lee

Rodriguez

S. F.

(2004). Does victim gender increase sentence severity? Further explorations of gender dynamics and sentencing outcomes. Crime & Delinquency, 50, 319-343.

Deering

Mellor

(2007). Female-perpetrated child sex abuse: Definitional and categorisational analysis. Psychiatry, Psychology and Law, 14, 218-226.

DeSantis

Kayson

W. A.

(1997). Defendants’ characteristics of attractiveness, race, and sex and sentencing decisions. Psychological Reports, 81, 679-683.

10.

Embry

Lyons

P. M.

Jr. (2012). Sex-based sentencing: Sentencing discrepancies between male and female sex offenders. Feminist Criminology, 7(2), 146-162.

11.

Finkelhor

Russell

(1984). Women as Perpetrators: Review of the Evidence. In David Finkelhor (ed.), Child Sexual Abuse: New Theory and Research. New York, NY: The Free Press.

12.

Frei

(2008). Media consideration of sex offenders: How community response shapes a gendered perspective. International Journal of Offender Therapy and Comparative Criminology, 52, 495-498.

13.

Fromuth

M. E.

Holt

A. R.

(2008). Perception of teacher sexual misconduct by age of student. Journal of Child Sexual Abuse, 17, 163-179.

14.

Gannon

T. A.

Rose

M. R.

(2008). Female child sexual offenders: Toward integrating theory and practice. Aggression and Violent Behavior, 13, 442-461.

15.

Gannon

T. A.

Rose

M. R.

Ward

(2008). A descriptive model of the offence process for female sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 20, 352-374.

16.

Gannon

T. A.

Rose

M. R.

Ward

(2010). Pathways to female sexual offending: Approach or avoidance? Psychology, Crime & Law, 16, 359-380.

17.

Grayston

A. D.

De Luca

R. V.

(1999). Female perpetrators of child sexual abuse: A review of the clinical and empirical literature. Aggression and Violent Behavior, 4, 93-106.

18.

Harris

D. A.

(2010). Theories of female sexual offending. In Gannon

T. A.

Cortoni

(Eds.), Female sexual offenders: Theory, assessment and treatment (pp. 31-51). Chichester, UK: John Wiley.

19.

Johansson-Love

Fremouw

(2006). A critique of the female sexual perpetrator research. Aggression and Violent Behavior, 11, 12-16.

20.

Johansson-Love

Fremouw

(2009). Female sex offenders: A controlled comparison of offender and victim/crime characteristics. Journal of Family Violence, 24, 367-376.

21.

Knoll

(2010). Teacher sexual misconduct: Grooming patterns and female offenders. Journal of Child Sexual Abuse, 19(4), 371-386.

22.

Lynch

Haney

(2000). Discrimination and instructional comprehension: Guided discretion, racial bias, and the death penalty. Law and Human Behavior, 24, 337-358.

23.

Mathews

Matthews

J. A.

Speltz

(1989). Female sexual offenders: An exploratory study. Orwell, VT: Safer Society Press.

24.

Mazzella

Feingold

(1994). The effects of physical attractiveness, race, socioeconomic status, and gender of defendants and victims on judgments of mock jurors: A meta-analysis. Journal of Applied Social Psychology, 24, 1315-1344.

25.

Miller

H. A.

Turner

Henderson

C. E.

(2009). Psychopathology of sex offenders: A comparison of males and females using latent profile analysis. Criminal Justice and Behavior, 36, 778-792.

26.

Nathan

Ward

(2002). Female sex offenders: Clinical and demographic features. The Journal of Sexual Aggression, 8(1), 5-21.

27.

O’Connor

A. A.

(1987). Female sex offenders. British Journal of Psychiatry, 150, 615-620.

28.

Robertiello

Terry

K. J.

(2007). Can we profile sex offenders? A review of sex offender typologies. Aggression and Violent Behavior, 12, 508-518.

29.

Rodriguez

S. F.

Curry

T. R.

Lee

(2006). Gender differences in criminal sentencing: Do effects vary across violent, property, and drug offenses? Social Science Quarterly, 87, 318-339.

30.

Saletan

(2006, January 16). Teachers pets? Are teachers sleeping with boys getting off? Slate News. Retrieved from http://www.slate.com/id/2134158

31.

Saradjian

(2010). Understanding the prevalence of female-perpetrated sexual abuse and the impact of that abuse on victims. In Gannon

T. A.

Cortoni

(Eds.), Female sexual offenders: Theory, assessment and treatment (pp. 31-51). Chichester, UK: John Wiley.

32.

Shaked-Schroer

Costanzo

Marcus-Newhall

(2008). Reducing racial bias in the penalty phase of capital trials. Behavioral Sciences & the Law, 26, 603-617.

33.

Steffensmeier

D. J.

(1977). The effects of judge’s and defendant’s sex on the sentencing of offenders. Psychology: A Journal of Human Behavior, 14(3), 3-9.

34.

Tabachnick

B. G.

Fidell

L. S.

(2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson Education.

35.

Vandiver

D. M.

Kercher

(2004). Offender and victim characteristics of registered female sexual offenders in Texas: A proposed typology of female sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 16, 121-137.

36.

Vandiver

D. M.

Walker

J. T.

(2002). Female sex offenders: An overview and analysis of 40 cases. Criminal Justice Review, 27, 284-300.

37.

Van Slyke

S. R.

Bales

W. D

. (2013). Gender dynamics in the sentencing of white-collar offenders. Criminal Justice Studies, 26, 168-196.

38.

Wijkman

Bijeveld

Hendriks

(2011). Female sex offenders: Specialists, generalists, and once-only offenders. Journal of Sexual Aggression, 17, 34-45.

39.

Wise

R. A.

Pawlenko

N. B.

Safer

M. A.

Meyer

(2009). What US prosecutors and defence attorneys know and believe about eyewitness testimony. Applied Cognitive Psychology, 23, 1266-1281.