Teacher–Student Sexual Relationships: The Role of Age,Gender,and Multiple Victims on Incarceration and Registration Outcomes

Abstract

Minimal research has examined sentencing outcomes in teacher–student sexual abuse cases, while the research that has been conducted has primarily focused on gender and attractiveness. Furthermore, the body of literature examining how gender affects sentencing has contradictory findings. Age of the victim and number of victims may also play a role in sentencing outcomes. The current study aimed to better understand which factors (by examining how gender of the victim and defendant, age of the victim and defendant, and having multiple victims) are associated with sentencing outcomes in cases where teachers engage in inappropriate sexual relationships with students. The sentencing outcomes that were assessed include length of incarceration, likelihood of having to register as a sexual offender, and likelihood of being incarcerated. The current study reviewed 250 cases that were covered in the media between 2008 and 2010 that involved a defendant who was accused of sexual contact with a student whom they had contact with through their role as an educator or school staff member. Results indicate that defendant age, victim age, and whether there was one or multiple victims had significant implications for sentencing outcomes. An increase in victim age was associated with a decreased likelihood of incarceration, shorter incarceration lengths, and a decreased likelihood to be required to register as a sexual offender. An increase in defendant age was also associated with increased likelihood of incarceration. Furthermore, having multiple victims was associated with increased incarceration lengths. This study has implications for school policies and teacher training specific to building appropriate relationships and boundaries with students. Given the rise in reporting of sexual abuse cases between teachers and their students, implementing more comprehensive ethics training is paramount.

Keywords

adolescent victims sexual assault female offenders offenders male victims

In recent years, there has been a notable increase in media coverage surrounding sexual relationships that have taken place between teachers and their students (Knoll, 2010). Conducting a quick search on any major search engine for “teachers arrested” will result in dozens of media reports detailing stories of teachers engaging in inappropriate sexual relationships with their students just within the past year (see Cerullo, 2018; Christenson, 2018; Miller, 2018 to name a few). In addition, high-profile cases such as that of Mary Kay Letourneau who was convicted in 1997 and later married the same student she had a sexual relationship with and that of Debra LaFave who pled guilty to a sexual relationship with a 14-year-old student in 2005 have drawn substantial attention to this issue and have fueled public concern (Bowerman, 2017; Goldenberg, 2006; Helling, 2017; Weaver, 2017).

Despite considerable media attention, there is a lack of empirical research examining teacher–student sexual relationships.¹ An emerging body of literature surrounding this topic has primarily focused on perceptions of teacher–student sexual relationships (Dollar, Perry, Fromuth, & Holt, 2004; Frketic & Easteal, 2010; Fromuth, Holt, & Parker, 2001; Geddes, Tyson, & McGreal, 2013). Most of the literature examining perceptions has focused on gender as a contributing factor influencing perceptions (Dollar et al., 2004; Frketic & Easteal, 2010; Fromuth et al., 2001; Howell, Egan, Giuliano, & Ackley, 2011). For example, research suggests that when sexual abuse occurs in an opposite-sex pairing, a male teacher perpetrating against a female student is typically perceived as being more negative in comparison to a female teacher with a male student (Fromuth et al., 2001; Howell et al., 2011).

Research in this area has been somewhat limited to examining perceptions, with fewer studies focusing on the actual consequences of teacher–student sexual relationships. In general, little research has been conducted to examine predictors of sentencing in child sexual abuse cases (Lewis, Klettke, & Day, 2014). Consequently, there is even less research on sentencing outcomes for sexual abuse cases in which the defendant was a teacher. Of the research that has been conducted, gender, alongside other factors (e.g., attractiveness), has been examined as a predictor of sentencing outcomes in cases of teacher–student sexual relationships (Mackelprang & Becker, 2017). Various studies have found that females are treated more leniently in comparison to males in sexual offense cases (Mackelprang & Becker, 2017; Rodriguez, Curry, & Lee, 2006; Van Slyke & Bales, 2013). Researchers have also found that appearance matters for female offenders (Mackelprang & Becker, 2017). The findings indicate that attractive female offenders receive more lenient judgments, but that attractiveness does not impact judgments of male offenders (Mackelprang & Becker, 2017).

The body of literature surrounding teacher–student sexual relationships suggests that gender is an important factor in understanding sentencing outcomes. Research also indicates that victim age may be relevant, as longer sentences are given to sexual offenders whose victims are younger children (Lewis et al., 2014). What remains unclear is whether victim age is also a predictor of sentencing outcomes in cases of teacher–student sexual relationships. The current study uses data from the Educator Sexual Misconduct Database (Mulligan, 2015) to assess how demographic and victim-associated factors are related to sentencing outcomes for teachers charged with sexual offenses against a student. This study expands upon previous research by examining the potential impact of (a) different opposite-gender teacher–student pairs (i.e., male teacher/female student or female teacher/male student), (b) both teacher and student age, and (c) multiple victims on sentencing outcomes for cases involving teachers charged with sexual offenses against students.

Previous Research

Research suggests that factors such as gender, age, and having multiple victims might be associated with sentencing outcomes (e.g., length of incarceration) in teacher–student sexual relationships. We will review the existing literature on each of these variables in the following.

Gender

Much of the media coverage on teacher–student sexual relationships has focused on cases involving a female teacher (Knoll, 2010). Furthermore, there appears to be a public debate surrounding whether female teachers are likely to receive more lenient sentencing outcomes compared to male teachers who have committed sexual offenses against their students (Cairns, 2012; Saletan, 2006). To date, most of the research conducted on gender has focused on perceptions or attitudes toward the teacher–student sexual relationship based on the gender of teacher (Dollar et al., 2004; Frketic & Easteal, 2010; Fromuth et al., 2001; Geddes et al., 2013; Howell et al., 2011), rather than on how gender impacts sentencing outcomes.

Studies examining perceptions of gender in teacher–student sexual relationship cases have found that respondents tend to rate male teacher/female student sexual relationships more negatively than when there is a female teacher and a male student (Fromuth et al., 2001; Howell et al., 2011). Research suggests that cases in which male teachers have committed sexual offenses against female students are considered to be more serious and are perceived to have a more negative impact on the student compared to female teachers with male students (Frketic & Easteal, 2010).

Furthermore, a number of studies have found that the gender of the survey respondent plays a role in perceptions (Broussard, Wagner, & Kazelskis, 1991; Fromuth et al., 2001; Geddes et al., 2013; Loredo, Reid, & Deaux, 1995) in cases of child sexual abuse. One study examining a sexual interaction between an adult and a 15-year-old student, regardless of the gender combination between the teacher and student, found that female respondents viewed the contact more negatively than did male respondents (Broussard et al., 1991). Consistent with gender-role stereotypes, female respondents perceived sexual relationships between teachers and their students as being more serious and responded to teacher–student sexual relationships with greater anger when compared to male respondents (Geddes et al., 2013). Similarly, Fromuth and colleagues (2001) found that male respondents tended to view a sexual relationship between a 16-year-old student and their teacher more positively than female respondents, regardless of the gender combination between teacher and student.

Together, these findings indicate that perceptions of teacher–student sexual relationships are impacted by gender (Dollar et al., 2004; Frketic & Easteal, 2010; Fromuth et al., 2001; Geddes et al., 2013; Howell et al., 2011). This is important, as gender-stereotypic perceptions may play a role in decision-making at the judicial level. However, far fewer studies have focused on how gender impacts actual sentencing outcomes. In one study that primarily focused on perceptions, the researchers also did a search of the 10 most recent sentences in teacher–student sexual offense cases and found that the average sentence for females was substantially shorter (2.1 years) than for males (10.4 years; Frketic & Easteal, 2010). Although they concluded that this may have been a result of the number of charges, different offense characteristics, or community attitudes, these and similar observations have piqued the interest of researchers.

Generally, researchers have noted that women tend to receive more lenient sentences compared to men across a variety of offenses (Curry, Lee, & Rodriguez, 2004; Rodriguez et al., 2006). Some researchers have posited that the chivalry hypothesis explains these gendered findings (Daly & Bordt, 1995; Embry & Lyons, 2012; Griffin & Wooldredge, 2006; Rodriguez et al., 2006). This hypothesis correlates with traditional gender roles, taking the perspective that women are the weaker gender, being stereotyped as being both childish and unpredictable (Rodriguez et al., 2006). This perspective finds that women are often not held to the same standards and are not generally perceived to be as responsible for their actions in comparison to men (Embry & Lyons, 2012; Rodriguez et al., 2006). As a result, women may be seen as individuals who need to be protected when they face the criminal justice system, which is primarily dominated by men (Embry & Lyons, 2012; Griffin & Wooldredge, 2006). However, it is important to note that the chivalry hypothesis has received scrutiny indicating that this theory may be more selective, only holding true for those who adhere to traditional and stereotypical gender roles (Belknap, 2007). Other researchers are proponents of the evil woman hypothesis, sometimes termed selective chivalry, which suggests that individuals who do not comply with traditional gender roles (e.g., violent crime) are given harsher sanctions because violence does not fall within the traditional female gender role (as discussed in Embry & Lyons, 2012).

Given the finding that suggests that women often receive more lenient treatment by the criminal justice system during sentencing (Curry et al., 2004; Rodriguez et al., 2006), we expect that gender disparities may also exist in sentencing outcomes for teacher–student sexual relationship cases such that female teachers will receive more lenient sentencing outcomes compared to their male counterparts. Recent research provides some support for this hypothesis; in one study, male teachers received harsher penalties than female teachers (Simmon, 2012). This study, however, was limited to only one county in the state of Arizona (Simmon, 2012). Consistent with these findings, researchers in a separate study found that female teachers who had committed sexual offenses against a male student were viewed more leniently and judged less punitively than males (Mackelprang & Becker, 2017). However, this study examined undergraduate student perceptions of how the cases should be judged and thus may not accurately reflect real sentencing outcomes (Mackelprang & Becker, 2017).

Sexual offending is unique in two ways, particularly when women commit sexual offenses. First, a sexual offense in general is often classified as one of the most heinous crimes an individual can commit. Second, when women commit sexual offenses, they are stepping outside of the traditional bounds of gender roles (Embry & Lyons, 2012). At this time, data suggest that less than .001% of reported forcible rapes and nearly 8% of reported sexual offenses other than forcible rape and prostitution are perpetrated by female offenders (U.S. Department of Justice, Federal Bureau of Investigation, 2013); however, researchers posit that sexual offending by women is underreported (Stemple, Flores, & Meyer, 2017). Previous researchers tested a hypothesis that female defendants would receive harsher sentencing outcomes compared to their male counterparts (selective chivalry/evil woman hypothesis; Embry & Lyons, 2012). However, when examining gender differences in sexual offense cases, the results suggested that males received harsher sentencing outcomes compared to their female counterparts, consistent with the chivalry hypothesis (Embry & Lyons, 2012). When offense category (i.e., rape, statutory rape, sexual assault, child sexual assault, and forcible sodomy) was held constant, men received longer sentences compared to women (Embry & Lyons, 2012).

Despite there being minimal recent literature that specifically pertains to sentencing outcomes for teacher–student sexual offense cases, there is reason to believe that gender might play a role in sentencing outcomes. First, it is possible that the individuals involved in decision-making at the judicial (Giles, Blackstone, & Vining, 2008) and prosecutorial (Pritchard, 1986) levels hold similar perceptions to those of the public regarding gender in teacher–student sexual offense cases. If this is true, gender is likely to impact sentencing outcomes. Although judges are the primary decision-makers in sentencing outcomes, prosecutors also have a high level of discretion in charging and sentencing requests. Second, research shows that females are consistently sentenced more leniently than their male counterparts across a variety of offenses (Butcher, Park, & Piehl, 2017; Koons-Witt, Sevigny, Burrow, & Hester, 2014; Rodriguez et al., 2006). Researchers examining sexual offense cases have found similar results, with women typically receiving shorter sentences than men (Deering & Mellor, 2009; Embry & Lyons, 2012). These results indicate that gender differences in sentencing for teacher–student sexual offense cases are likely to emerge. Consequently, this study aims to examine the impact of gender on sentencing outcomes in opposite-sex teacher–student pairs. Due to minimal data on same-sex teacher–student pairs, this study will only examine opposite-sex teacher–student pairs.

Age

The existing body of literature on teacher–student sexual relationships lacks an examination of victim age as a potential variable impacting sentencing outcomes. More general research on sentencing outcomes of convicted child sexual offenders (who were not necessarily teachers who had sexual relationships with their student) suggests that the age of the victim significantly affects sentencing outcomes such that the younger the victim, the more severe the offender’s sentence (Faller, Birdsall, Vandervort, & Henry, 2006; Patrick & Marsh, 2011). In addition, researchers have studied the age of child victims of general sexual abuse and its role on whether or not the victim was perceived as credible (Bottoms, Davis, & Epstein, 2004). These researchers found that children below the age of 13 years were perceived as more credible (more honest/trustworthy) than those above the age of 13 years (Bottoms et al., 2004). However, given that our dataset was restricted to high-school and middle-school teachers who were charged with sexual abuse, our lowest victim age is 13 years. Another study, which examined age, had a similar comparison (children below the age of 12 years vs. children above the age of 12 years) and also found that age significantly contributed to sentencing length with offenders receiving longer sentences when the victim was younger (Lewis et al., 2014). Offender age has also been found to be related to sentencing outcomes. Offender age was positively correlated with actually being sent to prison as well as the length of the prison sentence (Patrick & Marsh, 2011). As such, the impact of victim and offender age on sentencing will be examined in addition to gender to determine how it may influence a variety of sentencing outcomes in teacher–student sexual offense cases.

Multiple Victims

To date, few studies have assessed the influence of multiple victims on sentencing outcomes in sexual abuse cases. One study found that multiple victims increased the likelihood of a defendant being prosecuted (Brewer, Rowe, & Brewer, 1997). In another study, researchers found that the number of offenses predicted sentencing length, while the number of victims did not (Lewis et al., 2014). It is possible that whether the defendant had multiple victims was not a significant predictor in these studies because the cases of each victim may have been processed individually and thus was not taken into consideration in sentencing decisions. The presence of multiple victims has demonstrated contradictory findings in the literature in sentencing outcomes (Brewer et al., 1997; Lewis et al., 2014); therefore, this study will examine whether multiple victims, in addition to gender and age, predict sentencing outcomes in teacher–student sexual offense cases.

Current Study

The current study examined the impact of age, gender, and multiple victims on sentencing outcomes for teachers who engaged in teacher–student sexual relationships. The study expands upon research by Mulligan (2014) by incorporating teacher age into analyses to determine its effect on sentencing outcomes. Sentencing decisions are important as they portray a message to victims, potential offenders, and the public regarding the acceptability of such offenses. Sentencing decisions are particularly important when sexual offenders target victims through their role as a teacher via the imbalance of power between teachers and students, as it can cause long-term trauma to the victims (Knoll, 2010; Sullivan & Beech, 2002). Childhood sexual abuse has shown to be a strong predictor of depression, low self-esteem (Griffing et al., 2006), suicide, increased sexual perpetration, and decreased academic achievement (Paolucci, Genuis, & Violato, 2001). As such, it is important to understand the factors that may predict sentencing outcomes in these cases. The following research question and hypotheses were addressed in this study.

Research Question

How do case factors (defendant age, victim age, multiple victims, and gender of teacher in opposite-sex teacher–student pairings) affect sentencing outcomes in teacher–student sexual offense cases?

Hypothesis 1: Increasing defendant age will be associated with harsher sentencing outcomes.

Hypothesis 2: Increasing victim age will be associated with more lenient sentencing outcomes.

Hypothesis 3: Having multiple victims will be associated with harsher sentencing outcomes.

Hypothesis 4: Females will receive more lenient sentencing outcomes compared to their male counterparts.

Method

The current study analyzed data from the Educator Sexual Misconduct Database (Mulligan, 2015), which was retrieved from the Inter-University Consortium for Political and Social Research (ICPSR). This database is made up of a sample of criminal cases which were identified as involving a defendant who was accused of sexual contact with a student whom they had contact with through their role as an educator or school staff member. These data only include cases in which there was “social consent” rather than physical coercion. Social consent was defined as criminal sexual contact that did not involve physical coercion but was illegal due to the victim being unable to provide legal consent because they were a minor (Mulligan, 2015). It was noted that the victimization was not necessarily without coercion as the perpetrator’s authority and role as a teacher may have exerted social pressure on the victim (Mulligan, 2015). Furthermore, only cases in which victims were 13+ years old and not considered by the media to qualify for special education services were included (Mulligan, 2015). The data were collected between October 2012 and October 2014 (Mulligan, 2015).

The sample is based upon print newspaper articles of the criminal proceedings in each case. Thus, the resulting sample was developed from a content analysis of newspaper archives available through the LexisNexis database (Mulligan, 2014, 2015). Information about each case was gathered from these media reports. Print newspaper articles were utilized to create the sample due to the lack of a national database which identifies cases of teachers accused of sexual misconduct with students (Mulligan, 2014). Furthermore, due to the nature of secondary data, the current researchers were not able to analyze any other sources of media coverage. Cases were selected through searches including the search terms of teacher and student paired with either sexual relationship, sexual contact, rape, or sexual assault. The sample was further constrained by including only cases those that were reported on between 2008 and 2010 with trials having occurred no earlier than 2005 (Mulligan, 2014). Finally, cases which involved same-sex teacher–student pairs (n = 29 out of N = 279) were removed due to minimal data. As a result, the final sample size included 250 total cases. Cases were representative of the United States and its newspaper circulation (Mulligan, 2014). The sample cases were only included if they generated media coverage (i.e., newspaper articles), but it is important to note that the included cases did vary in the amount of coverage they received (Mulligan, 2014). As such, the sample may over-represent cases which were viewed as newsworthy or serious (Mulligan, 2014). A team of undergraduate research assistants coded cases, and inter-coder agreement was deemed acceptable at above 90% (Mulligan, 2014). The dependent, independent, and control variables are more thoroughly described in the subsequent sections. More detailed information on the procedures and methods used in this research are explicated elsewhere (see Mulligan, 2014).

Dependent Variables

This study examined three dependent variables that captured sentencing outcomes. The dependent variables included incarceration, incarceration length, and registration. The incarceration variable captured whether the teacher was sentenced to incarceration with 0 representing was not incarcerated and 1 was incarcerated. Incarceration length is a continuous dependent variable which was measured by the number of days the teacher was incarcerated. Last, the registration variable measured whether the teacher was mandated to publicly register as a sexual offender with 0 representing did not register as a sexual offender and 1 registered as a sexual offender on online sexual offender registration database.

Independent Variables

Four independent variables were used in this study to assess sentencing outcomes. The first independent variable captures whether the defendant/teacher had multiple victims using the multiple victims variable with 0 representing defendant/teacher was not alleged to have had sexual contact with multiple students and 1 defendant/teacher was alleged to have had sexual contact with multiple students. Next, gender of the teacher in opposite-sex teacher–student pairs was captured; defendant/teacher gender with 0 representing male teacher, female student pair and 1 female teacher, male student pair. The last two independent variables were continuous variable measurements of age: victim age and defendant/teacher age, which were both measured in years.

Control Variable

Defendant race is a demographic variable that has been found to predict sentencing outcomes across a variety of offenses (Kutateladze, Andiloro, Johnson, & Spohn, 2014); however, due to the disproportionally low number of non-White participants in the sample, we chose to control for race in our regression analyses. The defendant race variable was dichotomized with 0 representing White and 1 all other non-White. Consistent with statistics of registered sexual offenders in general (Ackerman, Harris, Levenson, & Zgoba, 2011), the majority of sexual offense perpetrators (defendants) in the sample were White. It should be noted that this coding scheme was utilized as the sample contained a disproportionally low number of non-White participants.

Descriptive Variable

Information on the educator’s role or school position (whether they were a teacher or held another school position such as coach/athletic employee, music/band/orchestra/dance/choir instructor, substitute teacher, teaching assistant/aide/student teacher, administrator/counselor, and other education position) was gathered to determine whether this was associated with differences in sentencing outcomes. However, due to the power differential and potential for coercion associated with a “teacher”–student sexual relationship, it was not hypothesized that there would be an association between school position and sentencing outcomes.

Results

Each of the sentencing outcomes (i.e., incarceration, incarceration length, required sexual offender registration) was examined separately. For each outcome, chi-square tests of case factors (e.g., defendant age, victim age, gender of the teacher in opposite-sex teacher–student pairs) are presented. Finally, ordinary least squares (OLS) and logistic regression models predicting sentencing outcomes based upon each case factor was examined. Race was controlled for in all analyses.

Descriptive Statistics

This sample consists of 250 criminal cases involving educator sexual misconduct that occurred between 2008 and 2010 and were covered in news media. In this sample, the defendants were predominantly White (82.4%) and male (60%; thus, the victims in the sample were 60% female due to only examining opposite-sex teacher–student pairs). Given the protection of anonymity that is typically provided to victims in the media, especially when the victim is a minor, race was unable to be captured for the students. The mean age of defendant was 34.63 years and ranged in age from 18 to 61 years. There was one 18- and one 20-year-old defendant (with victims aged 14 and 13 years, respectively), both of which were entrusted with the role of substitute teacher. The mean age of victim was 15.52 years and ranged in age from 13 to 18 years (10 victims were aged 18 years). Although 18-year-olds are legally able to provide consent in sexual encounters, there is an issue of power differentials and the potential for coercion when their sexual relationships are with teachers (Knoll, 2010; Sullivan & Beech, 2002). Furthermore, teachers engaging in sexual relationships with their students is often considered unethical and a potential violation of school policies, regardless of age. The analyses were conducted including and excluding 18-year-olds in the dataset; the results did not change significantly.

Most defendants were teachers (78.1%), with the additional defendants holding other educator positions (e.g., sports coach). Furthermore, 15.6% of the cases included a defendant who was alleged to have had sexual contact with multiple victims. In 86.8% of the cases, the defendant was convicted of a felony and in 81.9%, the defendant plead guilty. In addition, 73.4% of the cases resulted in the defendant being incarcerated and 76.3% of the cases resulted in the defendant being required to register as a sexual offender. See Table 1 for a more comprehensive overview of the sample characteristics of the final sample.

Table 1.

Descriptive Statistics.

	N = 250
	N	%	M	Range
Defendant characteristics
Gender
Male	150	60.0
Female	100	40.0
Race
White	206	82.4
Hispanic	20	8.0
Black	20	8.0
Other	4	1.6
Age			34.63	18-61
Victim characteristics
Gender
Male	100	40.0
Female	150	60.0
Age			15.52	13-18
Case characteristics
Defendant alleged to have multiple student victims	39	15.6
Defendant convicted of a felony	217	86.8
Defendant plead guilty^a	204	81.9
Defendant was incarcerated	182	73.4
Defendant required to register as sexual offender	171	76.3
Educator role
Teacher	193	78.1
Coach or athletic employee	61	24.7
Music, band, orchestra, dance, or choir instructor	15	6.1
Substitute teacher	13	5.3
Teaching assistant, aide, or student teacher	8	3.2
Administrator or counselor	11	4.5
Other educational position	8	3.2
Age difference between defendant and victim (in years)			19.11	4-47

Also includes any cases ending in plea bargains or in deferred adjudication.

Impact of Case Factors on Likelihood of Being Incarcerated

Chi-square tests were conducted to examine associations between male–female and female–male defendant–victim pairings, multiple victims, and school position and the likelihood of being incarcerated. All associations were nonsignificant (see Table 2).

Table 2.

Chi-Squares on Sentencing Outcomes.

	Defendant Gender		Multiple Victims		Race^a		School Position
	χ²	Effect Size	χ²	Effect Size	χ²	Effect Size	χ²	Effect Size
Incarceration length^b	−1.20	−.08	−2.63**	−.18	−0.29	−.02	−0.48	−.03
Register as sexual offender	0.17	.03	2.27	.10	2.45	−.10	0.01	−.01
Incarcerated or not	0.01	−.01	2.99^†	.11	1.53	−.08	0.14	−.02

†

p < .10. ** p < .01.

Race is a controlled variable.

Mann–Whitney U tests were run due to violations of normal distribution (and in some cases, violations to homogeneity of variance).

A binomial logistic regression was conducted to determine the effects of defendant age, victim age, male–female and female–male defendant–victim pairings, and multiple victims on the likelihood of being incarcerated or not. Linearity of the continuous variable with respect to the logit of the dependent variable was assessed via the Box–Tidwell (Box & Tidwell, 1962) procedure. A Bonferroni correction was applied using all six terms in the model resulting in statistical significance being accepted when p < .008333 (Tabachnick & Fidell, 2014). Based on this assessment, all continuous independent variables were linearly related to the logit of the dependent variable. There were 22 cases that were determined to be outliers based having two of the following: standardized residuals greater than ±3 standard deviations (SD), leverage value greater than 0.0403, or Cooks distance greater than 0.0165. The logistic regression model was statistically significant, χ²(5) = 41.35, p < .001. The model explained 24.5% (Nagelkerke R²) of the variance in being incarcerated or not and correctly classified 79.2% of the cases. Sensitivity was 26.8% and specificity was 96.5%. Of the five predictor variables, two were statistically significant at a p = .05 level, victim age, and defendant age (as shown in Table 3). An increase in victim age was associated with a decreased likelihood of being incarcerated. An increase in defendant age was associated with an increased likelihood of being incarcerated. These results suggest that gender, p = .56, did not contribute to the likelihood of being incarcerated in this sample.

Table 3.

Regression Analysis of Impact of Case Factors on Sentencing Outcomes.

	Incarcerated or Not^a				Incarceration Length^b				Register as Sexual Offender^a
	b	SE ^c	B	95% CI	b	SE	EXP (B)	95% CI	b	SE	EXP (B)	95% CI
Defendant age	0.06*	0.02	1.06	[1.01, 1.11]	12.95	11.36	0.07	[–9.44, 35.34]	0.02	0.03	1.02	[0.97, 1.07]
Victim age	−0.81***	0.17	0.45	[0.32, 0.62]	−510.18***	72.35	−0.43	[–652.81, –367.55]	−0.65***	0.16	0.52	[0.38, 0.72]
Defendant gender	0.22	0.37	1.24	[0.60, 2.57]	−176.79	203.56	−0.06	[–578.08, 224.50]	0.06	0.40	1.06	[0.49, 2.31]
Multiple victims	−0.63	0.60	0.53	[0.16, 1.73]	518.88*	262.77	0.12	[0.86, 1,036.89]	−1.27	0.78	0.28	[0.06, 1.30]
Race (White versus other)^d	1.15*	0.54	3.15	[1.10, 9.01]	290.18	272.04	0.07	[–246.12, 826.48]	0.99^†	0.60	2.68	[0.83, 8.64]
Constant	11.31***	2.51	81,491.24		8,520.47***	1,200.31		[6,154.21, 10,886.73]	11.02***	2.65	61,338.83
N	226				215				201
R²/pseudo R²	.25				.22				.18
Effect size					.28
F					11.93***
Wald chi²	41.35***								25.09***

Note. CI = confidence interval; OLS = ordinary least squares.

Logistic regression.

OLS regression.

Robust standard errors in all models.

Race is a controlled variable.

†

p < .10. *p < .05. ***p <.001.

Impact of Case Factors on Incarceration Length

Mann–Whitney U tests were conducted to examine associations between male–female and female–male defendant–victim pairings, multiple victims, and school position and incarceration length. The Mann–Whitney U tests were run due to violations of normal distribution (and in some cases violations to homogeneity of variance); we chose to run nonparametric statistics to examine differences on incarceration length. All associations were nonsignificant except the association between having multiple victims and incarceration length (see Table 2). Individuals who had multiple victims (median days incarcerated = 845, n = 36) received longer incarceration lengths compared with those who only had one victim (median days incarcerated = 365, n = 186), U = 2,424.50, z = −2.63, p = .008, with a small effect size of r = −.18.

A multiple regression was conducted to predict incarceration length from defendant age, victim age, male–female and female–male defendant–victim pairings, and multiple victims. The data demonstrated linearity as assessed by partial regression plots and a plot of standardized residuals against the predicted values. Independence of residuals was confirmed, as assessed by a Durbin–Watson statistic of 2.05. Homoscedasticity was present, as assessed by visual inspection of a plot of standardized residuals versus unstandardized predicted values. No evidence of multicollinearity emerged, which was assessed by tolerance values greater than 0.1. Seven cases were determined to be outliers based having two of the following: standardized residuals greater than ±3 SD, leverage value greater than 0.0450, or Cooks distance greater than 0.0185. The assumption of normality was met, as assessed by a Q–Q plot. The multiple regression model statistically significantly predicted incarceration length, F(5, 209) = 11.93, p < .001, adjusted R² = .22. Victim’s age and having multiple victims added statistically significantly to the prediction, p < .05. Regression coefficients and standard errors can be found in Table 3. A 1-year increase in victim age was associated with a decrease in incarceration length of 510.18 days. In addition, incarceration length for individuals who had more than one victim was 518.88 days greater than for individuals who only had one victim. These results suggest that gender, p = .39, did not contribute to the length of incarceration in this sample.

Impact of Case Factors on Required Registration as a Sexual Offender

Chi-square tests were conducted to examine associations between male–female and female–male defendant–victim pairings, multiple victims, and school position and whether the defendant was required to register as a sexual offender. All associations were nonsignificant (see Table 2).

A binomial logistic regression was conducted to determine the effects of defendant age, victim age, male–female and female–male defendant–victim pairings, and multiple victims on the likelihood of being required to register as a sexual offender. All previously mentioned assumptions were assessed and passed. There were 23 cases that were determined to be outliers based on having two of the following: standardized residuals greater than ±3 SD, leverage value greater than 0.0446, or Cooks distance greater than 0.0183. The logistic regression model was statistically significant, χ²(5) = 25.09, p < .001. The model explained 18.3% (Nagelkerke R²) of the variance in being required to register as a sexual offender and correctly classified 81.6% of cases. Sensitivity was 16.7% and specificity was 98.7%. Of the five predictor variables, one was statistically significant at a 0.05 level, victim age (as shown in Table 3). An increase in victim age was associated with a decreased likelihood of being required to register as a sexual offender. These results suggest that gender, p = .89, did not contribute to being required to register as a sexual offender in this sample.

Discussion

The current study examined the effect of age, gender, and multiple victims on sentencing outcomes in cases of teacher–student sexual relationships. With increased media attention (Knoll, 2010) and public concern around teacher–student sexual relationships, it is important that researchers examine factors that influence sentencing decisions. Understanding these decisions is critical, as these outcomes portray important messages about how these offenders will be punished and treated in society. However, research is still lacking in this area. Consequently, this study aimed to fill in some of the gaps in the current knowledge base by examining the effect of the defendant having multiple victims, defendant and victim age, and gender of the defendant/student in opposite-sex teacher–student pairs on sentencing outcomes in teacher–student sexual relationship cases.

Analyses suggest that victim age is significantly associated with sentencing outcomes, such that when the victim’s age increased, defendants were less likely to be incarcerated. In addition, as the victim’s age increased, there was a decrease in incarceration length by almost 1.4 years, which indicates that teachers who have sexual relationships with younger students receive longer sentences. This is consistent with previous research in which individuals who offend against younger children received harsher penalties (Lewis et al., 2014). In addition, as the victim’s age increases, there was a decreased likelihood of being required to register as a sexual offender. This may be due to state sexual offender registration laws that tend to have automatic registration for offenders with victims under a certain age (Cal. Penal Code § 288). In addition, analyses suggest that defendant age was significantly associated with an increased likelihood of being incarcerated. These results could similarly be due to mandatory sentencing for victims under a certain age (Cal. Penal Code § 288). All states are required to utilize sexual offender registration. However, states vary greatly in their specific legislation (e.g., offense specific, age factors) that leads to registration. Thus, it is possible that the findings of this study could be confounded by the specific mandatory registration requirements that had to be followed in each jurisdiction. Unfortunately, this dataset did not provide jurisdiction information, and as a result, we were unable to control for this factor in the analyses. However, the results suggest that teacher age has no effect on being required to register as a sexual offender or not.

Analyses suggest that teachers who had multiple victims were significantly associated with some incarceration outcomes. Specifically, for teachers with multiple victims, there is an increase in incarceration length by more than 1.4 years, which indicates that teachers who have sexual relationships with multiple students receive longer sentences. These findings were expected, as having multiple victims often results in multiple charges, thus impacting sentence length.

Our results generally support our hypotheses that increasing defendant age would be associated with harsher sentencing outcomes as it pertains to likelihood of being incarcerated; however, it is not associated with incarceration length or being required to register as a sexual offender. In addition, our results support our hypothesis that increasing victim age would be associated with more lenient sentencing outcomes including incarceration length, likelihood of being incarcerated, and being required to register as a sexual offender. Our results also support our hypothesis that having multiple victims would be associated with harsher sentencing outcomes as it pertains to an increase in incarceration length; however, it is not associated with likelihood of being incarcerated or being required to register as a sexual offender.

Our hypotheses about defendant and victim gender pairings were not supported. Despite the most recent literature surrounding sentencing differences in these cases, gender does not appear to be as important in determining sentencing outcomes in our sample. We speculate that there may be several possible explanations for our null findings. We chose to align this hypothesis with the most recent findings in the literature. Taking an optimistic perspective, it is possible that our findings are indicative of more recent sentencing patterns and that judges are making less gender-biased decisions in these cases. However, it is also possible that our findings are specific to this dataset. Furthermore, many of the previous research in this area did not control for the same factors as we did in this study. Based on our findings, it is possible that other case factors such as defendant age, victim age, and multiple victims are more predictive of sentencing outcomes than gender of the defendant in opposite-sex teacher–student pairings.

Recommendations and Future Directions

The current research points us toward a clearer understanding of sentencing outcomes for individuals charged with having sexual contact with a student whom they had contact with through their role as a teacher or school staff member. An important direction for future research is to examine a larger sample to address factors such as race with more confidence. Although the sample in the present study aligned with the overall racial composition of sexual offenders with the majority being White (Ackerman et al., 2011), future research should examine a larger sample to examine racial differences with more power. Future research should also examine a sample that includes both defendant and victim race so that the effect of race pairings on sentencing can be explored. In addition, future analyses will be more powerful if they are able to incorporate and control for jurisdictional differences in mandatory sentencing and registration requirements. This will allow for researchers to parse out whether the findings are being driven by location. Finally, future research should further examine the effects of multiple victims in individual cases (for each victim) and combined cases (for multiple victims) to better understand how number of victims impacts sentencing outcomes.

Research such as ours adds to the emerging body of literature surrounding consequences for teacher–student sexual relationships. However, future research needs to focus more on understanding what leads to the initiation of such relationships. Over the past years, investigations of inappropriate contact between education professionals and students have been on the rise (Chang, 2017). Beyond an understanding of why these offenses occur, researchers also need to focus efforts into interventions, training, and policies that reduce rates of teachers sexually offending against students. In terms of educator preparation, there are very few specific classes on ethics and maintaining boundaries in teaching (Zarra, 2013, 2016). Furthermore, programs and policies for teachers working in the school system are lacking or are unclear. In fact, there is no set federal or state code surrounding teacher professional conduct with students (Zarra, 2016). The lack of clearly defined professional expectations might introduce a blurring of boundaries for some teachers, which may in turn contribute to the formation of inappropriate relationships. For example, the rise in technology and social media use can lead to poor boundaries between teachers and their students (Zarra, 2016).

Given this line of research and previous findings, it is pertinent that universities and colleges incorporate more comprehensive training for teachers specific to building appropriate relationships and boundaries with students. Furthermore, ongoing professional development for teachers working in the school system is absolutely necessary. Professional development can prepare teachers to appropriately deal with situations as they arise and be familiar with the most recent policies. These programs can address informational and strategic ways to combat such issues before they occur (Zarra, 2016). Finally, policies which are clear, consistent, and comprehensive regarding teacher–student sexual relationships must be implemented. Policies should clearly outline all expectations of school personnel and the consequences for any inappropriate behavior. Furthermore, these policies need to be updated over time for the inclusion of new gray areas (e.g., social media) that may lead teachers to cross these boundaries.

Limitations

The current study expanded upon research pertaining to teacher–student sexual relationships; however, it is limited in several ways. The database chosen did not include several potentially important factors which might affect sentencing outcomes (e.g., jurisdiction of sentencing, race of victim). Due to the nature of the dataset, there is a lack of causal explanation; the present study was only able to determine associations between factors and sentencing outcomes. However, little research has been done in this area and the identification of associations is an important starting point. Another limitation is that the number of victims a teacher had was not indicated beyond noting that there was more than one. It is plausible that as the number of victims increase, sentencing outcomes might differ. However, there is limited research that has contradictory findings on the effects of having multiple victims; the present study’s findings add to the limited research. The multiple victims variable presents another limitation in that data were gathered from newspaper articles, and it is possible that the newspaper did not mention or have knowledge of multiple victims.

The findings were further limited by the small size of the data and thus an even smaller sample of minorities which led to race being a control variable. Defendant and victim age also presented limitations within the dataset. The data suggested that age interaction and age difference were both highly correlated with defendant age only and not victim age. As such, only defendant and victim ages were used instead of including age interaction and difference. Conceptually, defendant and victim ages seemed to be more important variables individually than in combination.

Conclusion

In conclusion, the current study demonstrated that age of defendant, age of victim, and whether there was one or multiple victims all have serious implications for sentencing outcomes in cases where teachers are charged with a sexual offense against a student. Specifically, an increase in victim’s age is associated with a decreased likelihood of being incarcerated, a decrease in incarceration length, and a decreased likelihood of being required to register as a sexual offender. The defendant’s age also contributes significantly to an increased likelihood of being incarcerated. In addition, having multiple victims is associated with an increase in incarceration length. Overall, these results indicate that there are a variety of factors that appear to be associated with sentencing outcomes across multiple domains, which suggests that sentencing decisions on teacher–student sexual relationships are multifaceted.

Footnotes

Acknowledgements

The authors would like to thank Shawn Marsh, PhD, for valuable feedback and input that greatly improved the manuscript.

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.

ORCID iDs

Victoria A. Knoche

Kristan N. Russell

Notes

Author Biographies

Victoria A. Knoche, MA, is a doctoral student in the Interdisciplinary Social Psychology PhD program at the University of Nevada–Reno. Her research interests include youth and families within the juvenile justice system. Her research has also focused on examining judicial impact on youth and families within the juvenile justice system.

Kristan N. Russell is a doctoral student in the Interdisciplinary Social Psychology PhD Program at the University of Nevada–Reno. She received her Master of Arts degree in Forensic Psychology from Roger Williams University. Her research has primarily focused on sex offender policies and judicial decision-making as it impacts offenders and their families.

References

Ackerman

A. R.

Harris

A. J.

Levenson

J. S.

Zgoba

(2011). Who are the people in your neighborhood? A descriptive analysis of individuals on public sex offender registries. International Journal of Law and Psychiatry, 34, 149-159. doi:10.1016/j.ijlp.2011.04.001

Belknap

(2007). The invisible woman: Gender, crime, and justice. Belmont, CA: Thomson/Wadsworth.

Bottoms

B. L.

Davis

S. L.

Epstein

M. A.

(2004). Effects of victim and defendant race on jurors’ decisions in child sexual abuse cases. Journal of Applied Social Psychology, 34, 1-33. doi:10.1111/j.1559-1816.2004.tb02535.x

Bowerman

(2017, May 31). A look back at Mary Kay Letourneau and student-turned-husband Vili Fualaau’s relationship. USA Today. Retrieved from https://www.usatoday.com/story/life/nation-now/2017/05/31/look-back-mary-kay-letourneau-and-student-turned-husband-vili-fualaaus-relationship/357605001/

Box

G. E. P.

Tidwell

P. W.

(1962). Transformation of the independent variables. Technometrics, 4, 531-550.

Brewer

K. D.

Rowe

D. M.

Brewer

D. D.

(1997). Factors related to prosecution of child sexual abuse cases. Journal of Child Sexual Abuse, 6, 91-111. doi:10.1300/J070v06n01_07

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. doi:10.1007/BF00980533

Butcher

K. F.

Park

K. H.

Piehl

A. M.

(2017). Comparing apples to oranges: Differences in women’s and men’s incarceration and sentencing outcomes. Journal of Labor Economics, 35(Suppl. 1), S201-S234. doi:10.1086/691276

Cairns

(2012, October 2). Hot for teacher: Gender bias in sentencing of teachers that have sex with their students. CLO: Campbell Law Observer. Retrieved from http://campbelllawobserver.com/hot-for-teacher-gender-bias-in-sentencing-of-teachers-that-have-sex-with-their-students/

10.

Cal. Penal Code § 288.

11.

Cerullo

(2018, April 10). Oklahoma teacher had sex with student before and after statewide school walkout. New York Daily News. Retrieved from http://www.nydailynews.com/news/crime/oklahoma-teacher-arrested-raping-student-day-walkout-article-1.3926705

12.

Chang

(2017, September 15). Statesman exclusive: Improper teacher-student relationship cases soar. Statesman. Retrieved from https://www.statesman.com/news/20170915/statesman-exclusive-improper-teacher-student-relationship-cases-soar

13.

Christenson

(2018, March 14). Marshall teacher arrested, accused of having sex with student. Battle Creek Enquirer. Retrieved from https://www.battlecreekenquirer.com/story/news/2018/03/14/marshall-teacher-arrested-accused-having-sex-student/425403002/

14.

Curry

T. R.

Lee

Rodriguez

S. F.

(2004). Does victim gender increase sentence severity? Further explorations of gender dynamic and sentencing outcomes. Crime & Delinquency, 50, 319-343. doi:10.1177/0011128703256265

15.

Daly

Bordt

R. L.

(1995). Sex effects and sentencing: An analysis of the statistical literature. Justice Quarterly, 12, 141-175. doi:10.1080/07418829500092601

16.

Deering

Mellor

(2009). Sentencing of male and female child sex offenders: Australian study. Psychiatry, Psychology and Law, 16, 394-412. doi:10.1080/13218710902930291

17.

Dollar

K. M.

Perry

A. R.

Fromuth

M. E.

Holt

A. R.

(2004). Influence of gender roles on perceptions of teacher/adolescent student sexual relations. Sex Roles, 50, 91-101. doi:10.1023/B:SERS.0000011075.91908.98

18.

Embry

Lyons

P. M.

Jr. (2012). Sex-based sentencing: Sentencing discrepancies between male and female sex offenders. Feminist Criminology, 7, 146-162. doi:10.1177/1557085111430214

19.

Faller

K. C.

Birdsall

W. C.

Vandervort

Henry

(2006). Can the punishment fit the crime when suspects confess child sexual abuse? Child Abuse & Neglect, 30, 815-827. doi:10.1016/j.chiabu.2005.12.009

20.

Frketic

Easteal

(2010). Public perceptions of teachers’ sexual misconduct with students: Does the sex of the teacher make a difference? Alternative Law Journal, 35, 142-146.

21.

Fromuth

M. E.

Holt

Parker

A. L.

(2001). Factors affecting college students’ perceptions of sexual relationships between high school students and teachers. Journal of Child Sexual Abuse, 10, 59-73. doi:10.1300/J070v10n03_04

22.

Geddes

R. A.

Tyson

G. A.

McGreal

(2013). Gender bias in the education system: Perceptions of teacher-student sexual relationships. Psychiatry, Psychology and Law, 20, 608-618. doi:10.1080/13218719.2012.728428

23.

Giles

M. W.

Blackstone

Vining

R. L.

Jr. (2008). The supreme court in American democracy: Unraveling the linkages between public opinion and judicial decision making. The Journal of Politics, 70, 293-306. doi:10.1017/S0022381608080316

24.

Goldenberg

(2006, March 23). Too pretty for prison. The Guardian. Retrieved from https://www.theguardian.com/world/2006/mar/24/usa.gender

25.

Griffin

Wooldredge

(2006). Sex-based disparities in felony dispositions before versus after sentencing reform in Ohio. Criminology, 44, 893-923. doi:10.1111/j.1745-9125.2006.00067.x

26.

Griffing

Lewis

Chu

Sage

Josprite

DMin

L. M.

Primm

(2006). The process of coping with domestic violence in adult survivors of childhood sexual abuse. Journal of Child Sexual Abuse, 15, 23-41. doi:10.1300/J070v15n02_02

27.

Helling

(2017, October 19). Inside the life of Debra Lafave, the notorious Florida teacher who had sex with a student. People. Retrieved from http://people.com/crime/debra-lafave-florida-teacher-sex-scandal-her-life-now/

28.

Howell

J. L.

Egan

P. M.

Giuliano

T. A.

Ackley

B. D.

(2011). The reverse double standard in perceptions of student-teacher sexual relationships: The role of gender, initiation, and power. The Journal of Social Psychology, 151, 180-200. doi:10.1080/00224540903510837

29.

Knoll

(2010). Teacher sexual misconduct: Grooming patterns and female offenders. Journal of Child Sexual Abuse, 19, 371-386. doi:10.1080/10538712.2010.495047

30.

Koons-Witt

B. A.

Sevigny

E. R.

Burrow

J. D.

Hester

(2014). Gender and sentencing outcomes in South Carolina: Examining the interactions with race, age, and offense type. Criminal Justice Policy Review, 25, 299-324. doi:10.1177/0887403412468884

31.

Kutateladze

B. L.

Andiloro

N. R.

Johnson

B. D.

Spohn

C. C.

(2014). Cumulative disadvantage: Examining racial and ethnic disparity in prosecution and sentencing. Criminology, 52, 514-551. doi:10.1111/1745-9125.12047

32.

Lewis

Klettke

Day

(2014). Sentencing in child sexual assault cases: Factors influencing judicial decision-making. Journal of Sexual Aggression, 20, 281-295. doi:10.1080/13552600.2013.804603

33.

Loredo

Reid

Deaux

(1995). Judgments and definitions of sexual harassment by high school students. Sex Roles, 32, 29-45. doi:10.1007/BF01544756

34.

Mackelprang

Becker

J. V.

(2017). Beauty and the eye of the beholder: Gender and attractiveness affect judgments in teacher sex offense cases. Sexual Abuse: A Journal of Research and Treatment, 29, 375-395. doi:10.1177/1079063215597646

35.

Miller

J. R.

(2018, March 26). Teacher allegedly performed oral sex on 13-year-old in school. New York Post. Retrieved from https://nypost.com/2018/03/26/teacher-allegedly-performed-oral-sex-on-13-year-old-in-class/

36.

Mulligan

(2014). Ideological resilience and disruption: Reproducing and resisting gender and racial inequality in cases of teacher sexual misconduct. Retrieved from https://escholarship.org/content/qt2b2653cv/qt2b2653cv.pdf

37.

Mulligan

(2015). Educator Sexual Misconduct Database 2008–2010 (ICPSR36137-v1) [Data file]. Retrieved from https://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/36137/version/1

38.

Paolucci

E. O.

Genuis

M. L.

Violato

(2001). A meta-analysis of the published research on the effects of child sexual abuse. Journal of Psychology, 135, 17-36. doi:10.1080/00223980109603677

39.

Patrick

Marsh

(2011). Sentencing outcomes of convicted child sex offenders. Journal of Child Sexual Abuse, 20, 94-108. doi:10.1080/10538712.2011.541356

40.

Pritchard

(1986). Homicide and bargained justice: The agenda-setting effect of crime news on prosecutors. The Public Opinion Quarterly, 50, 143-159. doi:10.1086/268971

41.

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. doi:10.1111/j.1540-6237.2006.00383.x

42.

Saletan

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

43.

Simmon

(2012). Sentencing disparities between male and female teacher sexual offenders: Do male offenders receive harsher penalties in Arizona? Retrieved from ProQuest Dissertations and Theses Global. (1268170683)

44.

Stemple

Flores

Meyer

I. H.

(2017). Sexual victimization perpetrated by women: Federal data reveal surprising prevalence. Aggression and Violent Behavior, 34, 302-311. doi:10.1016/j.avb.2016.09.007

45.

Sullivan

Beech

(2002). Professional perpetrators: Sex offenders who use their employment to target and sexually abuse the children with whom they work. Child Abuse Review, 11, 153-167. doi:10.1002/car.737

46.

Tabachnick

B. G.

Fidell

L. S.

(2014). Using multivariate statistics (6th ed.). Essex, UK: Pearson.

47.

U.S. Department of Justice, Federal Bureau of Investigation. (2013, October). Crime in the United States, 2012. Retrieved from https://ucr.fbi.gov/crime-in-the-u.s/2012/crime-in-the-u.s.-2012/resource-pages/about-cius/about-cius

48.

Van Slyke

S. R.

Bales

W. D

. (2013). Gender dynamics in the sentencing of white-collar offenders. Criminal Justice Studies, 26, 168-196. doi:10.1080/1478601X.2012.729707

49.

Weaver

(2017, May 30). Mary Kay Letourneau and former student end their 12-year marriage. Vanity Fair. Retrieved from https://www.vanityfair.com/style/2017/05/mary-kay-letourneau-and-vili-fualaau-separate

50.

Zarra

E. J.

III . (2013). Teacher-student relationships: Crossing into the emotional, physical, and sexual realms. Lanham, MD: Rowman & Littlefield.

51.

Zarra

E. J.

III . (2016). Addressing appropriate and inappropriate teacher-student relationships: A secondary education professional development model. CLEARvoz Journal, 3, 15-29.