Prison Culture as Rape Supportive: Applying Importation and Deprivation Models to Examine Inmate Beliefs

Abstract

While research examining sexual violence in prison has increased over the past 15 years, relatively scant attention has been paid to rape supportive beliefs and the factors influencing inmate adherence to these beliefs. Given the demonstrated role from studies outside the prison context that rape supportive beliefs have on sexual violence, important parallels can be drawn from studying the inmate population. Importation and deprivation models have been used to explain how inmates adapt to prison life—whether attitudes and behaviors are imported from their lives outside prison or are developed in prison due to the conditions of incarceration. Using a sample of male and female inmates (n = 875) from a large state prison system in the southern United States, the researchers explored the degree to which inmate rape supportive beliefs (IRSB) were influenced by variables indicative of importation or deprivation models of prison adaption. Findings revealed greater support for importation variables among both male and female inmates. With some variation, measures such as gender, age, ethnicity, and education were significant in explaining IRSB, similar to findings of studies on rape supportive beliefs among noninmate samples. Because these beliefs can manifest in problematic attitudes and behaviors among inmates, such as prison sexual violence and the underreporting of such violence, it is important that correctional administrators understand this relationship and take steps to lessen IRSB. Since IRSB largely correlate with factors unrelated to conditions of confinement, efforts to educate and sensitize inmates to prison sexual violence can replicate best practices based on populations outside of prisons.

Keywords

rape supportive beliefs prison culture importation deprivation

With prison inmates continuing to experience and/or allege sexual violence at increased rates, examining potential causes of such violence is important (Beck et al., 2013; Rantala, 2018). According to Rantala (2018), there was not only an increase in the rate of allegations of sexual violence in correctional institutions across the United States from 3.9 to 11.0 per 1,000 inmates from 2011 to 2015 but also the rate of substantiated cases doubled from 0.4 per 1,000 inmates to 0.8 per 1,000 inmates in the same time frame. Descriptions of prison culture suggest the carceral environment corresponds to characteristics of rape supportive cultures: traditional gender roles, hostility toward women, and acceptance of violence. Little work exists in describing prison culture as a rape supportive culture. Intercultural studies of rape supportive beliefs (Boakye, 2009; Xue et al., 2019), as well as intracultural studies within the United States (Haywood & Swank, 2008; Murnen & Kohlman, 2007), exist. The application of a rape supportive belief framework to prison culture may therefore be a fruitful area to advance explanations of prison sexual violence.

While prisons have been described as “total institutions,” largely disconnected from mainstream society (Goffman, 1961), current understandings of prisons suggest institutional boundaries are instead porous (Farrington, 1992; Jenness et al., 2019; McCorkle et al., 1995). Given that prison culture may be a rape supportive culture, it is important to examine the role that incarceration and prison conditions play in rape supportive beliefs among inmates as a research objective (Sisco & Becker, 2007). This is especially relevant as recent research demonstrates that male inmates subscribe to more rape supportive beliefs than male noninmates (Debowska et al., 2015; Debowska et al., 2016). Additionally, higher endorsements of rape supportive beliefs have been linked to increased blaming of victims for sexual violence and decreased likelihood in reporting sexual violence among incarcerated women (Grubb & Turner, 2012; Heath et al., 2013).

This research fills a gap in the literature by applying a rape supportive belief framework to prison culture. This research attempts to determine from where inmate rape supportive beliefs (IRSB) are derived by assessing if inmates import IRSB or if the prison environment affects the adoption of these beliefs. Such an examination is justified as prison administrators continue to address the issue of prison sexual violence (Rantala, 2018). Determining how best to prevent and respond to sexual violence could rely on a better understanding of rape supportive beliefs and how these beliefs impact inmate behavior. Additionally, determining if these beliefs vary by gender would allow for gender-specific responses on behalf of correctional administrators.

Review of the Literature

Prison Culture as Rape Supportive

To explain levels of sexual violence in the United States, US culture has been labeled “rape supportive” (Brownmiller, 1975) and this concept has been explored further through examinations of male identities shaped by US culture (Keith, 2017). These cultures are theorized to be the products of patriarchal social structures, which tolerate the sexual exploitation of women, and are reinforced by a patriarchal value system that differentially socializes men and women to their respective gender roles (Johnson & Johnson, 2021). A patriarchal value system promotes rape supportive beliefs, sometimes referred to as rape myths. Lonsway and Fitzgerald (1994, p. 134) define rape myths as, “attitudes and beliefs that are generally false but are widely and persistently held, and that serve to deny and justify male sexual aggression against women.” Researchers have examined rape supportive beliefs among male and female college students, finding that rape supportive beliefs remain prevalent on college campuses. In a recent study, Cannan et al. (2018) found that male university students involved in fraternities were significantly more likely to believe rape myths. Rape supportive beliefs also instruct women to blame themselves if sexually victimized as well as decrease the likelihood of reporting victimization due to fears of not being believed (Deming et al., 2013; Heath et al., 2011).

Evidence suggests that incarceration may be associated with an increased subscription to rape supportive beliefs based on studies comparing male university students to male inmates (Debowska et al., 2015; Debowska et al., 2016). There is also evidence to support that sexual victimization in prison is due in part to the fact that prison culture is rape supportive. Even though 93% inmates are men (Bronson & Carson, 2019, p. 3), in national and state-based studies incarcerated women still report higher rates of prison sexual violence (Beck et al., 2010; Wolff et al., 2006; Wolff et al., 2007). Further, the lifetime sexual violence prevalence rates of incarcerated females range between 56% and 82% (Karlsson & Zielinski, 2021) compared to nonincarcerated women who experience lifetime sexual violence prevalence rates around 21.3% (Smith et al., 2018). This demonstrates a potent difference between sexual victimization of those incarcerated and those who are not. Also, patriarchal features and other hallmarks of traditional masculinity are still present in prisons (Morse & Wright, 2021). Some penologists (Lutze, 2003) refer to prison as an “ultra-masculine” environment where policy and practices were put in place by men to govern mostly male inmates. This leads women to adapt to an environment that is more patriarchal than the community outside prison. Also, inmates who do not adapt to this ultra masculine cultural regime may experience difficulties in prison. In national and state-level examinations of prisons, lesbian, gay, bisexual, transgender, and queer inmates are at increased risk for prison sexual violence victimization (Beck et al., 2013; Hensley et al., 2003; Jenness et al., 2019).

Stories of prison sexual violence play an important role in inmate acculturation. While acclimating to prison, new inmates socially situate prison sexual violence after hearing others share stories of prison sexual violence (Fleisher & Krienert, 2009). Prison sexual violence becomes a social fact; if one wants to avoid prison sexual violence, inmates must learn the commonly held lessons of prison culture. Inmate cultural beliefs about prison sexual violence resemble many of the same rape supportive beliefs found in society at large. Face-to-face interviews with a nationally representative sample of male and female general population inmates in 30 prisons led Fleisher and Krienert (2009, p. 83) to conclude, “Assumptions [about prison sexual violence] are predicated on physical and mental weakness, a ‘blame the victim’ sexual victimization philosophy, and antipathy toward victims’ pain and suffering.” This is supported even with samples of incarcerated women; inmates view prison sexual violence as victim precipitated for “some personal, social, or economic transgression” (Owen et al., 2017, p. 126). Inmates’ cultural beliefs also contextualize the acceptability of prison sexual violence for retaliation for a violation of prison norms (Fleisher & Krienert, 2009; Owen et al., 2017).

As with rape supportive beliefs outside of prison, inmates reclassify and excuse some forms of prison sexual violence, allowing it to continue. Additionally, the impetus for the prevention of sexual victimization is placed on victims. Being a victim of sexual assault is to some degree seen as an issue of character, specifically in the inability to protect ones’ self from victimization on the outside (Johnson & Johnson, 2021) or while incarcerated (Fleisher & Krienert, 2009; Owen et al., 2017). Also, if an inmate fails to retaliate violently for being attacked, inmates may interpret the initial act of prison sexual violence as consensual (Fleisher & Krienert, 2009). When inmates fail to conform to the lessons found in prison cultural messages, they may be blamed by others for being assaulted, similar to victims outside of prison (Fleisher & Krienert, 2009; Owen et al., 2017). While these sorts of rape supportive beliefs have been studied in wider mainstream US culture, they have been much less frequently applied to prisons and inmates (see Fleisher & Krienert, 2009).

Prison Adjustment

The prison experience is inherently different than life outside of prison. If inmates subscribe to rape supportive beliefs, a major issue focuses on whether these beliefs derive from inmate characteristics and prior dispositions or from the prison experience. Two general models have been used to explain inmates’ adaptation to prison, importation and deprivation.

Importation. It suggests that inmates adapt to prison using strategies or characteristics that existed prior to entering prison (Irwin & Cressey, 1962). Because theoretical concepts used to directly test the importation model are rarely available prior to an inmate’s entry into prison, researchers have relied on proxy measures. For instance, traits such as race, age, marital status, and education are commonly used to demonstrate support for the importation model (Goncalves et al., 2014; Lahm, 2008; Mears et al., 2013). Homophobic attitudes in prison were predicted in the same manner as race, gender, and sexual orientation predict homophobic attitudes outside of prison (Hensley, 2000). Because those variables relate to homophobic attitudes in the same manner in prison and outside of prison, the missing key theoretical concepts not measured prior to inmates entering prison were believed to be present in the same manner.

If importation variables explain IRSB, previously found characteristics that predict acceptance of rape supportive beliefs outside of prison would predict IRSB. Previous researchers have found variables linked to importation to be important in explaining rape supportive beliefs. Gender has emerged as a strong and consistent predictor in meta-analyses, with men much more likely to adhere to rape supportive beliefs (Anderson et al., 1997; Suarez & Gadalla, 2010). White respondents are less likely to endorse rape supportive beliefs compared to members of other groups (Anderson et al., 1997; Suarez & Gadalla, 2010). Also, among meta-analyses, other factors associated with the importation model but not consistently related to rape supportive beliefs, or not yet tested for relationships with rape supportive beliefs, include age, education level, prior incarceration, and marital status (Anderson et al., 1997; Kassing et al., 2005; Suarez & Gadalla, 2010). Additionally, factors such as age, prior incarceration, and education level are variables representative of the importation model (Dhami et al., 2007; Gover et al., 2008).

Deprivation. It proposes that when inmates enter prison, they have restricted access to the outside world and experience loss of their former rights, lifestyle, and goods within a unique set of circumstances found only in this environment (Goffman, 1961; Sykes, 1958). As a result, they develop a culture around these shared deprivations and experiences. The degree of deprivation is believed to impact an inmate’s adjustment to the prison environment. How much time served in prison and the amount of time remaining on their sentence impacts their adjustment. Other factors related to the deprivation model include perceived relationship with staff and the security level/restrictiveness of the housing assignment (Dhami et al., 2007; Gover et al., 2008).

Fleisher and Krienert (2009) stated that over time inmates are acculturated to the prison worldview. Due to the sex-segregated nature of prison, inmates have adopted institutional-specific cultural beliefs that provide a framework for informing inmates of the accepted beliefs surrounding prison sexual violence. While most inmates detest prison sexual violence, there are highly contextualized definitions of prison sexual violence, even though sex is unwanted or forced. These beliefs have been found to be consistently held across prisons. In the deprivation model, there may be little or no difference between men’s and women’s rape supportive beliefs if the prison experience is uniform and pervasive (Fleisher & Krienert, 2009). As time served increases it would be expected that inmates may be more accepting of the contextualized construction of what constitutes sexual assault and the victim blaming beliefs found in prison (Fleisher & Krienert, 2009), thus subscribe to more IRSB.

The Current Study

There have been calls to examine inmates’ attitudes about prison sexual violence (Sisco & Becker, 2007). The objective of this study was to determine whether variables associated with importation and deprivation models of prison adjustment impacted the acceptance of IRSB. If the importation perspective affects IRSB, then it would be expected that preprison characteristics predict these beliefs in ways consistent with results from nonincarcerated samples. If the deprivation model influences IRSB, then variables related to confinement should significantly predict IRSB. Additionally, we explore whether importation and deprivation factors impact men and women inmates differently when it comes to their rape supportive beliefs.

Method

Data Collection

The project was designed to examine the intersection of prison culture and prison sexual violence, and the method has been described in previously published research using these data (Fowler et al., 2010).¹ Sample size was not calculated prior to the beginning of the study. Instead, all state-run prisons from a large southern prison system were stratified by the number of officially reported sexual assaults into one of three equally distributed strata of lowest, moderate, and highest numbers of assaults. Convenience sampling was utilized to select three prisons from each stratum. Inmates from each prison were sampled randomly; simple and systematic random sampling procedures were used for women and prison administrators randomly selected men from the general population.

Data collection began with groups of inmates, between 30 and 101, being issued voluntary passes to participate in the study. If an inmate agreed, they were escorted to available rooms in the prison for participation. After arriving to the survey location, inmates issued informed consent after researchers offered written and verbal descriptions of the project and gave assurances of confidentiality. Inmates not consenting were returned to their regularly scheduled activities. Correctional officers remained outside the survey area or near the exits, out of eyesight of the inmates’ responses. Inmates completed the self-administered questionnaire offered in English or Spanish, with most completing the survey in less than 45 minutes. In all, 935 inmates participated, representing 58.5% of those sampled.

Measures

Inmate rape supportive beliefs. The dependent variable is a composite scale measuring IRSB, with higher scores indicating greater adherence to rape supportive beliefs. This scale has been used before and was developed to address the prison context not present in other measures of rape supportive beliefs and correlated with how inmates define incidents of prison sexual violence (Fowler et al., 2010). The IRSB scale items were based on modified versions of both Burt’s (1980) Rape Myth Acceptance Scale and Deitz, et al.’s (1982) Rape Empathy Scale; and derived from face-to-face interviews with male and female inmates in the prison system under study regarding prison sexual violence. Interviewed inmates later critiqued the proposed questionnaire during a series of pilot tests among different inmates. Responses to IRSB questionnaire statements were 5- and 7-point Likert-scales with anchors of “strongly agree” and “strongly disagree.” Of the initial 28 items, 9 were omitted from the scale. Three items exhibited negative item-to-total correlations and six additional items were excluded iteratively as their item-to-total correlations were low (< .2) and they did not reliably contribute to the scale (Burgess, 2007). A parallel analysis resulted in four eigenvalues exceeding the 95% critical value of the corresponding generated means. As such, four components were extracted in a principal components analysis with varimax rotation explaining 43% of the variance. Item factor loadings greater than .4 were considered for interpretative purposes for individual factors. One item with the highest loading of .37 was retained in the analysis as it increased the reliability of the total scale (Burgess, 2007). Subscales consisted of 3, 4, 4, and 8 items with intercorrelations ranging from .26 to .47. The final 19 items were standardized and summed to form a scale that produced a Cronbach’s α = .80. The IRSB items are listed in the Appendix. Following Burgess’ (2007) approach, respondents with more than two missing values on the IRSB scale items were omitted from the analysis, resulting in a final sample of 875 participants. The expectation likelihood method in SPSS 15.0 was used to impute missing values for these items among the remaining participants.

Importation. These variables represent characteristics that were present when inmates entered prison or allowed for a current connection to the outside community. Here, they represent mostly demographic variables. Researchers identified inmates’ gender by adding this variable based on whether the prison selected was a men’s or women’s prison (male = 1, female = 0). Also, inmates were asked to indicate their race/ethnicity with their options as White, African-American/Black, Hispanic/Latino(a), and Other. Each of the resulting four categories was dummy-coded for later analyses. Inmates were asked about marital status. The single attribute was dummy-coded and used in the analyses. Inmates self-reported their age in years. Respondents’ level of education was measured by asking inmates to indicate the highest level of education or grade completed. Response categories included less than high school, high school diploma/GED, and more than high school. Each of these three categories was dummy-coded for later analyses. Prior incarceration was established by asking if this was their first time as an inmate in the state system where the study was conducted; no was coded 1 and yes was coded 0. Also, a previous sexual victimization experience (Vonderhaar & Carmody, 2015) have in some research been shown to significantly relate to rape supportive beliefs. Lifetime victim is a variable gauging inmates’ sexual abuse at any point in their lives. Respondents were asked if they had ever been sexually abused or assaulted in their lives.² Inmates that answered yes (1) were considered lifetime victims, while those answering no or not sure were coded 0.

Deprivation. Deprivation variables represent the restrictiveness of the prison environment and the degree to which an inmate is removed from society and likely to identify with the prison culture. Inmates were asked how long of a sentence they were serving. Respondents self-reported their overall sentence length in years and months, which was converted to the number of months. For sentence left, inmates were asked how much time they had left on their sentence and responded with the number of years and months, which was converted to number of months. Inmates were asked what year they were processed and admitted to the prison system. The difference between that year and survey year represents time served in years. Longer sentence lengths and years in prison represent greater degrees of deprivation. The greater length of sentence left suggests that an inmate is further from being released from prison and thus experiencing greater deprivation. Furthermore, inmates responded to several items regarding their beliefs about officers similar to those in previous research (Gover et al., 2008): staff members treat inmates fairly, staff members care about inmates, and staff members can be trusted. Response categories were anchored by (1) strongly disagree and (5) strongly agree. The responses were summed to form staff treatment, where higher scores represent more favorable perceptions of staff–inmate interactions and a less harshly perceived regime. These items’ internal consistency was α = .69. Also, inmates were asked to indicate their living arrangements. Dorms typically represent less restrictive security environments and are intended to function as a proxy for security. Their responses were coded dorm (1) or cell (0). In-prison victim captures whether inmates were victims of sexual assault by asking them if anyone had attempted to sexually assault them or actually sexually assault them while in custody. If inmates answered yes (1), they were considered victims; those answering no or not sure were coded 0. For the variable know a victim, inmates were also asked if they knew anyone, not including themselves, who had ever been the victim of sexual assault in prison. Inmates who answered yes were coded 1, and answers of no or not sure were coded 0. Past year was determined if the inmate knew someone who was a victim of prison sexual violence and if the sexual assault happened within the past 12 months. Yes responses were coded 1 and no and not sure were coded 0. Answering “yes” to any of the above three items indicates that events occurring in the prison environment result in the deprivation of security, thus, indicating higher degrees of deprivation.

Plan of Analysis

To assess how variables associated with the importation or deprivation models explain IRSB, a hierarchical linear regression strategy was used. First, IRSB was regressed over importation variables. Second, IRSB was regressed over deprivation variables. Third, IRSB was regressed over both sets of variables, and changes in the models were examined to determine if one model is better able to predict IRSB. Last, gender-specific linear regression models were estimated to determine if importation and deprivation variables impact male and female inmates differently.

Results

Sample Description

Table 1 displays the descriptive statistics for the inmate sample. A little over half of the sample are male (54%), and 40% of the inmates have never been married. There is a wide range in ages (19–78), but the average is 40 years. Nearly equal amounts of inmates are divided between having less than a high school diploma, diploma as the highest level of education, and obtaining more education after their diploma. Most inmates were non-White; 40% of inmates chose their race/ethnicity as Black, 37% indicated that they were White, 16% indicated they were Hispanic, and 7% chose some other category as their race/ethnicity. Nearly 45% of the sample had previously spent time in prison. On average, the inmates were sentenced to over 21.5 years and had 13.8 years remaining (depicted in months in Table 1). The average amount of time served was about 8 years. The mean of the staff treatment variable was below the hypothetical midpoint, indicating that inmates view staff–inmate relations in a negative light and see the prison regime as restrictive. Over 60% of the inmates were housed in lighter security dormitories as opposed to cells. The dependent variable, IRSB, is the result of adding standardized items, and the mean is zero. Over half of the inmates, 53%, had scores less than the mean. This is a large and diverse sample of inmates. As such, an analysis of this sample could provide a better understanding of how characteristics such as gender, race, and age impact IRSB.

Table 1.
Descriptive Statistics for Inmates.

N % M SD Range

IRSB 0.0 8.82 –19.20–37.02

Gender (male) 475 54.3

Single 352 40.2

Prior incarceration 382 43.7

Race

White 327 37.4

Black 346 39.5

Hispanic 142 16.2

Other 60 6.9

Age 40.34 10.63 19–78

Education

Less than HS/GED 271 31.3 .31

HS/GED 306 35.2 .35

More than HS 292 33.6 .33

Sentence length^a 259.12 167.17 3–480

Sentence left^a 165.82 149.13 0–480

Time served^b 8.02 6.83 0–38

Staff treatment 6.00 2.78 3–15

Dorm 537 61.4

Lifetime victim 388 44.3

In-prison victim 133 15.2

Know a victim 345 39.4

Past year 149 17.0

Note. IRSB = inmate rape supportive beliefs; HS = high school diploma; GED = general education development diploma.

^aMeasured in months. ^bMeasured in years.

Bivariate Relationships and Multicollinearity

An examination of correlations between IRSB and the importation variables (gender, race/ethnicity, single, education, and age³) yielded significant correlations in the expected direction based on findings from nonincarcerated samples. Additionally, prior incarceration positively correlated with IRSB. Lifetime victim of sexual abuse was significantly and negatively correlated with IRSB. Among the deprivation variables, there was much less support with few of the correlations emerging as significant. Living in a dormitory significantly and negatively correlated with IRSB. Also, knowing a victim of prison sexual assault significantly and negatively correlated with IRSB. Sentence length, sentence left, time served, and staff treatment did not significantly correlate with the dependent variable. Knowing prison sexual assault victims and knowing a victim within the past year were not significantly correlated with IRSB.

After inspecting the bivariate correlations, there was a concern for multicollinearity as sentence length substantially correlated with sentence left (r > .8). The variance inflation factors (VIFs) for these two variables were 4.9 for sentence length and 3.6 for sentence left, with 4 suggested as a cut-off value for diagnosing problematic levels of multicollinearity (Fox, 1991). All other VIFs were well below 4, with an average VIF below 2. Sentence length also strongly correlated with time served (r > .5) and more weakly (and nonsignificantly) correlated with IRSB compared to sentence left. Sentence length was eliminated from the regression analysis. All predictors’ VIFs remained under 4, with an average VIF below 2 after the variable deletion, and multicollinearity was assumed not to be problematic.

Regression Model

The results of the hierarchical OLS regression models are displayed in Table 2.⁴ The importation regression model significantly predicted IRSB, F(11, 857) = 9.69, p < .001. The model R² = .11. Five of the importation variables exhibited significant relationships with IRSB—having never married, self-identifying as Hispanic, age, and both categories of education. Inmates that have never been married had increased levels of IRSB and explained approximately 1% of the variance in IRSB. Identifying as Hispanic compared to those who identified as White non-Hispanic had higher levels of IRSB, which explained approximately 1% of the variance in IRSB. Age was positively correlated with IRSB and explained about 1% of the variance in IRSB. Compared to those who had not received a high school education, having a high school diploma decreased levels of IRSB and accounted for less than 1% of the variance in IRSB. Having an education level higher than a high school diploma was significant compared to not having a high school diploma. This educational category explained about 3.5% of the variance in IRSB.

Table 2.
Linear Regressions of Importation and Deprivation Variables on Inmate Rape Supportive Beliefs.

Importation Deprivation Full

B (95% CI) β sr B ( 95% CI) β sr B (95% CI) β sr

Gender^a 1.20 (–0.13, 2.53) 0.07 .06 – – – 1.71* (0.19, 3.23) 0.10 .07

Single 2.17 (0.89, 3.46) 0.12 .11 – – – 2.25 (0.97, 3.54) 0.13 .11

Prior incarceration 0.34 (–0.90, 1.58) 0.02 .02 – – – 0.36 (–0.88, 1.60) 0.02 .02

Black^b 1.03 (–0.35, 2.41) 0.06 .05 – – – 0.71 (–0.69, 2.11) 0.04 .03

Hispanic^b 2.88 (1.14, 4.62) 0.12 .11 – – – 2.82 (1.08, 4.55) 0.12 .10

Other^b 2.33 (–0.02, 4.68) 0.07 .06 – – – 2.26 (–0.07, 4.60) 0.06 .06

Age 0.12 (0.06, 0.18) 0.15 .12 – – – 0.13 (0.06, 0.20) 0.16 .12

Age² 0.00 (–0.00, 0.01) 0.04 .04 – – – 0.00 (–0.00, 0.01) 0.03 .03

HS/GED^c –1.72* (–3.11, –0.34) –0.09 –.08 – – – –1.57* (–2.96, –0.19) –0.09 –.07

More than HS^c –4.43 (–5.90, –2.97) –0.24 –.19 – – – –4.11 (–5.57, –2.64) –0.22 –.18

Lifetime victim –0.70 (–2.08, 0.61) –0.04 –.03 – – – –0.72 (–2.22, 0.79) –0.04 –.03

Sentence left^d – – – –0.00 (–0.01, 0.00) –0.03 –.03 –0.00 (–0.01, 0.00) –0.04 –.04

Time served^e – – – 0.11* (0.20, 0.20) 0.09 .08 0.00 (–0.09, 0.10) 0.00 .00

Staff treatment – – – 0.13 (–0.08, 0.34) 0.04 .04 0.09 (–0.12, 0.29) 0.03 .03

Dorm – – – –1.32* (–2.52, –0.12) –0.07 –.07 –0.34 (–1.59, 0.91) –0.02 –.02

In-prison victim – – – –0.28 (–1.94, 1.38) –0.01 –.01 0.73 (–1.10, 2.55) 0.03 .03

Know a victim – – – –3.02 (–4.47, –1.56) –0.17 –.14 –2.77 (–4.19, –1.36) –0.15 –.12

Past year – – – 1.49 (–0.35, 3.32) 0.06 .05 2.01* (0.24, 3.79) 0.09 .07

Constant –0.53 – – 0.44 – – –0.11 – –

R² .111 .034 .130

Note. B = unstandardized regression coefficient; β = standardized regression coefficient; CI = confidence interval; sr = semi-partial correlation; HS = high school diploma; GED = general education development diploma.

^aFemale is the reference category. ^bWhite is the reference category. ^cLess than high school is the reference category. ^dMeasured in months. ^eMeasured in years.

p < .05. p < .001.

The deprivation regression model in Table 2 significantly predicted IRSB, F(7, 867) = 4.37, p < .001. The model R² = .03. Three of the variables were significantly correlated with the dependent variable—time served, living in a dorm, and knowing a victim of prison sexual assault. Time served was positively correlated with IRSB and explained about 1% of the variance in IRSB. Living in a dorm compared to a cell resulted in lower levels of IRSB and accounted for approximately 1% of the variance in IRSB. Knowing a victim of prison sexual assault resulted in inmates displaying lower levels of IRSB and explained almost 2% of IRSB’s variance.

The full regression model included variables from the importation and deprivation regression models in Table 2. This model significantly predicted IRSB, F(18, 850) = 7.04, p < .001. The model R² = .13. In the full regression model, each of the variables exhibiting significance with IRSB in the importation regression model remained significant. Also, the proportion of variance explained by each of those variables remained relatively stable. When considering the importation regression model as the first block, then adding the variables from the deprivation regression model in the second block, the change in R² = .02 was significant, F(7, 850) = 2.67, p < .01. Of the significant variables in the deprivation regression model, only knowing a victim of prison sexual assault remained significant in the full regression model. The semi-partial r² associated with knowing a victim was also relatively unchanged. When considering the deprivation regression model as the first block, then adding the variables from the importation regression model in the second block, the change in R² = .10 was significant, F(11, 850) = 8.65, p < .001. Also, in the full regression model, two additional variables exhibited significance not present in the previous two models. Male inmates had significantly higher levels of IRSB and explained approximately 1% of the variance in the outcome. Knowing a victim of prison sexual assault within the past year had significantly higher levels of IRSB than those who did not. This variable also explained approximately 1% of the variance in IRSB.

The results of gender-specific regression models are depicted in Table 3. The men-only regression model is significant, F(17, 455) = 2.74, p < .001. Also, the model explained roughly 9% of the variance in IRSB among male inmates. Two variables significantly predicted male inmates’ IRSB. Inmates who identified as having never been married had higher levels of IRSB compared to those who had been married at some point, which explained almost 2% of the variance in the outcome. Also, having more than a high school diploma significantly decreased levels of IRSB compared to those who did not have a high school diploma. Having attained education beyond high school explained about 1% of the variance in the dependent variable (gender-specific importation and deprivation models available upon request).

Table 3.
Inmate Rape Supportive Beliefs Linear Regression Models by Gender.

Men Women

B (95% CI) β sr B β sr

Single 2.86 (1.11, 4.61) 0.16 .14 0.95 (–0.97, 2.87) 0.05 .05

Prior incarceration 0.08 (–1.59, 1.75) 0.01 .00 1.18 (–0.71, 3.08) 0.06 .06

Black^b 0.65 (–1.29, 2.59) 0.04 .03 0.74 (–1.30, 2.77) 0.04 .03

Hispanic^b 1.28 (–1.07, 3.62) 0.06 .05 4.90 (2.30, 7.51) 0.20 .17

Other^b 3.31 (–0.07, 6.68) 0.09 .09 1.19 (–2.02, 4.41) 0.04 .03

Age 0.07 (–0.04, 0.17) 0.08 .06 0.19 (0.10, 0.28) 0.21 .18

Age² 0.00 (–0.00, 0.01) 0.04 .04 0.01 (–0.00, 0.01) 0.09 .08

HS/GED^c –0.18 (–2.10, 1.75) –0.01 –.01 –3.08 (–5.08, –1.08) –0.16 –.14

More than HS^c –2.66* (–4.73, –0.60) –0.15 –.11 –5.51 (–7.62, –3.40) –0.29 –.24

Lifetime victim –1.27 (–3.95, 1.42) –0.06 –.04 0.16 (–1.77, 2.09) 0.01 .01

Sentence left^d –0.01 (–0.01, 0.00) –0.09 –.09 0.01 (–0.00, 0.01) 0.06 .06

Time served^e 0.01 (–0.11, 0.13) 0.01 .01 0.03 (–0.15, 0.21) 0.02 .02

Staff treatment 0.09 (–0.20, 0.38) 0.03 .03 0.08 (–0.22, 0.37) 0.03 .02

Dorm 0.08 (–1.55, 1.72) 0.01 .00 –1.05 (–3.10, 1.00) –0.05 –.05

In-prison victim 0.89 (–2.41, 4.19) 0.04 .02 0.52 (–1.76, 2.80) 0.02 .02

Know a victim –1.72 (–3.50, 0.07) –0.10 –.09 –4.38 (–6.77, –2.00) –0.24 –.17

Past year 2.04 (–0.44, 4.51) 0.08 .07 2.49** (–0.22, 5.20) 0.12 .08

Constant 0.68 0.11

R² .093 .207

Note. B = unstandardized regression coefficient; β = standardized regression coefficient; CI = confidence interval; sr = semi-partial correlation; HS = high school diploma; GED = general education development diploma.

^aFemale is the reference category. ^bWhite is the reference category. ^cLess than high school is the reference category. ^dMeasured in months. ^eMeasured in years.

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

The women-only regression model was also significant, F(17, 378) = 5.80, p < .001. The model R² = .21. Six variables in the women-only regression model were significantly associated with IRSB (Table 3). Women who identified as Hispanic had higher levels of IRSB than women who identified as White. This explained nearly 3% of the variance in IRSB. Older women inmates had higher levels of IRSB, which accounted for 3% of the variability in the outcome. Women who received a high school diploma and women who had more education than a high school diploma had significantly lower levels of IRSB than women who did not have a diploma. Having a high school diploma explained almost 2% of the variability in IRSB while having more education than a high school diploma accounted for 5%. Knowing a victim of prison sexual assault significantly decreased women’s IRSB and explained nearly 3% of the variance in the outcome. Knowing a victim of prison sexual assault within the past year significantly increased levels of IRSB among women and accounted for less than 1% of the variance in IRSB. Despite the pattern of results from the gender-specific regression models, coefficient comparison test results suggest that two variables impact men and women differently. Being a Hispanic inmate had a greater impact on IRSB for women (z = 2.28, p < .05), and having a high school diploma led to a greater impact in predicting women’s IRSB levels than it did for male inmates (z = 2.06, p < .05).

Discussion

Based on the findings of the present study, importation factors seem to matter more in determining inmates’ beliefs than deprivation factors. This suggests that inmates’ link to the outside world and the characteristics they bring into prison with them, while generally static in the present study, influence how they view prison sexual violence. These data suggest inmates import these types of patriarchal beliefs and then impose them on the prison environment. The results are less supportive of the idea that time spent in prison and the adaptations to their deprivations affect how they think about rape and view assault victims, except for when inmates know victims of prison sexual violence.

The relationships between IRSB and sex, age, ethnicity (Hispanic), marital status, and education mirror general patterns of relationships found among heterosexual based beliefs among free community samples (Anderson et al., 1997; Lonsway & Fitzgerald, 1994; Suarez & Gadalla, 2010), which suggests that rape supportive beliefs may be a robust concept across varying cultural contexts. One prison-related issue to further consider is the temporal difference between knowing a victim and knowing someone who was recently victimized. Over time, knowing someone who was victimized may lead to less rape supportive beliefs, but it would seem that those inmates who knew someone who was recently victimized led to greater adherence to rape supportive beliefs. This could be related to a shift in empathy after observing a victim over time or possibly a cognitively jarring effect of hearing such information about someone they know that leads to the person relying on prison cultural beliefs. Future studies should investigate whether there are different effects over time, related to those whom victims disclose news of their assault.

Distinct variables predicted men and women’s acceptance of IRSB. Only two variables impacted men’s IRSB levels. Being single increased IRSB and having an education greater than a high school degree decreased IRSB, which also impacted women. Among women, identifying as Hispanic, age, higher levels of educational achievement, knowing a victim ever, and knowing a victim within the past year significantly predicted IRSB. Except for being single, the women’s model closely resembles the full model for all inmates. Also, the impact of certain variables differed in only two instances. Identifying as Hispanic impacted women’s acceptance of IRSB compared to men. Additionally, having a high school diploma impacted women’s rejection of IRSB more so than men. Further investigating the link between being single and greater acceptance of IRSB among men could lead to the identification of important factors that can be targeted to reduce IRSB among male inmates. Additional investigation among inmates that identify as Hispanic appears important, too. Examining attitudinal measures toward women, sexism, traditional sex roles, and the acceptance of aggression may prove useful in beginning to disentangle these relationships with IRSB (Suarez & Gadalla, 2010).

Future studies could also address the effects of prison programming on IRSB, especially given that education level was a significant predictor in this study. A difference may exist between inmates who advance their education inside of prison versus those that enter prison with degrees in hand. Also, prison programs aimed at reducing IRSB should be developed and evaluated. Peer education programs are becoming more common and have been found to be effective in decreasing rape myth acceptance (Foubert & Marriott, 1997). Multilevel samples that rely on large numbers of prisons and inmates would further aid in identifying prison-level characteristics related to the deprivation model—crowding, programs, and resources available, or security level—that may influence IRSB.

Practical Implications

One practical implication is that, since these beliefs appear to be imported, prison administrators could implement prison sexual violence awareness and education programs. If new inmates are screened for predictors of rape supportive beliefs upon entry into prison, to include not only inmate demographic characteristics but also rape supportive attitudes found in research with noninmates (Murphy & Hine, 2019), correctional administrators could aim programs and other efforts to reduce rape supportive beliefs at these inmates making such efforts more efficient and beneficial. By changing these beliefs, one consequence could be the discouragement of sexually exploitive, coercive, and assaultive behaviors and the increased recognition that these behaviors are harmful. Another implication could build off of current prison programs offering opportunities for further education as higher education led to lower IRSB was consistent in all models. Additionally, and perhaps related to the finding regarding marital status, programs focusing on increasing empathy among inmates, particularly those who are not married may be useful. Finally, continued efforts to change people’s beliefs outside of prison are just as important as any prison program since these results support the idea that IRSB are imported into prison.

Limitations

Some limitations may affect the applicability of the results. For one, these results cannot be seen as causal as these data were collected as part of a cross-sectional study. Sampling issues may affect the generalizability of the results. Prison officials were contracted to provide a random sample for the male inmates; however, as this task was outside the researchers’ control, it cannot be entirely verified. Also, sampled male inmates were chosen from the general population; whereas all women inmates from administrative segregation and general population were allowed to participate. It is possible that male inmates in administrative segregation are either more or less likely to report sexual victimization. Finally, this sample focused on the prison units with the highest reported levels of prison sexual assault at the time of data collection. Therefore a sample size was not calculated at the outset of this study. However, with over 900 initial inmate respondents, this study represents one of the larger existing studies on prison sexual violence and IRSB.

Further, past research has indicated that one reason inmates do not disclose sexual victimization is to prevent themselves from potentially being transferred to protective custody (Fleisher & Krienert, 2009; Owen et al., 2017). Regarding the subject at hand, the sensitive nature of some of the measures and pressure to conform to prison normative attitudes may have resulted in inaccurate self-report responses. Past research suggests that inmates may be less likely to report victimization due to the nature of prison culture and the potential for negative consequences (Garland & Wilson, 2013). Also, measuring past incarceration was limited to the same state prison system as opposed to asking a more general question that would allow for imprisonments in other places to be captured.

Another limitation is that the IRSB scale has not been replicated. The IRSB scale relied on previously published survey items and their respective 5- and 7-point Likert scales (Burt, 1980; Deitz et al., 1982). As such, because each item was standardized prior to summing and not all scale items have the same response options there could be limitations generalizing the IRSB scale across study samples. Despite rigorous development procedures and acceptable internal consistency, there were no measures of criterion-based validity to support the IRSB scale in this study. However, Fowler et al. (2010) found that the IRSB scale positively correlated with vignette-based definitions of prison sexual assault such that higher levels of IRSB led to decreased chances of labeling a sexual assault as such. The IRSB measure did relate to other variables in the same manner as variables in free community research; however, further study is warranted to establish criterion-based and predictive validity in a similar manner to what others have found. Additionally, theoretical measures used to predict rape supportive beliefs outside of prison were not available for inmates prior to entering prison. Rather, this study relied on known demographic characteristics that are associated with rape supportive beliefs outside of prison to serve as proxy measures to assess if importation may serve as an explanation of IRSB. As a research implication, future studies could include variables measuring rape supportive beliefs in a longitudinal design. For instance, attitudinal “code of the street” variables used to predict violent behavior outside of prison have been demonstrated to predict violent prison misconduct among inmates through a prospective longitudinal design (Mears et al., 2013). Incorporating theoretically-derived variables collected outside of prison significantly predicted prison outcomes even with other proxy measures for importation were present (Mears et al., 2013).

Conclusion

While these results do not address IRSB-related behaviors, the assumption is that as long as the beliefs are present and pervasive, prison rape will continue and victims—because of lack of acknowledgment, self-blame, or embarrassment present in rape culture—may not seek the services needed to recover from this event. To address prison sexual violence, strategies that stress the importance of multiple causes may be required. Increasing supervision and decreasing opportunities to offend could be just as fruitful as efforts to change inmates’ cultural beliefs.

Footnotes

Appendix

Inmate Rape Supportive Beliefs Scale Items with Scale Points in Parentheses

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

ORCID iDs

Shannon K. Fowler

Wesley T. Smith

Author Biographies

Shannon K. Fowler, PhD, is an associate professor in the Department of Criminal Justice and Social work at the University of Houston-Downtown. His research interests include intimate partner violence, prison sexual violence, victimology, and the community context of victimization. His work has appeared in Criminal Justice Review, Journal of Criminal Justice, and Journal of Offender Rehabilitation.

Ashley G. Blackburn, PhD, is chair of the Department of Criminal Justice and Social Work and Professor of Criminal Justice at the University of Houston-Downtown. Her areas of interest include institutional corrections and victimization. She has published in journals such as Prison Journal and Women & Criminal Justice and has co-authored texts on research methods and victims’ rights.

Wesley T. Smith, MS, is a doctoral student in the College of Criminal Justice at Sam Houston State University. He is a 12-year veteran of the United States Armed Forces with a background in forensic collection, investigation, and information analysis. His research interests include correctional staff management, programming, and specialty populations, as well as veterans in the criminal justice system.

Janet L. Mullings, PhD, is associate dean and professor of Criminal Justice and Criminology at Sam Houston State University. Her research topics include women in prison, sexual assault, domestic violence, and child maltreatment. Her recent research has appeared in Journal of Child and Adolescent Trauma, Deviant Behavior, and The Prison Journal. She teaches victimology, victim studies, and family violence.

References

Aiken

L. S.

, & West

S. G.

(1991). Multiple regression: Testing and interpreting interactions . Sage Publications.

Anderson

K. B.

, Cooper

, & Okamura

(1997). Individual differences and attitudes toward rape: A meta-analytic review. Personality and Social Psychology Bulletin , 23(3), 295–315.

Basile

K. C.

, & Saltzman

L. E.

(2002). Sexual violence surveillance: Uniform definitions and recommended data elements: Version 1.0 . National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. https://stacks.cdc.gov/view/cdc/6545/cdc_6545_DS1.pdf

Beck

A. J.

, Berzofsky

, Caspar

, & Krebs

(2013). Sexual victimization in prisons and jails reported by inmates, 2011–12 (NCJ 241399). U.S. Department of Justice Office of Justice Programs. https://www.bjs.gov/content/pub/pdf/svpjri1112.pdf

Beck

A. J.

, Harrison

P. M.

, Berzofsky

, Caspar

, & Krebs

(2010). Sexual victimization in prisons and jails reported by inmates, 2008-09 (NCJ 231169). U.S. Department of Justice Office of Justice Programs. https://www.bjs.gov/content/pub/pdf/svpjri0809.pdf

Boakye

K. E.

(2009). Attitudes toward rape and victims of rape: A test of feminist theory in Ghana. Journal of Interpersonal Violence , 24(10), 1633–1651.

Bronson

, & Carson

E. A.

(2019). Prisoners in 2017 (NCJ 252156 U.S. Department of Justice Office of Justice Programs. https://www.bjs.gov/content/pub/pdf/p17.pdf

Brownmiller

(1975). Against our will: Men, women and rape . Fawcett Columbine.

Burgess

G. H.

(2007). Assessment of rape-supportive attitudes and beliefs in college men: Development, reliability, and validity of the rape attitudes and beliefs scale. Journal of Interpersonal Violence , 22(8), 973–993.

10.

Burt

M. R.

(1980). Cultural myths and supports for rape. Journal of Personality and Social Psychology , 38(2), 217–230.

11.

Cannan

S. N.

, Jozkowski

K. N.

, & Crawford

B. L.

(2018). Sexual assault supportive attitudes: Rape myth acceptance and token resistance in Greek and non-Greek college students from two university samples in the United States. Journal of Interpersonal Violence , 33(22), 3502–3530.

12.

Debowska

, Boduszek

, Dhingra

, & DeLisi

(2016). The effect of male incarceration on rape myth acceptance: Application of propensity score matching technique. Deviant Behavior , 37(6), 634–643.

13.

Debowska

, Boduszek

, Dhingra

, Kola

, & Meller-Prunska

(2015). The role of psychopathy and exposure to violence in rape myth acceptance. Journal of Interpersonal Violence , 30(15), 2751–2770.

14.

Deitz

S. R.

, Blackwell

K. T.

, Daley

P. C.

, & Bentley

B. J.

(1982). Measurement of empathy toward rape victims and rapists. Journal of Personality and Social Psychology , 43(2), 372–384.

15.

Deming

M. E.

, Covan

E. K.

, Swan

S. C.

, & Billings

D. L.

(2013). Exploring rape myths, gendered norms, group processing, and the social context of rape among college women: A qualitative analysis. Violence Against Women , 19(4), 465–485.

16.

Dhami

M. K.

, Ayton

, & Loewenstein

(2007). Adaptation to imprisonment: Indigenous or imported? Criminal Justice and Behavior , 34(8), 1085–1100.

17.

Farrington

(1992). The modern prison as total institution? Public perception versus objective reality. Crime & Delinquency , 38(1), 6–26.

18.

Fleisher

M. S.

, & Krienert

J. L.

(2009). The myth of prison rape: Sexual culture in American prisons . Rowman & Littlefield Publishers.

19.

Foubert

J. D.

, & Marriott

K. A.

(1997). Effects of a sexual assault peer education program on men’s belief in rape myths. Sex Roles , 36(3/4), 259–268.

20.

Fowler

S. K.

, Blackburn

A. G.

, Marquart

J. W.

, & Mullings

J. L.

(2010). Inmates’ cultural beliefs about sexual violence and their relationship to definitions of sexual assault. Journal of Offender Rehabilitation , 49(3), 180–199.

21.

Fox

(1991). Regression diagnostics . SAGE Publications.

22.

Garland

, & Wilson

(2013). Prison inmates’ views of whether reporting rape is the same as snitching: An exploratory study and research agenda. Journal of Interpersonal Violence , 28(6), 1201–1222.

23.

Goffman

(1961). Asylums . Anchor Books.

24.

Goncalves

L. C.

, Goncalves

R. A.

, Martins

, & Dirkzwager

A. J. E.

(2014). Predicting infractions and health care utilization in prison: A meta-analysis. Criminal Justice and Behavior , 41(8), 921–942.

25.

Gover

A. R.

, Pérez

D. M.

, & Jennings

W. G.

(2008). Gender differences in factors contributing to institutional misconduct. The Prison Journal , 88(3), 378–403.

26.

Grubb

, & Turner

(2012). Attribution of blame in rape cases: A review of the impact of rape myth acceptance, gender role conformity and substance use on victim blaming. Aggression and Violent Behavior , 17(5), 443–452.

27.

Haywood

, & Swank

(2008). Rape myths among Appalachian college students. Violence and Victims , 23(3), 373–389.

28.

Heath

N. M.

, Lynch

S. M.

, Fritch

A. M.

, McArthur

L. N.

, & Smith

S. L.

(2011). Silent survivors: Rape myth acceptance in incarcerated women’s narratives of disclosure and reporting of rape. Psychology of Women Quarterly , 35(4), 596–610.

29.

Heath

N. M.

, Lynch

S. M.

, Fritch

A. M.

, & Wong

M. M.

(2013). Rape myth acceptance impacts the reporting of rape to the police: A study of incarcerated women. Violence Against Women , 19(9), 1065–1078.

30.

Hensley

(2000). Attitudes toward homosexuality in a male and female prison: An exploration study. The Prison Journal , 80(4), 434–441.

31.

Hensley

, Tewksbury

, & Castle

(2003). Characteristics of prison sexual assault targets in male Oklahoma correctional facilities. Journal of Interpersonal Violence , 18(6), 595–606.

32.

Irwin

, & Cressey

D. R.

(1962). Thieves, convicts and the inmate culture. Social Problems , 10(2), 142–155.

33.

Jenness

, Sexton

, & Sumner

(2019). Sexual victimization against transgender women in prison: Consent and coercion in context. Criminology , 57(4), 603–631.

34.

Johnson

N. L.

, & Johnson

D. M.

(2021). An empirical exploration into the measurement of rape culture. Journal of Interpersonal Violence , 36(1–2), NP70–NP95. https://doi.org/10.1177/0886260517732347

35.

Karlsson

M. E.

, & Zielinski

M. J.

(2020). Sexual victimization and mental illness prevalence rates among incarcerated women: A literature review. Trauma, Violence, & Abuse , 21(2), 326–349. https://doi.org/10.1177/1524838018767933

36.

Kassing

L. R.

, Besley

, & Frey

L. L.

(2005). Gender role conflict, homophobia, age, and education as predictors of male rape myth acceptance. Journal of Mental Health Counseling , 27(4), 311–328.

37.

Keith

(2017). Masculinities in contemporary American culture: An intersectional approach . Taylor & Francis.

38.

Lahm

K. F.

(2008). Inmate-on-inmate assault: A multilevel examination of prison violence. Criminal Justice and Behavior , 35, 120–37.

39.

Lonsway

K. A.

, & Fitzgerald

L. F.

(1994). Rape myths: In review. Psychology of Women Quarterly , 18(2), 133–164.

40.

Lutze

F. E.

(2003). Ultramasculine stereotypes and violence. In Zaitzow

B. H.

& Thomas

(Eds.), Women in prison: Gender and social control (pp. 183–203). Lynne Reinner.

41.

McCorkle

R. C.

, Miethe

T. D.

, & Drass

K. A.

(1995). The roots of prison violence: A test of the deprivation, management, and “not-so-total” institution models. Crime & Delinquency , 41(3), 317–331.

42.

Mears

D. P.

, Stewart

E. A.

, Siennick

S. E.

, & Simons

R. L.

(2013). The code of the street and inmate violence: Investigating the salience of imported belief systems. Criminology , 51(3), 695–728.

43.

Morse

S. J.

, & Wright

K. A.

(2022). Imprisoned men: Masculinity variability and implications for correctional programming. Corrections , 7(1), 23–45. https://doi.org/10.1080/23774657.2019.1694854

44.

Murnen

S. K.

, & Kohlman

M. H.

(2007). Athletic participation, fraternity membership, and sexual aggression among college men: A meta-analytic review. Sex Roles , 57(1–2), 145–157.

45.

Murphy

, & Hine

(2019). Investigating the demographic and attitudinal predictors of rape myth acceptance in U.K. Police officers: Developing an evidence-base for training and professional development. Psychology, Crime & Law , 25(1), 69–89. https://doi.org/10.1080/1068316X.2018.1503663

46.

Owen

, Wells

, & Pollock

J. M.

(2017). In search of safety: Confronting inequality in women’s imprisonment . University of California Press.

47.

Rantala

R. R.

(2018). Sexual victimization reported by adult correctional authorities, 2012-15 (NCJ 251146). U.S. Department of Justice Office of Justice Programs. https://www.bjs.gov/content/pub/pdf/svraca1215.pdf

48.

Sisco

M. M.

, & Becker

J. V.

(2007). Beyond predicting the risk of sexual victimization in prison—Considering inmate options and reporting avenues for addressing an inherent problem. Criminology & Public Policy , 6(3), 573–584.

49.

Smith

S. G.

, Zhang

, Basile

K. C.

, Merrick

M. T.

, Wang

, Kresnow

M. J.

, & Chen

(2018). The national intimate partner and sexual violence survey: 2015 data brief–updated release . Center for Disease Control and Prevention. https://www.cdc.gov/violenceprevention/pdf/2015data-brief508.pdf

50.

Suarez

, & Gadalla

T. M.

(2010). Stop blaming the victim: A meta-analysis on rape myths. Journal of Interpersonal Violence , 25(11), 2010–2035. https://doi.org/10.1177/0886260509354503

51.

Sykes

G. M.

(1958). The society of captives: The study of a maximum security prison . Princeton University Press.

52.

Vonderhaar

R. L.

, & Carmody

D. C.

(2015). There are no “innocent victims”: The influence of just world beliefs and prior victimization on rape myth acceptance. Journal of Interpersonal Violence , 30(10), 1615–1632.

53.

Wolff

, Blitz

C. L.

, & Shi

(2007). Rates of sexual victimization in prison for inmates with and without mental disorders. Psychiatric Services , 58(8), 1087–1094.

54.

Wolff

, Blitz

C. L.

, Shi

, Bachman

, & Siegel

J. A.

(2006). Sexual violence inside prisons: Rates of victimization. Journal of Urban Health , 83(5), 835–848.

55.

Xue

, Fang

, Huang

, Cui

, Rhodes

K. V.

, & Gelles

(2019). Rape myths and the cross-cultural adaptation of the Illinois Rape Myth Acceptance Scale in China. Journal of Interpersonal Violence , 34(7), 1428–1460.

	N	%	M	SD	Range
IRSB			0.0	8.82	–19.20–37.02
Gender (male)	475	54.3
Single	352	40.2
Prior incarceration	382	43.7
Race
White	327	37.4
Black	346	39.5
Hispanic	142	16.2
Other	60	6.9
Age			40.34	10.63	19–78
Education
Less than HS/GED	271	31.3	.31
HS/GED	306	35.2	.35
More than HS	292	33.6	.33
Sentence length^a			259.12	167.17	3–480
Sentence left^a			165.82	149.13	0–480
Time served^b			8.02	6.83	0–38
Staff treatment			6.00	2.78	3–15
Dorm	537	61.4
Lifetime victim	388	44.3
In-prison victim	133	15.2
Know a victim	345	39.4
Past year	149	17.0

	Importation			Deprivation			Full
	B (95% CI)	β	sr	B ( 95% CI)	β	sr	B (95% CI)	β	sr
Gender^a	1.20 (–0.13, 2.53)	0.07	.06	–	–	–	1.71* (0.19, 3.23)	0.10	.07
Single	2.17** (0.89, 3.46)	0.12	.11	–	–	–	2.25** (0.97, 3.54)	0.13	.11
Prior incarceration	0.34 (–0.90, 1.58)	0.02	.02	–	–	–	0.36 (–0.88, 1.60)	0.02	.02
Black^b	1.03 (–0.35, 2.41)	0.06	.05	–	–	–	0.71 (–0.69, 2.11)	0.04	.03
Hispanic^b	2.88** (1.14, 4.62)	0.12	.11	–	–	–	2.82** (1.08, 4.55)	0.12	.10
Other^b	2.33 (–0.02, 4.68)	0.07	.06	–	–	–	2.26 (–0.07, 4.60)	0.06	.06
Age	0.12** (0.06, 0.18)	0.15	.12	–	–	–	0.13** (0.06, 0.20)	0.16	.12
Age²	0.00 (–0.00, 0.01)	0.04	.04	–	–	–	0.00 (–0.00, 0.01)	0.03	.03
HS/GED^c	–1.72* (–3.11, –0.34)	–0.09	–.08	–	–	–	–1.57* (–2.96, –0.19)	–0.09	–.07
More than HS^c	–4.43** (–5.90, –2.97)	–0.24	–.19	–	–	–	–4.11** (–5.57, –2.64)	–0.22	–.18
Lifetime victim	–0.70 (–2.08, 0.61)	–0.04	–.03	–	–	–	–0.72 (–2.22, 0.79)	–0.04	–.03
Sentence left^d	–	–	–	–0.00 (–0.01, 0.00)	–0.03	–.03	–0.00 (–0.01, 0.00)	–0.04	–.04
Time served^e	–	–	–	0.11* (0.20, 0.20)	0.09	.08	0.00 (–0.09, 0.10)	0.00	.00
Staff treatment	–	–	–	0.13 (–0.08, 0.34)	0.04	.04	0.09 (–0.12, 0.29)	0.03	.03
Dorm	–	–	–	–1.32* (–2.52, –0.12)	–0.07	–.07	–0.34 (–1.59, 0.91)	–0.02	–.02
In-prison victim	–	–	–	–0.28 (–1.94, 1.38)	–0.01	–.01	0.73 (–1.10, 2.55)	0.03	.03
Know a victim	–	–	–	–3.02** (–4.47, –1.56)	–0.17	–.14	–2.77** (–4.19, –1.36)	–0.15	–.12
Past year	–	–	–	1.49 (–0.35, 3.32)	0.06	.05	2.01* (0.24, 3.79)	0.09	.07
Constant	–0.53	–	–	0.44	–	–	–0.11	–	–
R²	.111			.034			.130

	Men			Women
	B (95% CI)	β	sr	B	β	sr
Single	2.86** (1.11, 4.61)	0.16	.14	0.95 (–0.97, 2.87)	0.05	.05
Prior incarceration	0.08 (–1.59, 1.75)	0.01	.00	1.18 (–0.71, 3.08)	0.06	.06
Black^b	0.65 (–1.29, 2.59)	0.04	.03	0.74 (–1.30, 2.77)	0.04	.03
Hispanic^b	1.28 (–1.07, 3.62)	0.06	.05	4.90** (2.30, 7.51)	0.20	.17
Other^b	3.31 (–0.07, 6.68)	0.09	.09	1.19 (–2.02, 4.41)	0.04	.03
Age	0.07 (–0.04, 0.17)	0.08	.06	0.19** (0.10, 0.28)	0.21	.18
Age²	0.00 (–0.00, 0.01)	0.04	.04	0.01 (–0.00, 0.01)	0.09	.08
HS/GED^c	–0.18 (–2.10, 1.75)	–0.01	–.01	–3.08* (–5.08, –1.08)	–0.16	–.14
More than HS^c	–2.66* (–4.73, –0.60)	–0.15	–.11	–5.51** (–7.62, –3.40)	–0.29	–.24
Lifetime victim	–1.27 (–3.95, 1.42)	–0.06	–.04	0.16 (–1.77, 2.09)	0.01	.01
Sentence left^d	–0.01 (–0.01, 0.00)	–0.09	–.09	0.01 (–0.00, 0.01)	0.06	.06
Time served^e	0.01 (–0.11, 0.13)	0.01	.01	0.03 (–0.15, 0.21)	0.02	.02
Staff treatment	0.09 (–0.20, 0.38)	0.03	.03	0.08 (–0.22, 0.37)	0.03	.02
Dorm	0.08 (–1.55, 1.72)	0.01	.00	–1.05 (–3.10, 1.00)	–0.05	–.05
In-prison victim	0.89 (–2.41, 4.19)	0.04	.02	0.52 (–1.76, 2.80)	0.02	.02
Know a victim	–1.72 (–3.50, 0.07)	–0.10	–.09	–4.38** (–6.77, –2.00)	–0.24	–.17
Past year	2.04 (–0.44, 4.51)	0.08	.07	2.49** (–0.22, 5.20)	0.12	.08
Constant	0.68			0.11
R²	.093			.207