Responding to Incidents of Sexual Victimization in Correctional Institutions: Correctional Officer Perspectives

Abstract

This study examined beliefs among correctional officers about responding to inmate-on-inmate sexual assault in jails. It is unique in its focus as no other published study has examined this topic using these variables in this setting. The authors utilize survey methodology to measure correctional officer perceptions about responding to sexual assaults among inmates. Results indicate that support for responding to instances of sexual assault among inmates differs somewhat by type of response and size of jail facility examined. Multivariate results suggest that while perceptions of inmate credibility are usually relevant, demographic factors matter more for those who work in larger jails, while beliefs about rape myths are somewhat more relevant in smaller jails. Implications for correctional policy and training and avenues for continued research are discussed.

Keywords

correctional officers perceptions of victims sexual assault in correctional institutions

The relatively recent political attention to the issue of sexual assault in correctional facilities is overdue, given the long-term existence of the problem. Since its passage in 2003, the Prison Rape Elimination Act (PREA) has allotted for increases in data collection about the incidence and prevalence of sexual assault in correctional facilities (S. Res. 1435, 2003). The law also intends for correctional agencies to implement standards to prevent, reduce, detect, and punish the occurrence of institutional sexual assault. Yet despite its name, correctional administrators have yet to achieve the prevention sought by the federal legislation. This is partially because reduction cannot be measured without proper estimates of its occurrence. Some of the earliest estimates by independent scholars about the prevalence of sexual assault in correctional facilities suggested that up to 30% of inmates were victims or targets of sexual assault (Hensley, Koscheski, & Tewksbury, 2005; Lockwood, 1980; Nacci & Kane, 1983; Struckman-Johnson & Struckman-Johnson, 2000; Struckman-Johnson, Struckman-Johnson, Rucker, Bumby, & Donaldson, 1996; Wooden & Parker, 1982). More recent attempts to determine its exact prevalence in nationwide samples suggests lower estimates, that fewer than 5% of state and federal prisoners and local jail inmates report experiencing at least one incident of sexual victimization within the last year (Beck, Berzofsky, Caspar, & Krebs, 2013). Although only about 2% of state and federal inmates and 1.6% of jail inmates indicated that the perpetrators were fellow inmates, this translates to about 41,000 inmate victims. While these recent estimates suggest that its occurrence may not be as common as previously thought, the sexual victimization of people incarcerated in our correctional facilities is still a problem that merits attention.

The Bureau of Justice Statistics’ third National Inmate Survey, conducted between 2011 and 2012, suggests that some factors make some inmates more susceptible to sexual victimization by other inmates. For both state and federal prisons and local jail inmates, being female, being White, and having a college degree created a greater risk for inmate-on-inmate sexual victimization (Beck et al., 2013). Victimization was also more likely for inmates who reported their sexual orientation as non-heterosexual, or who had experienced sexual victimization prior to their current incarceration. Inmates were also more likely to be sexually victimized by other inmates if they were incarcerated for a violent offense or had a history of mental health problems (Beck et al., 2013). An earlier finding from this survey also suggests that inmates may be particularly vulnerable during specific time periods during an incarceration. A large percentage of inmates in both prisons and jails reported being victimized by other inmates during the first 24 hr of their incarceration, and between the hours of 6:00 p.m. and midnight (Beck, Harrison, Berzofsky, Caspar, & Krebs, 2010).

Another mandate of the PREA legislation is to increase the accountability of correctional officials charged with ensuring the eighth amendment rights of their wards. Standards of accountability suggest that there are expectations of behavior for agency administrators and employees as well as sanctions for behavior outside of those standards. The PREA legislation formalizes the expectations of agencies regarding the issue of sexual victimization of inmates, and in so doing, clarifies standards for the employees of those agencies. Although some correctional employees may resent legislative and judicial meddling in their affairs (Jacobs, 1980), we do not have a lot of data about how correctional staff feel about the issue of sexual assault within correctional facilities. A few studies (discussed below) find that correctional staff are quite progressive in their attitudes about their role in preventing these incidents in correctional facilities. Given PREA legislative mandates and the limited information we have about how correctional staff feel about sexual assault within correctional facilities, it is increasingly vital to examine how those who work within these agencies feel about the issue, as they are best situated to address the problem. The purpose of this research is to examine correctional officer beliefs about responding to acts of sexual assault in correctional environments, specifically jails.

Research About Perspectives

To thoroughly explore this issue, it is imperative to review two distinct fields of research. First, we examine existing literature that helps us understand how sexual assault victims are perceived. Although outlined in more detail below, this review suggests that when considering attitudes about victims of sexual assault and responses to victimization incidents, beliefs about gender, gender roles, and the culture of sexual assault should be examined. Second, an examination of prior research on correctional officer perceptions suggests that there may be several factors relevant to beliefs generally, or relevant to the beliefs about responding to sexual assault victims more specifically. This body of research, reviewed in depth below, indicates that demographic factors, job-related factors, and correctional ideology are all important considerations in correctional officer beliefs, and may also be influential when considering victims of sexual assault within jails.

Perceiving Victims of Sexual Assault, Generally

Demographic factors

Research about perceptions of sexual assault and sexual assault victims has roots in many fields, including criminology and psychology. Scholars in both fields have found that subject demographic variables often influence our perceptions of victims. One of the more established lines of research suggests that females, in general, feel differently about sexual assault and sexual assault victims than their male counterparts. These beliefs are related to both blaming victims as well as perceptions of their credibility; specifically, females are less likely to blame victims and more likely to perceive them as credible (Bottoms & Goodman, 1994; Cook & Lane, 2015; Goodman, Bottoms, Herscovici, & Shaver, 1989; H. Jackson & Nuttall, 1994; Jimenez & Abreu, 2003; Mori, Bernat, Glenn, Selle, & Zarate, 1995; O’Donohue, Elliott, Nickerson, & Valentine, 1992; Xenos & Smith, 2001). Some researchers have also found that while helping behaviors generally cannot be attributed to one gender, females are more likely to intervene in instances of sexual assault within the community (Banyard, 2008; Burn, 2009).

Race and age are other demographic variables that are sometimes relevant in determining attitudes about victims. Research about victims of sexual assault indicates that racial minorities may be more likely to hold negative perceptions about victims of sexual assault (Fischer, 1987; Jimenez & Abreu, 2003; Mori et al., 1995). Another study that examined the impact of age found that older subjects hold less favorable attitudes toward victims of sexual assault than younger subjects do (Nagel, Matsuo, McIntyre, & Morrison, 2005). Higher levels of education have been found to be associated with a lower likelihood of victim blaming and rape myth acceptance, and a greater tendency to perceive victims as credible (M. Burt, 1980; Gray, Palileo, & Johnson, 1993; Xenos & Smith, 2001). Finally, although religion is not definitively related to attitudes toward victims, we know of one study that found more fundamentalist religious attitudes were related to negative attitudes toward rape victims (Sheldon & Parent, 2002).

Attitudes about gender and the culture of sexual assault

In addition to demographic factors, beliefs about sexual assault victims are also impacted by beliefs about gender roles and expectations, as well as the nature of sex and sexual assault. When examining beliefs about victims of and responses to sexual assault, many scholars have attempted to determine the unique impact of attitudes about homosexuality, the credibility of victims, and beliefs about rape myths. Prior research suggests that disapproval of homosexuality may also indicate greater likelihood of victim blaming (Anderson, 2004; D. L. Burt & DeMello, 2003; Wakelin & Long, 2003; White & Robinson Kurpius, 2002).

Ideas about gender roles and myths about the culture of sexual assault are also often related to perceptions about sexual assault victims, as well as the likelihood of adopting helping behaviors in instances of sexual assault (see Banyard, 2011). For example, traditional attitudes about the roles of men and women have been found to be associated with beliefs in the culpability of, and diminished support for, victims (Brownmiller, 1975; M. Burt, 1980; Jenkins & Dambrot, 1987; Kopper, 1996; Pollard, 1992). Research about gender role expectations for males suggests that they are expected to act sexually aggressively and assertively, and these expectations emphasize physical responses to confrontation or unwanted advances (Davies & Rogers, 2006; Davies, Rogers, & Bates, 2008; Howard, 1984a, 1984b; Kassing & Prieto, 2003; Perrott & Webber, 1996). For example, in a case of attempted sexual assault, males are expected to fight off an offender, suggesting to some that they cannot be physically forced to engage in sex (Davies, 2002; Groth & Burgess, 1980). Culpability then increases for males who do not physically resist (Davies & Rogers, 2006; Davies et al., 2008; Howard, 1984a, 1984b; Kassing & Prieto, 2003; Perrott & Webber, 1996), and subjects who exhibit support for rape myths are less likely to label a situation as problematic and less likely to intervene to help victims (Frese, Moya, & Megias, 2004; Frye, 2007; McMahon, 2010). In addition, some literature about helping behaviors suggests that an emotional connection to victims (i.e., knowing a victim) can influence likelihood of intervention (McMahon, 2010). Other research has found that interpreting sexual assault as a problem may also impact the likelihood that someone will intervene in these situations (Banyard, 2008; McMahon, 2010). Although the reviewed research offers a useful backdrop for understanding how sexual assault victims are perceived, it is also imperative to review what is known about how correctional officers perceive the issue of sexual assault.

Correctional Staff Beliefs

There have been some efforts to determine how prison and jail staff perceive the issue of sexual assault in correctional facilities. A series of focus groups conducted in a dozen jails and prisons suggests that staff (officers, supervisors, and administrators) who work in correctional facilities agreed that preventing the occurrence of sexual assault is an essential part of the job (National Institute of Corrections & the Moss Group, Inc., 2006). Some of the early and only existing research about correctional officer perspectives suggests that officers perceive the sexual assault of inmates as a common occurrence; in one study, only 9% stated that it is a rare occurrence (Eigenberg, 1989). Yet more recent studies contradict this finding, suggesting that although the actual prevalence is unknown, correctional staff believe it is an infrequent occurrence and that few inmates are at high risk for victimization (Gonsalves, Walsh, & Scalora, 2012; National Institute of Corrections & the Moss Group, Inc., 2006). Similar studies of wardens in prisons also suggest that they believe it is an infrequent occurrence (Hensley & Tewksbury, 2005a, 2005b; Monster & Jeglic, 2009).

A more consistent finding among studies is that correctional staff, including wardens, recognize it as an underreported event (Eigenberg, 1989; Hensley & Tewksbury, 2005a; National Institute of Corrections & the Moss Group, Inc., 2006). Underreporting in the context of a correctional environment may be functional for the victim, as it might prevent further harassment or victimization (Dumond, 2000; Fishman, 1934; Wooden & Parker, 1982). Yet some researchers have also found that officers who were aware of sexual behavior between inmates may ignore it, unless it causes a disturbance (Eigenberg, 2000b; Wooden & Parker, 1982). Scholars who have studied this issue suggest that officers might also have trouble distinguishing what is “consensual” behavior from that which is forced (Eigenberg, 1989, 2000b; National Institute of Corrections & the Moss Group, Inc., 2006).

Demographic factors and correctional ideology

Other studies of correctional officers in particular have explored various topics other than institutional sexual assault to determine which individual officer characteristics might be correlated with certain attitudes. To understand the unique experience and perspective of correctional officers, scholars have examined several types of demographic factors to determine what impacts their perspectives about their jobs generally or about inmates specifically. Some studies have found that officer gender, race, age, and education are all correlates of officer attitudes about their job or about inmates (Arthur, 1994; Crouch & Alpert, 1982; Cullen, Lutze, Link, & Wolfe, 1989; Farkas, 1999; E. Jackson & Ammen, 1996; Jacobs & Kraft, 1978; Jurik, 1985; Maahs & Pratt, 2001; Poole & Regoli, 1980; Teske & Williamson, 1979; Toch & Klofas, 1982). Specifically, some research suggests that perceptions about particular aspects of correctional work can differ across gender. This research is increasingly relevant given increases in the employment of female officers in correctional institutions (Kifer, Hemmens, & Stohr, 2003). Race and age are other demographic variables that are sometimes relevant in determining attitudes about correctional work. In contrast to research on attitudes toward victims, one study of correctional officers found that racial minorities hold more positive perceptions of inmates than Whites do (Jurik, 1985). Although one study found that education predicted lower levels of support for a custodial orientation (Poole & Regoli, 1980), others suggest that education has no significant impact on attitudes toward inmates or custodial work (Farkas, 1999; Jurik, 1985). In research that is quite relevant to the present study, Eigenberg’s (1994) foundational research on correctional officers incorporated religion as an independent variable and found that officers who are religious are more willing to respond to sexual assault incidents among inmates.

Job-related factors

Job-related factors might also be relevant when considering how officers think about managing inmates and responding to victimization incidents on the job. Seniority, job satisfaction, and job stress have been shown to affect officer attitudes about their profession or managing inmates (Arthur, 1994; Farkas, 1999; Jurik, 1985; Poole & Regoli, 1980; Teske & Williamson, 1979). Although fewer studies have examined the relationship between these variables and beliefs about responding to victimization incidents, one study found that correctional officer stress was correlated with attitudes about correctional work in general. Specifically, this study found that stress was correlated with a custody orientation rather than a rehabilitative orientation (Poole & Regoli, 1980). Although there is not a substantial amount of research about whether attitudes of correctional officers vary by size of jail, there is research that distinguishes the small jail environment from the large jail environment (Applegate & Sitren, 2008; Beck, 2002; Kellar, Jaris, & Manboah-Roxin, 2001; Mays & Thompson, 1988; Ruddell & Mays, 2007). For example, most research that does assess staff perceptions of small versus large jails used administrator samples (sheriffs, wardens, etc.) and found that perceptions of resources, programming, and overcrowding differ by jail size or location (urban or rural; Kellar et al., 2001; Kinkade, Leone, & Semond, 1995).

Attitudes about sexual assault among inmates

The relatively few studies that have considered correctional officer attitudes about sexual assault victims in correctional institutions suggest that officers are, indeed, willing to respond to protect inmates (Eigenberg, 1994, 2000b; Nacci & Kane, 1984). The types of responses they are comfortable adopting vary; some research has found that officers are more comfortable using direct actions as opposed to talking to inmates about prevention practices (Cook & Lane, 2012; Eigenberg, 1994, 2000b). A few studies have found that correctional ideology (e.g., attitudes about rehabilitation, punitiveness, relationships with inmates, etc.) is associated with attitudes about inmate victims of sexual assault among correctional officers. Specifically, being punitive (Eigenberg, 2000a) and preferring social distance from inmates (Cook & Lane, 2015) has been found to be predictive of greater likelihood of blaming inmate sexual assault victims. More specifically related to the current topic, Eigenberg (1994) examined the relationship between officers’ correctional ideology and willingness to respond to instances of sexual assault. Findings from this study suggest that officers who embrace counseling and are comfortable with social proximity to inmates are more willing to respond to instances of sexual assault victimization in correctional settings.

Attitudes about gender and the culture of sexual assault

In addition to demographic correlates, job-related factors, and correctional ideology, ideas about gender and the culture of sexual assault also impact correctional officer perceptions of this issue. Similar to research among more general populations, research among correctional officer samples also suggests that disapproval of homosexuality may indicate greater likelihood of victim blaming (Cook & Lane, 2015; Eigenberg, 2000a). One interesting finding more relevant to the present research is that in predicting willingness to respond to sexual assault in prison, officers who condemn homosexuality are more willing to respond to instances of sexual assault in correctional institutions (Eigenberg, 1994). So while disapproval of homosexuality may impact beliefs about victim culpability, there may also be a tendency to want to respond to and prevent instances of sexual assault victimization among inmates.

It is somewhat uncertain whether belief in the credibility of sexual assault victims affects officer willingness to respond to instances of sexual assault. One recent study suggests that correctional officer attitudes about inmate victim credibility may influence the tendency to blame inmate victims of sexual assault (Cook & Lane, 2015). This research also suggests that perceptions about the culpability of incarcerated male sexual assault victims is related to acceptance of male rape myths (Cook & Lane, 2015). We know of no other research that has examined whether these factors influence correctional officer perceptions about responding to these incidents in correctional facilities. The literature discussed thus far has informed our choice of variables to include in the study. We refrain from making many specific hypotheses, given the inconclusive and limited nature of the research about this specific topic. Therefore, we include many of the previously mentioned variables in this study to examine whether they specifically influence officer perceptions about responding to sexual victimization in jail facilities.

Contributions of the Current Study

Clearly, there is a paucity of research regarding how correctional officers, those most closely situated to those incarcerated, perceive the issue of sexual violence within correctional facilities. There is practically no research about how officers feel about responding to sexual violence within correctional institutions. Research with prison administrators, however, suggests the importance of staff supervision in responding to and preventing sexual assault within correctional facilities (Monster & Jeglic, 2009). Research conducted in correctional facilities has largely been conducted in prison facilities, specifically research about correctional officer beliefs. Jails themselves are not the only understudied phenomena; correctional officers in jails have also been traditionally overlooked. The dynamic nature of jails and their populations, however, necessitates continued research in this context. Jail populations experience more frequent turnover and more admissions than populations in prison facilities (Sabol, 2008a, 2008b). This means that there are more inmates who cycle in and out of jail facilities who are potentially affected by the issue of sexual victimization. This trend underscores the importance of understanding how officers feel about responding to sexual victimization in jails. In addition, administrators and agencies are somewhat disadvantaged in implementing standards to prevent and respond to these incidents if there is a lack of information about the culture of officers who work within jails.

Method

Data Collection and Sample

All of Florida’s 67 jail jurisdictions were contacted to request participation in the study. Of those, 13 jail jurisdictions (approximately 19% of Florida jail agencies) agreed to participate by allowing the researcher to visit the facility, meet with correctional officers during briefings, and distribute surveys to the officers. Survey distribution and data collection were partly driven by the nature of the setting. For example, we chose to allow participants to use return mail to submit their completed surveys rather than conduct the surveys on-site to limit disruption to the agencies, which also increased access to the facilities. Each shift reporting to the jail within one 24-hr data collection period in the summer of 2008 was targeted for the study. During data collection, packets containing a survey, consent form, and a pre-addressed and stamped return envelope were distributed to certified correctional officers of all ranks. The participating facilities varied in geographic setting, as well as facility and staff characteristics. Geographically, facilities were located in all regions of Florida, and varied from large jails in urban areas containing a million or more people to small jails located in rural counties. A total of 1,161 surveys were distributed to officers at the participating jails, which represented almost half (41%, n = 2,822) of officers staffed at these jails during data collection.¹ We received 376 completed surveys; while the individual facility response rate ranged from 11% to 63%, the individual response rate was 32.4%.² Although this response rate is lower than we would prefer, the sampling method was intentionally a purposive non-probability sample as the study was intended to be exploratory. Social exchange theory suggests that this type of survey methodology should net return rates between 30% and 50% (D. Dillman, 2000; F. Dillman, Dolsen, & Machlis, 1995). Although our response rate was on the low end of this continuum, we believe the sample (as discussed below) is still somewhat representative of the facility populations sampled.

Measures

Dependent variables

The survey asked subjects to reflect on their beliefs about expectations of correctional officers in responding to incidents of sexual assault among inmates. The questions were adopted from H. Eigenberg (1994; personal communication, June 7, 2006) and include the following items: “officers should talk to inmates about the risk of sexual assault,” “officers should encourage inmates to report sexual assaults,” “officers should do everything they can to prevent sexual assaults,” “officers should use cell assignments to safeguard inmates from sexual assault,” and “officers should refer inmates to protective custody to safeguard them from sexual assault.”

Although Eigenberg (1994) reports unidimensionality and reliability for these items, these data in this analysis do not support unidimensionality when items were submitted to confirmatory factor analysis using Varimax rotation. They come close, however, the item that deals with “talking to inmates about the risk of sexual assault” loads more strongly on a separate factor (see Table 1 for factor loadings). Thus, we use this variable as our first dependent variable.

Table 1.

Variable and Index Descriptives, Factor Loadings, and Reliability Analyses.

	N = 376	Minimum, maximum, skew	M^a (SD)	Cronbach’s alpha	Factor loading
Dependent variable: talk to inmates
Jail officers should talk to inmates about the risk of sexual assault in jail	368	1, 6, −.78	4.41 (1.42)
Dependent variable: 4-item index		2, 6, -.64		.70
Jail officers should encourage inmates to report sexual assaults that occur in jail	369	2, 6	5.48 (0.71)		.81
Jail officers should do everything they can to prevent sexual assaults in jail	368	2, 6	5.60 (0.62)		.81
Jail officers should use cell assignments to safeguard inmates from sexual assault	365	1, 6	4.92 (1.04)		.71
Jail officers should refer inmates to protective custody to safeguard them from sexual assault	366	1, 6	4.72 (1.20)		.67
Independent variables
Correctional ideology
Counseling roles^a	369	1, 6, .09	3.33 (1.26)	.77
Rehabilitation programs should be left to mental health professionals^b	371	1, 6	3.30 (1.41)		.63
Counseling is a job for counselors, not correctional officers^b	374	1, 6	3.22 (1.55)		.92
If a correctional officer wants to do counseling, he should change jobs^b	375	1, 6	3.47 (1.61)		.91
Punitive orientation^a	361	1.33, 6, .23	3.58 (1.08)	.58
Rehabilitation programs are a waste of time and money	371	1, 6	3.42 (1.44)		.75
Improving jails for inmates makes them worse for officers	368	1, 6	3.15 (1.56)		.71
A military regime is the best way of running a jail	368	1, 6	4.20 (1.41)		.76
Corruption of authority^a	369	2, 6, −1.21	5.13 (0.79)	.74
You cannot ever completely trust an inmate	374	1, 6	5.06 (1.24)		.65
A good principle is not to get “close” to inmates	373	1, 6	5.36 (1.02)		.71
A personal relationship with an inmate invites corruption	372	1, 6	5.45 (0.94)		.75
You must keep conversations with inmates short and businesslike	372	1, 6	4.80 (1.33)		.72
If an officer is lenient with inmates, they will take advantage of him or her	373	1, 6	4.96 (1.11)		.69
Social distance^a	358	1.4, 6, −.06	3.73 (0.88)	.67
A correctional officer should work hard to earn trust from inmates^b	371	1, 6	3.39 (1.54)		.55
It is important for a correctional officer to have compassion^b	371	1, 6	3.03 (1.33)		.68
You get to like the inmates you supervise^b	372	1, 6	4.09 (1.28)		.66
Sometimes a correctional officer should be an advocate for an inmate^b	368	1, 6	3.97 (1.33)		.64
The way to get respect from inmates is to take an interest in them^b	371	1, 6	4.21 (1.24)		.77
Attitudes toward homosexuality (ATH)^a	338	1, 6, .02	3.13 (1.37)	.94
Male homosexual couples should be allowed to adopt children the same as heterosexual couples	368	1, 6	2.75 (1.70)		.82
I would not be too upset if I learned that my son was a homosexual	370	1, 6	2.84 (1.64)		.77
Homosexual behavior between two men is just plain wrong^b	368	1, 6	2.75 (1.70)		.83
Male homosexuality is merely a different kind of lifestyle that should not be condemned	368	1, 6	3.15 (1.66)		.71
Just as in other species, male homosexuality is a natural expression of sexuality in human men	366	1, 6	2.81 (1.55)		.80
I think male homosexuals are disgusting^b	366	1, 6	3.24 (1.72)		.79
Male homosexuals should not be allowed to teach school^b	360	1, 6	3.91 (1.66)		.76
If a man has homosexual feelings, he should do everything he can to overcome them^b	361	1, 6	3.27 (1.62)		.87
Male homosexuality is a perversion^b	359	1, 6	3.36 (1.69)		.88
The idea of male homosexual marriages seems ridiculous to me^b	363	1, 6	2.81 (1.71)		.90
Credibility of victims^c An inmate tells you he was raped. How likely are you to believe him if he:	357	1.44, 5, .46	3.55 (0.71)	.91
Is a muscular inmate	364	1, 5	3.34 (1.05)		.80
Is a homosexual inmate	363	1, 5	3.77 (0.88)		.71
Is a young inmate	364	2, 5	4.08 (0.65)		.69
Is a gang member	364	1, 5	3.34 (1.06)		.82
Is an inmate who owes money	364	1, 5	3.76 (0.84)		.75
Is drunk or high on drugs	358	1, 5	3.29 (0.99)		.83
Delayed reporting the incident	362	1, 5	3.53 (0.90)		.78
Has previously consented to sex with other inmates	361	1, 5	3.30 (1.01)		.81
Has mental health problems	361	1, 5	3.48 (0.89)		.70
Male rape myth^a		1, 6
Most men who are raped by a man are very upset by the incident^b (Myth 1)	372	1, 6, 1.63	1.96 (1.09)
It is impossible for a man to rape a man (Myth 2)	371	1, 6, 2.63	1.65 (1.17)
Most men who are raped by a man do not need counseling after the incident (Myth 3)	372	1, 6, 2.12	1.91 (1.13)
Even a big, strong man can be raped by another man^b (Myth 4)	371	1, 6, 2.03	2.12 (1.35)
Estimate of sexual assault	346	1, 80, 0.69	28.2 (22.2)

Codes: 1 = strongly disagree, 2 = disagree, 3 = not sure, probably disagree, 4 = not sure, probably agree, 5 = agree, 6 = strongly agree.

Reverse-coded item.

Codes: 1 = never, 2 = rarely, 3 = sometimes, 4 = generally, 5 = always.

The remaining index of four items includes items that focus on officer encouragement of reporting and using proactive measures to safeguard inmates from sexual assault. Both the factor loadings (see Table 1) and the reliability analysis suggest that officers view these issues as conceptually similar. The items are summed and averaged to form a four-item index measuring support for encouraging reporting and using proactive measures, and the index is reliable (α = .70). All dependent variable items are coded so that higher scores reflect respondent agreement that officers should respond in these ways to sexual assault. Agreement was measured using a 6-point Likert-type scale, ranging from strongly disagree to strongly agree.

Independent variables

These two dependent variables were examined in relation to several demographic and other independent variables. The inclusion of variables in the analysis is premised on those that other research has suggested may be important in determining attitudes about responding to sexual assault in general or within correctional institutions specifically. Prior research on correctional officer perceptions indicates that several demographic and job-related factors are important to consider when analyzing attitudes about victims of sexual assault, which may also impact perceptions about willingness to respond to these victimization incidents. These factors also extend to attitudinal factors such as beliefs about correctional ideology and the culture of sexual assault.

We examine the following relevant demographic and job-related factors: gender, race, age, education, religiosity, seniority, job satisfaction, job stress, and whether their family or friends had ever been incarcerated. Both gender and race were dummy coded so that females and non-Whites were coded as one (1). Education is coded ordinally so that higher values represent higher levels of completed education (where 1 = high school graduate/general educational development [GED], and 7 = doctorate or professional degree). Age is also coded as an ordinal variable, so that higher responses indicate older age groups (where 1 = 18-25, and 9 = over 60). Subjects were also asked to indicate on a scale of 1 (not at all) to 10 (extremely), how religious they considered themselves. Subjects were asked to indicate their seniority, or job experience, by estimating the number of years they have been working as a correctional officer (to the nearest year). For job satisfaction and job stress, respondents indicated on a scale of 1 (not at all or none) to 10 (extreme) both their level of satisfaction with their job and how much stress they experience on the job, respectively. We also used dummy coding to assess whether their friends or family members had ever been incarcerated, using “no” as the reference group (where 1 = yes).

In addition, we measured the following correctional ideology indices as independent variables: beliefs about counseling roles, punitive orientation, corruption of authority, and social distance. Questions on the survey asked subjects to respond with their level of agreement from response options ranging from strongly disagree to strongly agree on a 6-point Likert-type scale (some items were reverse coded as indicated in Table 1). All four indices were created by including questions as indicated below, and summing and averaging subject responses to these questions. The indices were identified by Whitehead, Linquist, and Klofas (1987) and originally adopted from Klofas and Toch (1982) and Toch and Klofas (1982). Confirmatory factor analysis suggests unidimensionality of each index and reliability analysis suggests moderate to strong internal consistency (see Table 1).

To gauge officers’ perceptions about whether they should adopt counseling roles, they were asked to respond to questions such as “counseling is a job for counselors, not correctional officers” (see Table 1 for a list of all items). This index is coded so higher responses reflect support for officers adopting counseling roles. Punitive orientation was measured by inclusion of questions such as “improving jails for inmates makes them worse for officers” (see Table 1 for a list of all items). These items are coded so higher scores reflect more punitiveness. The third correctional ideology topic measured, corruption of authority, showed the officer’s concern about the ability of an inmate to corrupt an officer’s authority. This five-item index (see Table 1 for a list of all items) included, for example, the statement: “a personal relationship with an inmate invites corruption.” The index is coded so that higher responses suggest more concern about the ability of inmates to corrupt their authority. The final correctional ideology topic measured is perceptions about maintaining social distance from inmates, and includes the following as an example of the five questions: “it’s important for a correctional officer to have compassion” (see Table 1 for a list of all items). These items are coded so that higher values reflect more preference for maintaining social distance in managing inmates (or less comfort with social proximity to inmates).

Finally, we measured the following beliefs about gender and the culture of sexual assault that prior literature has found relate to either perceptions about victims or responding to victimization instances: beliefs about homosexuality, perceptions of victim credibility, and beliefs about male rape myths. We measured attitudes toward homosexuality using Herek’s (1988) 10-item published index of attitudes toward gay men. Subjects were asked to indicate their level of agreement on a 6-point Likert-type scale where higher responses indicate more acceptance (or approval) of homosexuality, and scores were summed and averaged to create an internally consistent index. This index includes statements such as: “male homosexuality is merely a different kind of lifestyle that should not be condemned,” and “I would not be too upset if I learned that my son was a homosexual.”

We also included perceptions of victim credibility in this study to examine whether, as discussed in the literature review, they also influence perceptions about responding to victimization incidents. The survey included a nine-item index measuring perceptions of victim credibility, adopted from Eigenberg (1989) and Jordan (2004). Respondents were asked whether certain characteristics of inmates or sexual assault incidents would affect the credibility of an inmate victim. Specifically, officers were asked if delays in reporting an assault, victim drug use, previous consent, victim mental health problems, victim debt, or various attributes (e.g., youth, muscularity, homosexuality, or gang affiliation) would affect the credibility of the inmate victim. Response options ranged from 1 = never believe to 5 = always believe, and are coded such that higher responses indicate belief in victim credibility. A summed and averaged index of these items indicates that they were unidimensional and highly reliable.

We also measured beliefs about male rape myths, which included four separate items adopted from Struckman-Johnson and Struckman-Johnson (1992). These myths were measured using the following questions: “most men who are raped by a man are very upset by the incident,” “it is impossible for a man to rape a man,” “most men who are raped by a man do not need counseling after the incident,” and “even a big, strong man can be raped by another man.” Factor and reliability analyses indicate that the items are neither unidimensional nor internally consistent, therefore, all four items were included separately in the analysis. Respondents were asked to respond using a 6-point Likert-type scale ranging from strongly disagree to strongly agree, and items are coded (and reverse coded where indicated in Table 1) so that higher responses suggest greater adherence to male rape myths and lower responses suggest rejection of rape myths about men. The final independent variable measured was used to determine whether officers perceive sexual assault as problematic within jails and was assessed by asking them to estimate the percentage of inmates that have been forced to have sexual contact against their will by other inmates.

Analysis and Results

We first present information about the study sample, including descriptives of the independent variables. We then conduct both bivariate and multivariate analyses by splitting the sample into small jails (fewer than 1,500 inmates) and large jails (1,500 or more inmates) according to the official count. The decision to assess the impact of the independent variables on willingness to respond to sexual assault in this way was based on the recognition that small and large jails are unique in both their environment, and may be unique in how staff perceives issues within them.³ Bivariate correlations are then examined to determine which demographic factors and independent variables are significantly correlated with each dependent variable. Those variables that are significantly associated with the dependent variables are subsequently entered into one of two separate, theoretically driven, nested ordinary least squares (OLS) regression models.

Sample

As Table 2 indicates, the majority of officers who responded were male (n = 260, 69%) and White (n = 293, 81%). The proportions of male and female respondents matched those employed by these facilities at the time of data collection. However, the proportion of White respondents in our sample is larger than the proportion of officers employed at these facilities (65%), and the proportion of Black respondents in our sample (16%) is smaller than the proportion employed at these facilities (29%).

Table 2.

Sample Characteristics.

	Codes	Sample (N = 376)	%	M	SD	Target population (N = 2,822)	%
Female	1	115	30.6			869	30.8
Non-White	1	69	19.1			983
Age				4.38	1.94
18-25	1	26	7.0
26-30	2	40	10.7
31-35	3	58	15.5
36-40	4	79	21.2
41-45	5	71	19.0
46-50	6	44	11.8
51-55	7	29	7.8
56-60	8	18	4.8
Over 60	9	8	2.1
Missing		3	0.8
Education^a				3.23	1.31
HS or GED	1	38	10.1			2,822
Some college <1 year	2	79	21.0
>1 year college	3	102	27.1
Associates	4	78	20.7			304	10.7
Bachelor’s	5	71	18.9			160	5.7
Master’s	6	6	1.6			7	0.2
Professional/doctorate	7	1	0.3			0
Missing		1	0.3
Religiosity	1 (not at all)-10			5.76	2.43
Seniority^b (mode = 2)				10.25	6.99
Four years or less		95	26.0
5-9 years		94	26.0
10-14 years		74	20.2
15-19 years		58	15.8
20-24 years		34	9.3
25 years or more		11	3.0
Missing		10	2.7
Satisfaction (job)	1 (not at all)-10			6.92	2.09
Stress (job)	1 (none)-10			5.96	2.17
Know a prisoner	0 (no)-1	253	67.5	0.67	0.47

Note. HS = high school; GED = general educational development.

Minimum education achieved.

Nearest year.

At the time of the survey, the majority of respondents were between the ages of 31 and 45 (n = 208, 56%). Most of the respondents reported education levels beyond high school at the time of the survey: about half (n = 181, 48%) completed some college credit, but did not yet graduate with a degree. Although not all facilities kept track of the number of officers with educational levels beyond the minimum requirements of the job (which were typically a high school diploma or GED), seven of the 13 facilities did record this information. For these seven facilities, 27% of officers employed in these facilities had an associate’s degree, while 14% had a bachelor’s degree. Among the sample, only 21% had an associate’s degree, while 19% had a bachelor’s degree. Regarding education level, then, it appears as if the sample may be similar to the target population, although we cannot be certain because of missing data. Given that the score of 10 represented “extremely” religious, the mean (5.76) suggests that overall, the sample is moderately religious. A larger percentage (n = 253, 68%) of respondents had a friend or family who had been incarcerated. In terms of job-related variables, the sample was somewhat inexperienced, satisfied with their job, and moderately stressed. The average response for seniority was 10 years, while the modal response was 2 years. On a scale of 1 to 10, the average response about job satisfaction was 6.92. The mean level of stress (5.96) suggests that while officers experienced stress on the job, they only report moderate levels.

Bivariate Correlations

Bivariate correlations indicate that several demographic and other independent variables are significantly associated with at least one dependent variable. In the full sample (see Table 3), the following variables are significantly and positively correlated with support for talking to inmates (the first dependent variable): age, seniority, counseling roles, and credibility. Specifically, older respondents, those with more experience, those who embrace counseling roles, and those who perceive inmates as credible are more willing to support talking to inmates about the risks of sexual assault. In addition, two variables are significantly negatively associated with the first dependent variable: corruption of authority and social distance. If officers express concern about the ability of inmates to corrupt their authority or prefer to have social distance from inmates, they are less supportive of talking to inmates about the risk of sexual assault. The following variables are significantly and positively associated with the four-item index for encouraging inmate reporting and proactive responses (the second dependent variable): age, job satisfaction, and credibility. Older respondents, those with more job satisfaction, and those who perceive inmates as credible are more willing to respond in this way. Two other variables are significantly negatively associated with this dependent variable. Officers who prefer social distance from inmates and those who endorse the myth that men cannot be raped are less willing to support responding in this manner. Although separate correlations are not presented by facility size in the table, they indicate that in small jails (fewer than 1,500 inmates), social distance and credibility of inmates were significantly associated with the “talk to inmates” dependent variable. Specifically, less preference for social distance and belief in the credibility of inmates were both correlated with willingness to talk to inmates about the risks of sexual assault. Similarly, when examining correlations for the four-item index within small jails, respondents who believed in the credibility of inmates were also more likely to be willing to respond to these items that focus on encouraging reporting and using proactive measures. Yet this second dependent variable was more distinct when analyzed among smaller jail respondents, as gender, age, perceptions about corruption of authority, and rape myths were also significantly correlated to this four-item dependent variable index. Females, older respondents, those who are concerned about corruption of authority, those who reject the myth that males cannot be raped, and those who reject the myth that victims do not need counseling following a rape incident are more likely to endorse inmate reporting and using proactive measures in response to sexual assault within small jails.

Table 3.

Willingness to Respond Zero-Order Correlations (Full Sample).

	Gender	Race	Age	Education	Religious	Seniority	Satisfaction	Stress	Know prisoner	Counsel roles	Punitive orient	Corrupt authority	Social distance
Gender
Race	.11*
Age	−.10	−.04
Education	−.01	.06	.16**
Religious	.06	.25**	.18**	−.03
Senior	−.11*	.06	.52**	.16**	.15**
Satisfaction	−.01	−.03	.10	−.04	.19**	.00
Stress	.07	.00	.13*	.02	.08	.15**	−.15**
Know prisoner	.07	.22**	−.16**	−.11*	.07	.02	−.04	.04
Counsel roles	−.03	.17**	.04	−.06	.16**	.03	.11*	−.07	.15**
Punitive orient	−.06	−.24**	−.05	−.01	−.15**	−.05	−.11*	.07	−.06	−.39**
Corrupt authority	.04	−.19**	−.01	.03	−.08	.06	−.02	.13*	−.09	−.36**	.38**
Social distance	.02	−.01	−.10	−.03	−.11*	.01	−.04	.05	−.04	−.29**	.27**	.28**
Talk to inmates (dv1)	.02	.07	.14**	.05	.07	.18**	.06	.05	.06	.15**	−.01	−.12*	−.20**
4-item index (dv2)	.05	−.05	.19**	.05	.03	.08	.11*	.06	−.02	.05	.02	.10	−.21**
	Attitude toward homosexuality^b	Credibility	Raped men are upset^c	Men cannot be raped^c	Raped men need no counseling^c	Strong men can be raped^c		Estimate		Talk to inmates (dv1)		4-item index (dv2)
Attitude toward homosexuality^b
Credibility	.18**
Raped men are upset	.01	.16**
Men cannot be raped	−.10	−.16**	.10
Raped men need no counseling	−.16**	−.12*	−.26**	.26**
Strong men can be raped	−.12*	.02	.01	−.01	.06
Estimate	.07	.06	.12*	.12	−.08	−.02
Talk to inmates (dv1) ^a	.04	.16**	.04	.02	−.03	.05		.09
4-item index (dv2) ^a	.03	.24**	−.08	−.23**	−.20**	.05		−.08		.32**

Willing to talk to inmates (dv1), willing to encourage reporting and use proactive measures (dv2).

Attitudes toward homosexuality.

Most raped men are very upset by the incident (Myth 1); it is impossible for a man to rape a man (Myth 2); most raped men do not need counseling after the incident (Myth 3); even a big, strong man can be raped by another man (Myth 4).

p < .05. **p < .01.

In contrast to small jails, respondents from larger facilities (more than 1,500 inmates) exhibit distinct attitudes when assessing willingness to talk to inmates about the risk of sexual assault. The following variables are significantly correlated with the “talk to inmates” dependent variable in large jails: age, seniority, job satisfaction, counseling roles, and social distance. Older respondents, those with more experience, those who are more satisfied with their jobs, those who embrace counseling roles, and those who prefer less social distance from inmates are more likely to endorse talking to inmates about the risk of sexual assault.

Similarly, in large jails, age and social distance are also significantly associated with the second dependent variable: the perception that officers should encourage inmate reporting and using proactive measures to protect inmates from sexual assault. Again, older officers and those who prefer less social distance from inmates are more supportive of these responses. Similar to respondents from small jails, credibility and rape myths are also significant. Specifically, belief in inmate credibility, rejection of the myths that males cannot be raped, and that they do not need counseling following a rape incident are all significantly correlated with support for measures encouraging inmate reporting and officers using proactive measures.

Regression Analysis

To further examine the impact that demographic and other relevant independent variables have on the two dependent variables, we estimated two separate theoretically built, nested OLS regression models.⁴ This model estimation allowed us to determine how the impact of each independent variable changed once we added subsequent independent variables to the model. We examine this change by comparing standardized coefficients and R² change statistics across models. In Tables 4 and 5, we feature nested models predicting each dependent variable in each regression analysis. The first model of each regression (Model 1) includes demographic variables (e.g., age) and job-related factors (e.g., seniority) that were significantly bivariately correlated with the specific dependent variable. The second model (Model 2) introduces relevant correctional ideology variables (e.g., social distance) that were significantly bivariately associated with the specific dependent variable. Finally, the last model (Model 3) incorporates variables relating to beliefs about gender and the culture of sexual assault (e.g., credibility or rape myths) that were significantly bivariately correlated with the specific dependent variable.⁵

Table 4.

Regression Predicting Willingness to Respond by “Talking to Inmates.”

	Model 1		Model 2
	b (SE)	β	b (SE)	β
Small jails: <1,500 inmates
(n = 124^a)
Social distance	−.47 (.14)	−.29**	−.40 (.14)	−.25**
Credibility			.38 (.17)	.19*
Constant	6.29 (.55)		4.67 (.90)
Model R²	.08		.12
Model R² adjusted	.08		.10
R² change	.08		.04
Model F	8.22**		5.70***
Large jails: 1,500+ inmates
(n = 215^b)
Age	.06 (.06)	.08	.04 (.06)	.05
Seniority	.03 (.02)	.15	.03 (.02)	.17*
Satisfaction	.09 (.05)	.13	.08 (.05)	.12
Counseling roles			.13 (.08)	.12
Social distance			−.25 (.11)	−.15*
Constant	3.19 (.39)		3.77 (.67)
Model R²	.06		.10
Model R² adjusted	.04		.08
R² change	.06		.05
Model F	4.27**		4.77***

Note. Standard errors in parentheses.

Incomplete cases were excluded through listwise deletion (n = 139).

Incomplete cases were excluded through listwise deletion (n = 237).

p < .05. **p < .01. ***p < .001.

Table 5.

Regression Predicting Willingness to Respond Using Four-Item Index “Encouraging Reporting and Using Proactive Measures.”

	Model 1		Model 2		Model 3
	b (SE)	β	b (SE)	β	b (SE)	β
Small jails: <1,500 inmates
(n = 120^a)
Gender	.34 (.15)	.20*	.31 (.15)	.19*	.19 (.15)	.11
Age	.06 (.03)	.17	.06 (.03)	.16	.06 (.03)	.15
Corruption of authority			.14 (.09)	.14	.08 (.09)	.08
Credibility					.19 (.09)	.18*
Men cannot be raped^b					−.14 (.05)	−.23*
Raped men need no counseling^c					−.08 (.05)	−.14
Constant	4.81 (.16)		4.08 (.48)		4.16 (.61)
Model R²	.06		.08		.20
Model R² adjusted	.04		.06		.16
R² change	.06		.02		.12
Model F	3.77*		3.39*		4.83***
Large jails: 1,500+ inmates
(n = 215^d)
Age	.06 (.02)	.18**	.05 (.02)	.16*	.05 (.02)	.15*
Social distance			−.17 (.05)	−.23***	−.14 (.05)	−.19**
Credibility					.16 (.06)	.18**
Men cannot be raped^b					−.05 (.04)	−.09
Raped men need no counseling^c					−.07 (.04)	−.11
Constant	4.93 (.11)		5.57 (.21)		5.12 (.30)
Model R²	.03		.09		.16
Model R² adjusted	.03		.08		.14
R² change	.03		.05		.07
Model F	7.15**		10.03***		7.82***

Note. Standard errors in parentheses.

Incomplete cases were excluded through listwise deletion (n = 139).

It is impossible for a man to rape a man (Myth 2).

Most raped men do not need counseling after the incident (Myth 3).

Incomplete cases were excluded through listwise deletion (n = 237).

p < .05. **p < .01. ***p < .001.

Talking to inmates

We first present the regression analysis for predicting whether officers support talking to inmates about the risks of sexual assault. In small jails, both factors that were bivariately correlated with this dependent variable were significant in the final multivariate model as well. Even when controlling for the other variable, both beliefs about social distance from inmates and perceptions of inmate credibility mattered: Officers who were comfortable with social proximity to inmates and those who believed in inmate credibility were more likely to support talking to inmates about the risk of sexual assault.

The final model of the analysis (Model 2) explains about 12% of the total variance in support for talking to inmates about the risks of sexual assault in small jails (adjusted R² = .10). Comparing R² change scores across models for this dependent variable highlights some important considerations. Beliefs about social distance between officers and inmates added twice the amount of explained variance (R² change is .08) compared with perceptions of credibility (R² change is .04). Comparing standardized coefficients also supports the conclusion that this variable is the most important factor in predicting this willingness to talk to inmates (β = −.25). Overall, officers in small jails are more supportive of talking to inmates about the risks of sexual assault if they are more comfortable with social proximity in their relationships with inmates and if they perceive inmate victims as credible.

In large jails, the predictive variables for officer support for talking to inmates are different (see Table 4). Of the demographic (age) and job-related (job satisfaction) factors that were bivariately correlated with this dependent variable in large jails, only seniority remains significant in the multivariate model, and it retains significance across all models. In large jails, respondents with more experience are more supportive of talking to inmates about the risks of sexual assault. One of the correctional ideology variables, social distance from inmates, is also relevant. In large jails, as in the model for small jails, officers who prefer more social distance from inmates are less likely to support talking to them about the risks of sexual assault. The total explained variance for the final model, which also incorporates support for counseling roles (which is not significant), is .10 (adjusted R² of .08). Analysis of R² change scores and standardized coefficients (β = .17) suggests that seniority is the most important predictor of support for talking to inmates about the risk of sexual assault among officers in larger jails.

Overall, when predicting support for talking to inmates about the risk of sexual assault, this analysis suggests that our models do a better job estimating probabilities among smaller jails. Within smaller jails, support for talking to inmates is strongly related to how officers feel about social distance from inmates and inmate credibility. Among officers in larger jails, support for talking to inmates about the risks of sexual assault is more likely among officers who have more experience and are comfortable with social proximity to inmates.

Encouraging reporting and using proactive measures

The second regression analysis examines the four-item indexed dependent variable that measures support for encouraging inmate reporting and the use of proactive measures (cell assignments and protective custody) to protect inmates from sexual assault. Within small jails, gender is the only demographic or job-related variable that is significantly predictive of support for proactive measures in the first two models. Gender remains significant after the inclusion of measures of correctional ideology (Model 2), yet loses significance when variables that measure beliefs about gender and sexual assault are added (Model 3). In Model 3, officers who perceive inmate victims as credible are also more willing to respond in this way. The third model also suggests that both male rape myths retain significance in the multivariate analysis. Officers who reject the myths that men cannot be raped and that men do not need counseling after rape are more likely to support encouraging inmate reporting and using proactive measures to protect inmates from sexual assault.

For small jails, about 20% of the total variance in this dependent variable is explained by the included independent variables (adjusted R² of .16). A comparison of explained variance across models suggests that independent variables added in Model 3 are more important predictors of whether officers are supportive of responding by encouraging reporting and using proactive measures. Specifically, while demographics explain about 6% of the variance in this dependent variable, the inclusion of beliefs about gender and sexual assault variables in the third model adds even more explained variance (R² change of .12). It seems, then, that the inclusion of beliefs about gender and sexual assault variables (credibility and male rape myths, specifically) add the most explained variance in this analysis. A comparison of standardized coefficients also illustrates important points about the relative importance of variables in explaining support for being proactive in protecting inmates. This analysis suggests that the most important predictors of this dependent variable are perceptions of inmate credibility (β = .18) and rejection of male rape myths (β = −.23 for the myth that men cannot be raped, and β = −.14 for the myth that they do not need counseling, respectively) when considering the relative importance of standardized coefficients. Again, officers in small jails are more supportive of encouraging inmate reporting and using proactive measures to protect inmates from sexual assault if they perceive inmate victims as credible and reject the myths that men cannot be raped or that they do not need counseling after rape.

In large jails, three of the five variables that were bivariately associated with support for encouraging inmate reporting and using proactive measures retain significance in the multivariate model. Age, beliefs about social distance from inmates, and perceptions about inmate credibility are all significant predictors of whether officers are supportive of using these measures in response to sexual assault. Specifically, older officers, those comfortable with less social distance from inmates, and those who believe in the credibility of inmates are all more likely to agree with using these measures in response to sexual assault.

The total explained variance for the final model is .16 (adjusted R² of .14). Analysis of R² change scores indicates that variables added later in the analysis (models 2 and 3) are more important predictors of officer support for this response. Specifically, the inclusion of age (Model 1) adds only 3% of explained variance, while the inclusion of the correctional ideology variable social distance adds another 5% of explained variance. Finally, adding measures of belief credibility and male rape myths adds an additional 7% of explained variance in this dependent variable for large jails. Comparing standardized coefficients across models also supports the notion that concern about social distance from inmates (β = −.19; Model 2) and perceptions of inmate credibility (β = .18; Model 3) are most important factors in determining support for these responses. The results of this regression analysis indicate that officers in large jails who support these responses are also more likely to be older, less concerned about social distance from inmates, and believe in the credibility of inmate victims.

Discussion

While the examination of coefficients and R² change statistics across models does not clearly delineate a specific factor as the most important either within small jails, within large jails, or between them, it does suggest that it is critical to consider some demographic factors, work-related variables, correctional ideology factors, and attitudes about gender and the culture of sexual assault. The specific variables in each category that matter seem to depend somewhat on the context and the dependent variable examined. The independent variables that are critical predictors of willingness to respond across three of the four analyses are officer beliefs about social distance from inmates and the credibility of inmate victims. It seems that if officers are comfortable with social proximity to inmates, and they believe that inmates are credible victims of sexual assault, they are more likely to support responses to sexual assault among inmates.

We will review other results here both by comparing results across facility size for each dependent variable examined, then we will examine differences and similarities across dependent variables within small jails and then within large jails. First, the support that officers indicate for talking to inmates about the risks of sexual assault differs only slightly according to whether they work in small or large facilities. While officers in both sizes of facilities are supportive of talking to inmates if they are comfortable with social proximity to inmates, officers in small jails are also supportive of talking to inmates if they perceive them as credible. Officers in larger facilities are then more likely to support talking to inmates when they have more experience. It may be the case that within both types of facilities, officers who get to know inmates better and have compassion toward them are, therefore, more comfortable talking to them. In larger jails, officers with experience may feel more comfortable talking to inmates, and while this may be a sign of maturity, confidence, or experience, it is a little unclear why this would be the case for larger facilities only.

When examining the analyses predicting support for the four-item index that measured officer support for encouraging inmate reporting and using proactive measures, the single consistency across jails is that perceiving inmates as credible increases the likelihood of supporting these types of responses. In smaller jails, officers who reject the idea that men cannot be raped are also willing to support these responses. Officers who work in larger jails are more supportive of adopting these responses if they are older and are more secure with less social distance from inmates.

In comparing results across smaller jail facilities only, if officers perceive inmate victims as credible, they are also more likely to support adopting responses to sexual assault. Officers in small jails who are comfortable with social proximity to inmates are also more likely to express comfort talking to them about sexual assault risks. Another interesting finding for small jails is that if officers believe male rape is a possibility, they will consider it important to be proactive to prevent it. It is not clear why this result emerges in small jails only, however. It may be that, for the proactive measures (such as using cell assignments and protective custody), the context of using these resources differs across facilities. Larger jails may differ from smaller jails in the accessibility and availability of these resources for the prevention of sexual assault.

For larger jails, age or experience are important predictors of support for responding to sexual assault. These may be indicators of adjustment to the job, experience talking with inmates, or learned comfort discussing difficult subjects. It may be that these officers are likely more comfortable managing inmates and addressing issues that affect them. Beliefs about social distance and perceptions of credibility are also relevant. This is not surprising given the results discussed above, and officers who are comfortable with social proximity to inmates and who believe inmate reports are more likely to encourage reporting and using resources to prevent sexual assault.

Perceptions of inmate credibility and beliefs about social distance from inmates are the most consistent predictors across models. Yet the context under which these effects are seen differs across jails. It may be that social distance from inmates makes it difficult for officers to feel as if they can talk to inmates about this issue. Social proximity to inmates in larger jails may be less acceptable to officers in that context, and, thus, more critical in terms of responding to instances of sexual assault. These may be considered rather progressive behaviors among officers in larger jails, and they may only feel comfortable adopting them if they feel more socially close to inmates. This seems intuitive, as social distance is one independent variable that measures professionalization of the officer job, as well as compassion and advocacy for inmates. In contrast to much of the literature discussed earlier, demographics and job-related characteristics are either not or are only rarely related to officer support for responding to sexual assault in these ways. Age and experience are the only variables that were relevant here, and they were only relevant for officers in larger jails. Contrary to what prior research would suggest about the relationship between age and attitudes toward victims, older officers in large jails were more progressive in supporting some responses to sexual assault in jails. Yet age was not relevant in all models. We also found a similarity to one of the very few prior studies to examine this issue: Similar to Eigenberg (1994), officers in our study who prefer less social distance from inmates are also more willing to support responding to instances of sexual assault among inmates. We did not find, however, that attitudes about homosexuality mattered in our sample.

A main limitation of this study is the lack of external generalizability. The sampling mechanism was intended to help further our understanding of the experiences and beliefs of correctional officers in jails. For example, as discussed in the review of the literature, attitudes about sexual assault are likely to vary significantly by education; yet, we cannot be certain that our sample is representative of the population from which it is drawn on this variable enough to impact the interpretation of findings. Regardless, the low response rate suggests that its findings can only be attributed to those officers who completed the survey. We do not have enough other information about the target population to determine if other factors make our sample unrepresentative and influence our findings. We are also unable to generalize findings to other correctional contexts in other areas, specifically to jail facilities outside Florida, or to prisons. The cross-sectional (versus ongoing) nature of the research is also problematic. These data were collected in 2008, and although we do not expect that jail contexts are dramatically different, more research needs to be conducted to continue to verify these findings. Therefore, more research needs to be continuously done in more jails nationwide.

Another limitation may be the reliance on self-reported attitudinal data, instead of reports of official actions taken regarding sexual assault. Although these data depend on accuracy and truthfulness of response, survey responses have been shown to be valid if their subjects believe their responses are anonymous (Aquilino, 1994; DeMaio, 1984; Jones & Forrest, 1992; Schwarz, Strack, Hippler, & Bishop, 1991; Turner, Lessler, & Devore, 1992). Finally, the overall models explain a relatively modest proportion of variance among the dependent variables. While the models are more robust in explaining the variance among dependent variables within smaller jails compared with larger jails, the explained variance within smaller jails is modest, and ranges only from 12% to 20%. Although the survey included many variables relevant to attitudes about sexual assault and correctional officer beliefs, there are additional variables that should be examined if we are to gain a better understanding of willingness to respond to instances of sexual assault in correctional facilities. One area that needs to be considered is research about helping behaviors, more specifically bystander intervention in relation to instances of sexual assault (see Banyard, 2011). It was beyond the scope of this survey, yet researchers should consider inclusion of this literature in informing survey design. Despite limitations, the results suggest important considerations for the unique context of jails. Jail facilities, unlike prisons, are likely to have fewer programmatic resources, given the turnover of its inmate population (Klofas, 1990; Miller, 1978; O’Toole, 1997; Sabol, 2008a, 2008b; Welsh, 1992). Given this, the results from this sample suggest that jail officers may be unexpectedly progressive in terms of their correctional ideology and attitudes toward inmates. That is, they are not the “uncaring guard” that defines traditional stereotypes of officers (Kauffman, 1981). This is critical in light of the fact that there are likely fewer treatment or programming staff in jails than in prisons (Stephan, 2001, 2008). In other words, correctional officers are not only well situated, but they may also be willing, to help battle the problem of sexual assault in jail facilities.

If inmate victims perceive that officials (including correctional officers) will respond to this problem, they may also increasingly come forward with reports of victimizations. In general, sexual assault is one of the most underreported crimes due partially to fear of stigma. Yet in correctional institutions, reluctance to report sexual assault can also be attributed to the fear of harassment and repeated victimization (Chonco, 1989; Lockwood, 1980; Weiss & Friar, 1974). Administrators who ensure that responses to these incidents are a priority are also likely to facilitate reporting and reduction of stigma. Training can assist both individuals and facilities to prepare themselves for responding to these incidents. Even with the provisions outlined in PREA, a report of sexual violence occurring in jails showed that jail facilities with high incidents of sexual assault did not have sufficient training about the law, or about how to investigate and prevent sexual assault (McFarland, Ellis, & Chunn, 2008). The findings yield themselves to other implications in terms of training. For example, given that older and more experienced officers are more willing to respond to these incidents in larger jails, a relatively simple training mechanism would be a mentoring program. Similar to a field training program, it would allow younger or inexperienced recruits to be matched up with more experienced, older officers for on the job training. In addition, training can be extended to emphasize believing reports of sexual assault, and the importance of having compassion for and advocating for inmates.

Despite the political attention this issue has received in the last decade, including the provision of resources for correctional agencies, there is little known about how correctional officers feel about responding to sexual assault in correctional facilities. Furthermore, we know of no other research that has examined how officers in jails feel about responding to these incidents. Although the culture of correctional environments may present a unique challenge to correctional officials attempting to combat these issues in their facilities, the provisions of PREA mandate action. There are increasing calls for administrators and those who work directly with inmates to take the issue of sexual assault seriously and to take necessary actions to protect inmates. Prior literature and the findings of the present research indicate that officers are not necessarily uncompassionate toward those who are incarcerated (Farkas, 1999; Toch & Klofas, 1982). The experience that people have while incarcerated is especially important to consider given that the majority, especially jail inmates, eventually return home to their communities (Petersilia, 2003; Travis, Soloman, & Waul, 2001).

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Notes

References

Anderson

(2004). Explaining negative rape victim perception: Homophobia and the male rape victim. Current Research in Social Psychology, 10, 43-57.

Applegate

B. K.

Sitren

A. H.

(2008). The jail and the community: Comparing jails in rural and urban contexts. The Prison Journal, 88, 252-269.

Aquilino

W. S.

(1994). Interview mode effects in surveys of drug and alcohol use: A field experiment. Public Opinion Quarterly, 58, 210-240.

Arthur

J. A.

(1994). Correctional ideology of Black correctional officers. Federal Probation, 58, 57-66.

Banyard

V. L.

(2008). Measurement and correlates of pro-social bystander behavior: The case of interpersonal violence. Violence and Victims, 23, 85-99.

Banyard

V. L.

(2011). Who will help prevent sexual violence: Creating an ecological model of bystander intervention. Psychology of Violence, 1, 216-229.

Beck

A. J.

(2002). Jail population growth: National trends and predictors of future growth. American Jails, 16, 9-14.

Beck

A. J.

Berzofsky

Caspar

Krebs

(2013). Sexual victimization in prisons and jails reported by inmates, 2011-12. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=263489

Beck

A. J.

Harrison

P. M.

Berzofsky

Caspar

Krebs

(2010). Sexual victimization in prisons and jails reported by inmates, 2008-09. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from http://www.bjs.gov/content/pub/pdf/svpjri0809.pdf

10.

Bottoms

B. L.

Goodman

G. S.

(1994). Perceptions of children’s credibility in sexual assault cases. Journal of Applied Social Psychology, 24, 702-732.

11.

Brownmiller

(1975). Against our will: Men, women, and rape. New York, NY: Simon & Schuster.

12.

Burn

S. M.

(2009). A situational model of sexual assault prevention through bystander intervention. Sex Roles, 60, 779-792.

13.

Burt

D. L.

DeMello

L. R.

(2003). Attribution of rape blame as a function of victim gender and sexuality, and perceived similarity to the victim. The Journal of Homosexuality, 43, 39-57.

14.

Burt

(1980). Cultural myths and supports for rape. Journal of Personality and Social Psychology, 38, 217-230.

15.

Chonco

(1989). Sexual assaults among male inmates: A descriptive study. The Prison Journal, 69, 72-82.

16.

Cook

C. L.

Lane

(2012). Examining differences in attitudes about sexual victimization among a sample of jail officers: The importance of officer gender and perceived inmate characteristics. Criminal Justice Review, 37, 191-213.

17.

Cook

C. L.

Lane

(2015). Blaming the victim: Perceptions about incarcerated sexual assault victim culpability among a sample of jail correctional officers. Victims & Offenders, doi: 10.1080/15564886.2015.1065531

18.

Crouch

B. M.

Alpert

G. P.

(1982). Sex and occupational socialization among prison guards: A longitudinal study. Criminal Justice and Behavior, 9, 159-176.

19.

Cullen

F. T.

Lutze

F. E.

Link

B. G.

Wolfe

N. T.

(1989). The correctional orientation of prison guards: Do officers support rehabilitation? Federal Probation, 53, 33-42.

20.

Davies

(2002). Male sexual assault victims: A selective review of the literature and implications for support services. Aggression and Violent Behavior, 7, 203-214.

21.

Davies

Rogers

(2006). Perceptions of male victims in depicted sexual assaults: A review of the literature. Aggression and Violent Behavior, 11, 367-377.

22.

Davies

Rogers

Bates

(2008). Blame toward male rape victims in a hypothetical sexual assault as a function of victim sexuality and degree of resistance. Journal of Homosexuality, 55, 533-544.

23.

DeMaio

T. J.

(1984). Social desirability and survey measurement: A review. In Turner

C. F.

Martin

(Eds.), Surveying subjective phenomena, Vol. 2 (pp. 257-281). New York, NY: Russell Sage.

24.

Dillman

(2000). Mail and Internet surveys: The tailored method (2nd ed.). New York, NY: John Wiley.

25.

Dillman

Dolsen

D. E.

Machlis

G. E.

(1995). Increasing response to personally-delivered mail-back questionnaires. Journal of Official Statistics, 11, 129-139.

26.

Dumond

R. W.

(2000). Inmate sexual assault: The plague that persists. The Prison Journal, 80, 407-414.

27.

Eigenberg

(1989). Male rape: An empirical examination of correctional officers’ attitudes toward rape in prison. The Prison Journal, 68, 39-56.

28.

Eigenberg

(1994). Rape in male prisons: Examining the relationship between correctional officers’ attitudes toward male rape & their willingness to respond to acts of rape. In Braswell

Montgomery

Lombardo

(Eds.), Prison violence (2nd ed., pp. 145-165). New York, NY: Anderson.

29.

Eigenberg

(2000a). Correctional officers’ definitions of rape in male prisons. Journal of Criminal Justice, 28, 435-449.

30.

Eigenberg

(2000b). Correctional officers & their perceptions of homosexuality, rape, & prostitution in male prisons. The Prison Journal, 80, 415-433.

31.

Farkas

M. A.

(1999). Correctional officer attitudes toward inmates and working with inmates in a “get tough” era. Journal of Criminal Justice, 27, 495-506.

32.

Fischer

G. J.

(1987). Hispanic and majority student attitudes toward forcible date rape as a function of differences in attitudes toward women. Sex Roles, 17, 93-101.

33.

Fishman

(1934). Sex in prison: Revealing sex conditions in American prisons. New York, NY: National Library.

34.

Frese

Moya

Megias

J. L.

(2004). Social perception of rape: How rape myth acceptance modulates the influence of situational factors. Journal of Interpersonal Violence, 19, 143-161.

35.

Frye

(2007). The informal social control of intimate partner violence against women: Exploring personal attitudes and perceived neighborhood social cohesion. Journal of Community Psychology, 35, 1001-1018.

36.

Gonsalves

V. M.

Walsh

Scalora

M. J.

(2012). Staff perceptions of risk for prison rape perpetration and victimization. The Prison Journal, 92, 253-273.

37.

Goodman

G. S.

Bottoms

B. L.

Herscovici

B. B.

Shaver

(1989). Determinants of the child victim’s perceived credibility. In Ceci

S. J.

Ross

D. F.

Toglia

M. P.

(Eds.), Perspectives on children’s testimony (pp. 1-22). New York, NY: Springer-Verlag.

38.

Gray

N. B.

Palileo

G. J.

Johnson

G. D.

(1993). Explaining rape victim blame: A test of attribution theory. Sociological Spectrum, 13, 377-392.

39.

Groth

A. N.

Burgess

A. W.

(1980). Male rape: Offenders and victims. American Journal of Psychiatry, 137, 806-810.

40.

Hensley

Koscheski

Tewksbury

(2005). Examining the characteristics of male sexual assault targets in a southern maximum-security prison. Journal of Interpersonal Violence, 20, 667-679.

41.

Hensley

Tewksbury

(2005a). Wardens’ perceptions of inmate fear of sexual assault: A research note. The Prison Journal, 85, 198-203.

42.

Hensley

Tewksbury

(2005b). Wardens’ perceptions of prison sex. The Prison Journal, 85, 186-197.

43.

Herek

G. M.

(1988). Heterosexual’s attitudes toward lesbians and gay men: Correlates and gender differences. The Journal of Sex Research, 25, 451-477.

44.

Howard

J. A.

(1984a). The “normal” victim: The effects of gender stereotypes on reactions to victims. Social Psychology Quarterly, 47, 270-281.

45.

Howard

J. A.

(1984b). Societal influences on attribution: Blaming some victims more than others. Journal of Personality and Social Psychology, 47, 494-505.

46.

Jackson

Ammen

(1996). Race and correctional officers’ punitive attitudes toward treatment programs for inmates. Journal of Criminal Justice, 24, 153-166.

47.

Jackson

Nuttall

(1994). Effects of gender, age, and history of abuse on social workers’ judgments of sexual abuse allegations. Social Work Research, 18, 105-113.

48.

Jacobs

J. B.

(1980). The prisoner’s rights movement and its impact, 1960-1980. In Morris

Tonry

(Eds.), Crime and justice: A review of research (Vol. 2, pp. 429-470). Chicago, IL: University of Chicago Press.

49.

Jacobs

J. B.

Kraft

L. J.

(1978). Integrating the keepers: A comparison of Black and White prison guards in Illinois. Social Problems, 25, 304-318.

50.

Jenkins

M. J.

Dambrot

F. H.

(1987). The attribution of date rape: Observer’s attitudes and sexual experiences and the dating situation. Journal of Applied Social Psychology, 17, 875-895.

51.

Jimenez

J. A.

Abreu

J. M.

(2003). Race and sex effects on attitudinal perceptions of acquaintance rape. Journal of Counseling Psychology, 50, 252-256.

52.

Jones

E. F.

Forrest

J. D.

(1992). Under-reporting of abortion in surveys of U.S. women: 1986-1988. Demography, 29, 113-126.

53.

Jordan

(2004). Beyond belief: Police, rape, and women’s credibility. Criminal Justice, 4, 29-59.

54.

Jurik

N. C.

(1985). Individual and organizational determinants of correctional officer attitudes toward inmates. Criminology, 23, 523-540.

55.

Kassing

L. R.

Prieto

L. R.

(2003). The rape myth and blame-based beliefs of counselors-in-training toward male victims of rape. Journal of Counseling & Development, 81, 455-461.

56.

Kauffman

(1981). Prison officers’ attitudes and perceptions of attitudes: A case of pluralistic ignorance. Journal of Research in Crime & Delinquency, 18, 272-294.

57.

Kellar

Jaris

Manboah-Roxin

(2001). Texas jail survey 2001: A status report. Austin: Texas Commission on Jail Standards. Retrieved from http://www.tcjs.state.tx.us/index.php?linkID=300

58.

Kifer

Hemmens

Stohr

M. K.

(2003). The goals of corrections: Perspectives from the line. Criminal Justice Review, 28, 47-69.

59.

Kinkade

Leone

Semond

(1995). The consequences of jail crowding. Crime & Delinquency, 41, 150-161.

60.

Klofas

(1990). The jail and the community. Justice Quarterly, 7, 69-102.

61.

Klofas

Toch

(1982). The guard subculture myth. Journal of Research in Crime & Delinquency, 19, 238-254.

62.

Kopper

B. A.

(1996). Gender, gender identity, rape myth acceptance, and time of initial resistance on the perception of acquaintance rape blame and avoidability. Sex Roles, 34, 81-93.

63.

Lockwood

(1980). Prison sexual violence. New York, NY: Elsevier.

64.

Maahs

Pratt

(2001). Uncovering the predictors of correctional officers’ attitudes and behaviors: A meta-analysis. Corrections Management Quarterly, 5, 13-19.

65.

Mays

G. L.

Thompson

J. A.

(1988). Mayberry revisited: The characteristics and operations of America’s small jails. Justice Quarterly, 5, 421-440.

66.

McFarland

S. T.

Ellis

C. A.

Chunn

(2008). Report on rape in jails in the U.S. Washington, DC: U.S. Department of Justice, Office of Justice Programs. Retrieved from http://www.ojp.usdoj.gov/reviewpanel/pdfs/prea_finalreport_081229.pdf

67.

McMahon

(2010). Rape myth beliefs and bystander attitudes among incoming college students. Journal of American College Health, 59, 3-11.

68.

Miller

E. E.

(1978). Jail management: Problems, programs, and perspectives. Lexington, MA: D.C. Heath.

69.

Monster

A. N.

Jeglic

E. L.

(2009). Prison warden attitudes toward prison rape and sexual assault. The Prison Journal, 89, 65-78.

70.

Mori

Bernat

J. A.

Glenn

P. A.

Selle

L. L.

Zarate

M. G.

(1995). Attitudes toward rape: Gender and ethnic differences across Asian and Caucasian college students. Sex Roles, 32, 457-467.

71.

Nacci

P. L.

Kane

T. R.

(1983). The incidence of sex & sexual aggression in federal prisons. Federal Probation, 7, 31-36.

72.

Nacci

P. L.

Kane

T. R.

(1984). Sex and sexual aggression in federal prisons: Inmate involvement and employee impact. Federal Probation, 48, 46-53.

73.

Nagel

Matsuo

McIntyre

K. P.

Morrison

(2005). Attitudes toward victims of rape. Journal of Interpersonal Violence, 20, 725-737.

74.

National Institute of Corrections & the Moss Group, Inc. (2006). Staff perspectives: Sexual violence in adult prisons and jails. Retrieved from http://nicic.gov/Library/021619

75.

O’Donohue

Elliott

Nickerson

Valentine

(1992). Perceived credibility of children’s sexual abuse allegations: Effects of gender and sexual attitudes. Journal of Interpersonal Violence, 7, 147-155.

76.

O’Toole

(1997). Jails and prisons: The numbers say they are more different than generally assumed. American Jails, 11, 27-31.

77.

Perrott

Webber

(1996). Attitudes towards male and female victims of sexual assault: Implications and services to the male victim. Journal of Psychology & Human Sexuality, 8, 19-38.

78.

Petersilia

(2003). When prisoners come home: Parole and prisoner reentry. New York, NY: Oxford University Press.

79.

Pollard

(1992). Judgments about victims and attackers in depicted rapes: A review. The British Psychological Society, 31, 307-326.

80.

Poole

E. D.

Regoli

R. M.

(1980). Role stress, custody orientation, and disciplinary actions: A study of prison guards. Criminology, 18, 215-226.

81.

Res

1435, 108 Cong. (2003) (enacted).

82.

Ruddell

Mays

G. L.

(2007). Rural jails: Problematic inmates, overcrowded cells, and cash-strapped counties. Journal of Criminal Justice, 35, 251-260.

83.

Sabol

W. J.

(2008a). Jail inmates at midyear 2007. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

84.

Sabol

W. J.

(2008b). Prison inmates at midyear 2007. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

85.

Schwarz

Strack

Hippler

H. J.

Bishop

(1991). The impact of administration mode on response effects in survey measurement. Applied Cognitive Psychology, 5, 193-212.

86.

Sheldon

J. P.

Parent

S. L.

(2002). Clergy’s attitudes and attributions of blame toward female rape victims. Violence Against Women, 8, 233-256.

87.

Stephan

J. J.

(2001). Census of jails, 1999. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

88.

Stephan

J. J.

(2008). Census of state and federal correctional facilities, 2005. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

89.

Struckman-Johnson

(1992). Acceptance of male rape myths among college men and women. Sex Roles, 27, 85-100.

90.

Struckman-Johnson

(2000). Sexual coercion rates in seven Midwestern prison facilities for men. The Prison Journal, 80, 379-390.

91.

Struckman-Johnson

Rucker

Bumby

Donaldson

(1996). Sexual coercion reported by men and women in prison. The Journal of Sex Research, 33, 67-76.

92.

Teske

R. H. C.

Williamson

H. E.

(1979). Correctional officers’ attitudes toward selected treatment programs. Criminal Justice and Behavior, 6, 59-66.

93.

Toch

Klofas

(1982). Alienation and desire for job enrichment among correction officers. Federal Probation, 46, 35-44.

94.

Travis

Soloman

A. L.

Waul

(2001). From prison to home: The dimensions and consequences of prisoner reentry. Washington, DC: The Urban Institute.

95.

Turner

C. F.

Lessler

Devore

(1992). Effects of mode of administration and wording on reporting of drug use. In Turner

C. F.

Lessler

J. T.

Gfroerer

J. C.

(Eds.), Survey measurement of drug use: Methodological studies (pp. 177-220). Washington, DC: Department of Health and Human Services.

96.

Wakelin

Long

K. M.

(2003). Effects of victim gender and sexuality on attributions of blame to rape victims. Sex Roles, 49, 477-487.

97.

Weiss

Friar

(1974). Terror in the prisons: Homosexual rape & why society condones it. Indianapolis, IN: Bobbs-Merrill.

98.

Welsh

W. N.

(1992). The dynamics of jail reform litigation: A comparative analysis of litigation in California counties. Law & Society Review, 26, 591-625.

99.

White

B. H.

Robinson Kurpius

S. E.

(2002). Effects of victim sex and sexual orientation on perceptions of rape. Sex Roles, 46, 191-200.

100.

Whitehead

Linquist

Klofas

(1987). Correctional officer correctional ideology: A replication of the Klofas-Toch measure. Criminal Justice and Behavior, 14, 468-486.

101.

Wooden

Parker

(1982). Men behind bars: Sexual exploitation in prison. New York, NY: Plenum.

102.

Xenos

Smith

(2001). Perceptions of rape and sexual assault among Australian adolescents and young adults. Journal of Interpersonal Violence, 16, 1103-1119.