Skip to main content
SearchLoginLogin or Signup

The Dark Figure of Prison Violence in Uruguay: An Exploratory Mixed-Method Study

Trajtenberg, N. & Sánchez de Ribera, O. (in press). The Dark Figure of Prison Violence in Uruguay: An Exploratory Mixed-Method Study. In L. Huey & D. Buil-Gil (Eds.), The Crime Data Handbook. Bristol: Policy Press.

Published onFeb 07, 2024
The Dark Figure of Prison Violence in Uruguay: An Exploratory Mixed-Method Study


Prisons are concerned with safety. One key piece of information is how much violence takes place in prison. This is important from a micro perspective – will this new inmate be at risk of victimization? And from a macro one too – are more resources needed in a prison to reduce the potential for violence? Yet our knowledge about the ‘dark figure’ of prison violence is limited by the nature of the act and our ability to research it. We use a mixed methods project in the Penitentiary Unit 4 (Uruguay) that combines: official records of prison incidents; survey data in a convenience sample of inmates (n=209); qualitative interviews with inmates, prison staff and key informants (n=33); and non-participant unstructured and structured observation. Our findings show that in two months only 12 violent incidents were officially recorded, compared to 82 incidents recorded in our observations. Of the total inmates that answered our survey (42% of the module), 60% were victimized but only 22% of victims reported to authorities. Reasons for not reporting were not wanting to break prison codes, fear of retaliation, lack of trust in the prison system and the perception of lack of punishment. Under-reporting was also associated with institutional conditions related to lack of resources, irrelevance of reporting in guard role, lack of trust in the importance of reporting, poor knowledge of how to report, and ‘naturalization’ of violence. We finish discussing limitations of our study, and making suggestions for estimation of dark figure and policy implications.

Corresponding author: Nicolás Trajtenberg. School of Social Science, School of Social Science, University of Manchester, Oxford Rd, Manchester, M13 9PL, UK. E-mail: [email protected]


Interpersonal violence among prisoners involves a broad range of acts such as physical attacks and injuries, homicides, sexual abuse, different types of psychological aggression, threats, humiliation, bullying, financial harms, misconduct or self-harm (Ireland, 2002; Gadon et al, 2006). Prison violence is an extended phenomenon worldwide involving a serious abuse of inmates’ human rights (Wolff et al, 2007). Inmates’ experience of violent victimization is significantly higher than in the community (Wolff et al, 2007; Turanovic and Pratt, 2019). Additionally, prison violence can have psychological and health costs for inmates, and can reinforce violence and recidivism after release (Steiner, Butler and Ellison, 2014; Auty and Liebling, 2020). Given that one of the aims of prisons is to provide a secure environment for inmates and staff (Dagget and Campt, 2009; Modvig, 2014), the measurement of safety is an important issue.

Measuring and estimating prison violence with official statistics faces a well-known challenge in crime statistics: the non-reporting and non-detection of violent incidents to/by authorities, or the ‘dark figure’ problem (Penney, 2014). Estimations of prison incidents from data generated by the records of  correctional institutions do not randomly sample violent incidents that take place in prison, and thus their representativeness are questioned in criminology (Bottoms, 1999; Dagget and Campt, 2009). Scarce available comparisons of official and unofficial data reveal that prison violence is significantly underestimated: only 10 per cent of violent incidents will be recorded in prison statistics (Kury and Smartt, 2002; Byrne and Hummer, 2007).

The gap between actual victimization and reported/recorded violent events in prisons is due to a number of factors. First, the thread of retaliation by the perpetrator or other inmates for violating ‘subcultural prison norms’ (Levan Miller, 2010; Modvig, 2014). Second, reporting a victimization might involve being sent to protective custody and losing benefits (for example, educational or work programmes, leisure activities) (Levan Miller, 2010). Third, victimized inmates might not trust that the system will punish the perpetrator (Gear, 2007). A final reason for underestimation might be the lack of standardized and reliable measures in penitentiary institutions, which are associated with staff discretion (Abbiati et al, 2019; Trajtenberg and Sánchez de Ribera, 2020).

The use of self-report surveys in prison can help to overcome some limitations of official statistics. Most of this type of research has been conducted in American prisons and has confirmed the existence of high levels of prevalence of physical violence, with estimates ranging from 7 per cent to 25 per cent depending on the study (Wolff et al, 2007; Wooldredge and Steiner, 2014; Steiner and Cain, 2016; Teasdale et al, 2016), with some comparable estimates from Europe (Hagemann, 2008). Closer geographical comparators in Latin American countries reveal substantial variability in the reports of physical victimization in prison: from 25 per cent in Chile to 3.5 per cent in El Salvador (Sánchez and Piñol, 2015[A1] [A2] ). A recent meta-analysis which combined 36 different samples estimated that physical victimization reported in prison was 18 per cent and sexual victimization reached almost 10 per cent (Caravaca-Sánchez et al, 2022).

Yet, self-reported measures have several limitations. Respondents may not recall many events they suffered and when events took place, particularly if they are less serious and distant in time (Connel and Farrington, 1996; Braga et al, 2019). Furthermore, there are no standardized measures and changes in the operationalization of violence; specificity of questions referring to behaviours, as well as the temporal frame, can change the rates of victimization (Wolff and Bachman, 2008). Reporting can also be affected by the combination of high levels of non-response and convenience sampling procedures frequently applied in surveys in prisons. Another source of bias is social desirability when reporting behaviours that are serious or shameful in the context of subcultural codes (Connel and Farrington, 1996). Machismo and hyper masculinity are related to higher reporting of offending (Gudjonsson et al, 2011) and could be related to higher reporting of victimization to signal toughness and power to ‘deal with violence’.

Prison violence research studies rarely combine official and self-report survey data (Logan, 1992). To the best of our knowledge only a Portuguese study by Braga and colleagues (2019) compared these two types of data to estimate the dark figure during a 12-month period using a sample of 219 (response rate of 49 per cent). While official records reported that 3.2 per cent of the inmates were victimized, self-report surveys revealed a prevalence rate of 78.5 per cent (Braga et al, 2019). The aforementioned problems show that if we want to know in a less biased way how much violence happens in prison and why this bias is taking place, we need to combine official records and self-report surveys with other quantitative and qualitative data.

In this chapter we advance in the exploration of the dark figure by using a mixed-methods design with two aims. First, we combine official records and structured observation to provide a less biased estimation of unreported violent incidents. As a second step, we include survey data, qualitative unstructured observation and qualitative interviews to identify reasons affecting reporting/recording of incidents.


Design and participants

Data used is part of a larger project, ‘Violencia Penitenciaria: el Caso Uruguayo’ (‘Penitentiary Violence: The Uruguayan Case’). The goal was to improve the measurement, diagnosis and understanding of correlates of prison violence. This project involved a mixed-methods strategy which combined different sources of information. As a first step we analysed official records of prison reports of incidents in the Penitentiary Management System for the last three years, and particularly during June and July of 2019 in Santiago Vazquez Prison. This data informed our selection of Module A for the next steps because it yielded the highest rate of incidents between 2016 and 2019.[i] The second step involved non-participant unstructured and structured observation of Module A, which allowed us to generate an alternative estimation of incidents during June and July and to develop trust and facilitate contacts with inmates/guards to carry out interviews. As a third step we conducted semi-structured interviews with inmates (n=21) and prison staff working in the module (n=8), and key informants from the prison service (n=4). Finally, we administered a self-report questionnaire survey to a convenience sample of male inmates in the module during August (n=209; 42 per cent response rate) which involved information related to their criminal antecedents, health problems, use of drugs, moral values, self-control, involvement in violent incidents in prison and so on. Four measures are relevant for this chapter (see later).

Research setting

Participants were housed in Module A of Penitentiary Unit 4 (Figure 1), one of the largest high-security prisons in Uruguay, which held a total of 3,124 inmates in 2019 (28 per cent of the prison population of Uruguay) and concentrates almost 20 per cent of all violent incidents of the prison system (Trajtenberg and Sanchez de Ribera, 2020). Module A is overcrowded, as it was designed for 200 inmates and houses 498. Organizationally, this means a staff—prisoner ratio of 50:1 rather than 5:1.

Figure 1: Picture of Penitentiary Institution No 4


The survey questionnaire included a screening measure of eight items of bullying and violent victimization adapted from the Direct and Indirect Prisoner Checklist (DIPC-SR) (Ireland, 2002). Inmates had to answer if they had robbed (or been robbed) in the last two months, the options being never, almost never, sometimes, frequently, always. Similar items were used in relation to: insult or verbally abuse another inmate; physical aggression of another inmate without a weapon; and physical aggression with a weapon. Additionally, we included one item asking them if any victimization took place, whether prison officers aware of this and whether it was reported to authorities (options: never, almost never, sometimes, frequently, always). Finally, we included five items on violent subculture and prison codes related to justifications used for assault on inmates that started a fight, the use of violence to defend honour and so on (Reisig and Mezko, 2009) (options: totally disagree, disagree, agree, totally agree).

The unstructured observation lacked any predetermined observation guide. For the structured observation we developed a checklist of violent and non-violent behaviours which included a narrative description of the incident and a list of behaviours including: not following rules but without violence (for example, selling or buying forbidden items, self-harm or threats to carry it out); thefts/robberies (for example, stealing goods from others); indirect aggressions that do not involve other persons as object of the aggression (for example, violently kick doors/walls, destroy prison materials or property, derogatory comments toward others); direct verbal aggression which has an identifiable object of aggression without physical aggression (for example, intimidation or threats to others, forcing another inmate to lie, verbally abusing inmates’ family); sexual aggression (for example, rape, sexual abuse); physical aggression or threat of physical aggression (for example, hit others, hurt others with or without gun, threaten others using violence); and riots or escapes.

Semi-structured interviews with inmates and staff followed a guide that included questions about relations with inmates and prison guards, problems and pains suffered during incarceration, personal experiences of violence and with reporting, frequency of violent incidents in the module, reasons for violence taking place and guards’ reactions to incidents.


Unstructured observation was conducted during three months (April to June) by one of the authors and a research assistant and involved going three days per week from 9 am to 2 pm or from 13 am to 18 pm. The survey questionnaire was administered to 42 per cent of the module population using a convenience sample between August and September 2017. The questionnaire took 20–25 minutes and was administered in the visits room of the module. Inmates were either recruited for the qualitative interviews based on our observations or were survey participants. Interviews lasted 45–60 minutes and were conducted by the authors and a research assistant in the visits room or an office within the module. Structured observation of incidents in the module was conducted from Monday to Friday between 8 am and 6 pm during June and July. It was conducted by a prison educator from the module trained to observe and register incidents using a structured observation checklist.

Data analysis

We used descriptive analysis to compare incidents in the module and official data, as well as to compare reporting among different groups from the survey data (victim, victim-perpetrator, type of violence suffered, support for subcultural honour values). We used thematic analysis to analyse qualitative data from observation and interviews. All analyses were performed using R (R Core Team, 2013) and the RDQA package for qualitative data (Huang, 2012).


Comparison between official records and observation

Only 15 per cent of incidents of the module in June and July 2019 ended up being reported officially (approximately 1 in 7). Our observations yielded 82 incidents broken down as follows: 35% physical aggressions (28 incidents), 23% direct non-physical aggressions (19 incidents); 20% breach of rules (16 incidents); 12% thefts (10 incidents); 9% indirect aggressions (7 incidents); and 1% riots or escapes (Table 1). In contrast, official records showed that only 12 events were recorded in the Prison Management Information system (two events refer to non-compliance with rules; one robbery; two events of indirect aggression; four events of direct non-physical aggression; two events of physical aggression; and one riot). Furthermore, the difference in under-reporting is not homogeneous across categories of incidents. While one of the most serious incidents (riot) showed no differences, indirect aggressions and direct physical aggressions showed lower levels of under-reporting (21 per cent and 29 per cent respectively), breach of rules and thefts (13 per cent  and 10 per cent respectively) and, finally, only 7 per cent of physical aggressions ended up being reported to authorities.

Table 1: Comparison of incidents in the module: observation versus official records in June and July 2019



Official records

Rate reported

Breach of rules








Indirect aggression




Direct not physical aggression




Physical aggression




Riots and escapes








The self-report victimization survey showed a high rate of not reporting in the last two months. Of a total of 209 inmates who completed the survey (42 per cent of the total population of the module), almost 60 per cent had been victimized at least once, but only 22 per cent of those victimized reported their victimization (Table 2). Likewise, almost half of the inmates that answered our survey had suffered at least one physical aggression, but again only one quarter of them reported it. Interestingly, the poly-victim group, that is, those that suffered multiple types of aggressions (physical, sexual and symbolic violence, robberies) showed the highest reporting rates (47 per cent).

It is worth noticing that there is a victimization–perpetration overlap and it is related to inmates’ reporting. One third declared to be neither a victim nor a perpetrator in the last two months (n=69), 18 per cent committed at least one type of aggression (n=39), 8 per cent were only victims (n=17) and 40 per cent (n=84) were both victims and perpetrators. Victim-perpetrators were the group that reported less to authorities (14 per cent) in relation to other types of victims.

Table 2: Victimization and reporting to authorities (two months)














Victims of physical aggressions





Poly victims





Only victims










Reasons for under-reporting and under-recording

Qualitative and quantitative data were useful for understanding some of the reasons that might explain non-detection of incidents in the module. We identified the following organizational and individual barriers for reporting/recording.

The lack of resources and working conditions. In overcrowded prisons, with scarce resources, and high prevalence of violence, recording these events is just one among many functions of guards, which is perceived as neither urgent nor relevant in relation to other daily tasks. Further, prison guards are often unsure about the relevance of the data reported or what role it has in the collection of information and decision making in the prison policies. For example, some interviewees told us how “electronic files die in module A […] because there are no resources, we have no internet”. There is no network system that connects all the modules of the prison, so the incidents forms need to be physically handed in by the guards at the central office. Once the form was handed in, guards did not know what happened with incidents recorded by them, which made them feel that reporting incidents was not worthwhile, thus reducing the chance of future reporting.

Inmates’ prison code/rules. In our informal talks during the ethnography and in our interviews, inmates who endorsed the ‘prison code’ had more conflicts and violent incidents because violence was a sign of respect and status in the module. As we were told in one interview: “Most of the violence is not reported, it remains between us […] it is a code that we have’ “Inmate 12). Additionally, many inmates also referred to the costs and negative connotations associated with reporting to authorities about any victimization. Those who ‘betrayed the codes’ and reported the incidents to the guards were called ‘rats’ and were physically punished. It is also difficult to report victimization to guards without everyone in the module quickly finding out ‘who betrayed the codes’ (“quickly […] it comes to light super-fast”). Answers to our survey were consistent, showing that, among inmates who were victimized, those who agreed with items related to ‘prison codes’ yielded lower levels of reporting to authorities (Table 3).

Table 3: Reporting victimization and prison codes among respondents


Reporting (%)


Disagreeing with item

Agreeing with item

It is alright to assault another inmate if they start a fight



If another inmate makes me really mad, they deserve to be assaulted



It is sometimes necessary to fight to protect my honour



I am always ready to fight if someone tries to take advantage of me



If someone tries to hurt me, I will try to get even



Lack of punishment and trust in the system. In conversations during the ethnographic fieldwork when we raised the issue of reporting victimization some inmates laughed and asked “what for?” Several respondents thought it very unlikely that reporting would lead to any punishment of perpetrators. In fact, when inmates were asked about whether they were punished for violent incidents, many told us that they received either no sanction or irrelevant sanctions (for example, not allowed visits when no one visited them): “I got into physical fights many times and I was never sanctioned” (Inmate 3). Absence of enforcement of punishment was corroborated by our survey: of 209 inmates who answered the survey 46 per cent had been involved in violent behaviours in the last two months, yet, only 2 per cent of those perpetrators were sanctioned (Figure 2), and of those three sanctioned inmates, only one received a severe sanction (approximately 1 per cent of perpetrators!).

Figure 2: Waffle chart of reported perpetration of violent incidents and sanctions, in Module 3 during a two-month period

Lack of staff motivation. Both inmates and prison managers referred systematically to the lack of motivation and the reluctance of some prison guards to carry out their work responsibilities, due to poor support: “I see prison guards […] particularly unwilling […] they’re unmotivated, like they’re tired, like they don’t feel supported” (Prison Manager 1). Job dissatisfaction in guards is associated with lack of recognition and a lack of involvement in decision making in the module even though they have extensive knowledge and several years of experience in the position.

Staff perceptions and attitudes towards inmates: Some prison guards described themselves as ‘victims’ of a system that has led them to face violence so frequently, to the point of considering ‘normal’ several violent behaviours in the module such as knife-fights where inmates are seriously injured. Prison guards stated that managers “neglected the […] physical security part […] They neglected this area. Instead of having twelve policemen [prison guards] in one module, you start having nine, eight […] and today with five or four” (Prison guard 6). Violence was normalized and justified by guards as conflicts between inmates to be sorted out by them and which are not the guards’ business. Conversely, inmates generally believed that ‘[Prison guards] are not interested […] because they’re cops! Let them kill each other! They are trashy people, they are chavs.’


Prison violence has been researched for almost five decades with a focus on macro and micro predictors, the role of the environment, management and individual characteristics and what programmes can reduce it (McGuire, 2018; Day et al, 2022). Yet, only recently has a Portuguese study (Braga et al, 2019) examined the dark figure of prison violence. The current study expands on Braga et al’s (2019) study by examining prison violence through official records and self-report survey data, but including also qualitative interviews and observations to explore not only the gap between official and non-official incidents but also the potential reasons for under-reporting and under-recording. Our analysis provided four key findings.

First, we found a gap between the under-reported victimization and the official reports. Of the incidents that took place in the module, 85%were not registered in official reports within a period of two months. Our results showed a higher gap than Braga et al (2019), who found a 75 per cent difference between the self-report of inmate-on-inmate victimization using the Prison Violence Inventory (PVI; Warren et al, 2002) and official data from prison. Differences might be associated with dissimilar time frame and type of methods used to measure under-reporting in both studies. Particularly, when inmates report victimization less serious incidents might not be perceived, or might tend to be forgotten if they happened several months ago (Connel and Farrington, 1996).

Second, our study shows that reporting is related with the inmate’s role in prison incidents and the victim–offender overlap. The existence of the victim—offender overlap in the community is well documented, but only recently it has been explored empirically in inmate populations (Daquin and Daigle, 2021). Our survey data corroborates the victim—offender overlap with regard to prison violence: 40 per cent of our sample were both victims and offenders, and they tended to report less incidents as compared to pure victims, and particularly in comparison to inmates who suffer multiple victimizations. This is consistent with routine activity theory: those involved in crime with frequent contact with other offenders also have more ‘victimogenic lifestyles’ because they are attractive victimization targets, since they are less likely to call and report to the police (Jensen and Brownfield, 1986).

Third, our quantitative and qualitative data provided evidence that non-reporting is related with inmates’ prison cultural codes. Two potential associated mechanisms seem to be in play. Dealing with conflicts using violence and without resorting to prison officers is a sign of respect and honour (Gambetta, 2011) and is a skill that inmates need to learn in order to survive in hostile and dangerous environments (Sykes, 1958). This is clear in qualitative testimonies but also in how those victims who endorse subcultural codes tend to show lower reporting rates in our survey results. However, fear of punishment from other inmates also deters inmates from reporting, in line with rational choice theories and the deterrent role of informal sanctions (Apel and DeWitt, 2018) and with some research that shows reasons for not reporting sexual assault on prison involve fear of retaliation and harassment from the perpetrator, together with embarrassment and stigma (Miller, 2010).

Finally, lack of trust in the functioning of the institution also affects reporting of violent incidents. Both inmates and prison guards reveal lack of trust in how reporting is used by the prison system; in particular, inmates don’t believe that it will lead to effective enforcement of punishment of the perpetrators. These results are consistent with evidence from outside prison, where victims in the community fail to report because of fear, feelings of helplessness and the perceived powerlessness of the police and the threat of further victimization from authorities (Kidd and Chayet, 1984). Improving trust and legitimacy of prison officers and authorities is a relevant organizational barrier to be overcome, which can improve the reporting but can also have positive externalities in terms of reduction of misconducts and violence (Reisig and Mesko, 2009; Steiner and Wooldredge, 2018).

Practical implications

This study highlights the importance of organizational factors in the reporting of prison violence. In order to prevent prison violence, it is key to tackle some organizational barriers so as to obtain a more valid estimation of prison incidents. Our recommendations are: first, reporting can increase by improving poor prison conditions, particularly overcrowding. In our study, the prisoner– guard ratio was 50:1, when it should be 5:1 (Zeng, 2018). Second, the submission process for incident forms and the tracking of cases could be improved by developing standardized and easy-to-fill forms, as well as a computerized system containing a database in which staff have easy access to the cases reported in addition to having periodic meetings with managers to inform about cases. Third, reporting can be improved by investing in training the staff to follow protocols for reporting incidents, and designing informative, short and easy-to-submit forms. Fourth, to increase staff motivation, managers should involve officers in the decision-making process for reporting incidents, and ensure that their voice is heard and considered. Finally, clear rules and a reward/punishment system need to be established that are tailored to what inmates’ value; encourage inmates to report incidents; and protect victims who report by providing certain and fast punishment to those inmates who retaliate against inmates who report.


The present study has some relevant limitations. First, the period of observation included only one observer, was short (two months) and excluded nights and weekends. Second, we used a small convenience sample in a very violent module which was not representative of other modules within the prison, let alone of other prisons of the country. Third, the measure of prison violence in the survey was a short screening version of eight items. Fourth, given that the analysis of reasons for under-reporting was based on qualitative data or descriptive analysis using a small non-representative sample, we were not able to adequately test their empirical relevance. Finally, we were not given access to prison health statistics, which would have been a very valuable source of information to triangulate with prison records and our data.


Contributions to prison violence research to date have been important but they have focused on factors associated with violence rather than on estimating the dark figure or understanding factors associated with under-reporting. This study provides a first step with a mixed-method design that has been lacking in previous studies. Furthermore, this study found several barriers affecting the reporting of prison incidents by both staff and inmates. Future studies should corroborate the presence of these barriers in other prisons and countries (and potential moderators) in order to develop programmes and strategies targeting inmates, staff and management to improve the reporting of prison violence. A less biased measurement of prison violence is a first necessary step to reduce the problem of prison violence.


This work was supported by the Sectorial Commission of Scientific Research [Comisión Sectorial de Investigación Científica] under Grant number FINCSOCCSIC38. The funders had no role in study design, data collection and analysis, interpretation of the data, decision to publish or preparation of the manuscript. We would like to thank the executive director at that time of the Uruguayan Prison System, Ana Juanche, for her trust and support. Additionally, we would also like to thank all the inmates who participated in the study, and Victoria Gambetta and Gustavo Medina for their help in collecting data.


Abbiati, M., Palix, J., Gasser, J. and Moulin, V. (2019) Predicting physically violent misconduct in prison: A comparison of four risk assessment instruments. Behavioral Sciences & The Law 37(1), 61–77.

Apel, R. and DeWitt, S.E. (2018) Informal and formal sanctions. In: Deterrence, choice, and crime (pp 139–156). Routledge.

Auty, K.M. and Liebling, A. (2020) Exploring the relationship between prison social climate and reoffending. Justice Quarterly 37(2), 358–381.

Bottoms, A.E. (1999) Interpersonal violence and social order in prisons. Crime and Justice 26, 205–281.

Braga, T., De Castro-Rodrigues, A., Severim Di Folca, S.M. and Abrunhosa Gonçalves, R. (2019) How dark are dark figures? Official and self report rates of inmate-on-inmate victimization. Victims & Offenders 14(6), 745–757.

Byrne, J.M. and Hummer, D. (2007) Myths and realities of prison violence: A review of the evidence. Victims and Offenders 2(1), 77–90.

Caravaca-Sánchez, F., Aizpurua, E. and Wolff, N. (2022) The prevalence of prison-based physical and sexual victimization in males and females: A systematic review and meta-analysis. Trauma, Violence, & Abuse 0(0).

Connell, A. and Farrington, D. (1996) Bullying among incarcerated young offenders: Developing an interview schedule and some preliminary results. Journal of Adolescence 19(1), 75–93.

Daggett, D.M. and Camp, S.D. (2009) Do official misconduct data tell the same story as the individuals who live in prison? Criminal Justice Review 34(3), 428–449.

Daquin, J.C. and Daigle, L.E. (2021) The victim–offender overlap in prison: Examining the factors associated with group membership. Journal of interpersonal violence 36(23–24), NP13439–NP13462.

Day, A., Newton, D., Cooke, D. and Tamatea, A. (2022) Interventions to prevent prison violence: A scoping review of the available research evidence. The Prison Journal 102(6), 745–769.

Gadon, L., Johnstone, L. and Cooke, D. (2006) Situational variables and institutional violence: A systematic review of the literature. Clinical Psychology Review 26, 515–534.

Gambetta, D. (2011) Codes of the Underworld. In: Codes of the Underworld. Princeton University Press.

Gear, S. (2007) Fear, violence and sexual violence in a Gauteng juvenile correctional centre for males. Centre for the Study of Violence and Reconciliation. Briefing Report No 2.

Gudjonsson, G., Sigurdsson. J.F., Skaptadottir, S. and Helgadottir, þ. (2011) The relationship of violent attitudes with self-reported offending and antisocial personality traits. Journal of Forensic Psychiatry & Psychology 22(3), 371–380.

Hagemann, O. (2008) Conditions of imprisonment – victimization and conflicts in European prisons. Journal of Ethnicity in Criminal Justice 6(4), 281–302.

Huang, R. (2012) RQDA: R-based qualitative data analysis. R package version 0.2–3. URL

Ireland, J.L. (2002) Direct and indirect prisoner behavior checklist – Revised (DIPC-R). Preston: University of Central Lancashire.

Jensen, G. F., & Brownfield, D. (1986). Gender, lifestyles, and victimization: Beyond routine activity. Violence and victims1(2), 85-99.

Kidd, R.F. and Chayet, E.F. (1984) Why do victims fail to report? The psychology of criminal victimization. Journal of Social Issues 40(1), 39–50.

Kury, H. and Smartt, U. (2002) Prisoner-on-prisoner violence: Victimization of young offenders in prison. Some German findings. Criminal Justice 2(4), 411–437.

Levan Miller, K. (2010) The darkest figure of crime: Perceptions of reasons for male inmates to not report sexual assault. Justice Quarterly 27(5), 692–712.

Logan, C.H. (1992) Well kept: Comparing quality of confinement in private and public prisons. The Journal of Criminal Law and Criminology 83, 577–613.

McGuire, J. (2018) Understanding prison violence: A rapid evidence assessment. Analytical Summary. H.M. Prison and Probation Service.

Miller, L.K. (2010) The darkest figure of crime: Perceptions of reasons for male inmates to not report sexual assault. Justice Quarterly 27(5), 692–712.

Modvig, J. (2014). Violence, sexual abuse and torture in prisons. in S. Enggist, L. Møller & C. Udesen (eds.). Prisons and Health. Copenhague: World Health Organizaztion, Europe, pp. 19-26.


Penney, T.L. (2014) Dark figure of crime (problems of estimation). The encyclopaedia of criminology and criminal justice, 1–6. Wiley.

Reisig, M.D. and Mesko, G. (2009) Procedural justice, legitimacy, and prisoner misconduct, Psychology, Crime & Law 15(1), 41–59.

Sánchez, M. y D. Piñol (2015). Condiciones de vida en los centros de privación de libertad en Chile. Análisis a partir de una encuesta aplicada a siete países de Latinoamérica. Santiago: Centros de Estudios Públicos en Seguridad Ciudadana.

Steiner, B. & C. M. Cain (2016). The Relationship between Inmate Misconduct, Institutional Violence, and Administrative Segregation: A Systematic Review of the Evidence. NCJ 250320. En National Institute of Justice (2016), Restrictive Housing in the United State: Issues, Challenges, and Future Directions. Washington DC: US Department of Justice, National Institute of Justice.

Steiner, B. and Wooldredge, J. (2018) Prison officer legitimacy, their exercise of power, and inmate rule breaking. Criminology 56(4), 750–779.

Steiner, B., Butler, D. and Ellison, J.M. (2014) Causes and correlates of prison inmate misconduct: A systematic review of the evidence. Journal of Criminal Justice 42(6), 462–470.

Sykes, G.M. (1958) The society of captives: A study of a maximum security prison. Princeton, NJ: Princeton University Press.

Teasdale, B., Daigle, L.E., Hawk, S.R. and Daquin, J.C. (2016) Violent victimization in the prison context: An examination of the gendered contexts of prison. International Journal of Offender Therapy and Comparative Criminology 60(9), 995–1015.

Trajtenberg, N. and Sanchez de Ribera, O. (2020) Prison violence in Uruguay. National Institute of Rehabilitation, Penitentiary Institutions, Ministry of Interior, Uruguay.

Turanovic, J.J. and Pratt, T.C. (2019) Thinking about victimization: Context and consequences. Routledge.

Warren, J.I., Hurt, S., Loper, A.B., Bale, R., Friend, R. and Chauhan, P. (2002) Psychiatric symptoms, history of victimization, and violent behavior among incarcerated female felons: An American perspective. International Journal of Law and Psychiatry 25, 129–149.

Wolff, N., Shi, J. and Bachman, R. (2008) Measuring victimization inside prisons. Questioning the questions. Journal of Interpersonal Violence 26(10), 1343–1362. doi:10.1177/0886260508314301

Wolff, N., Blitz, C.L., Shi, J., Siegel, J. and Bachman, R. (2007) Physical violence inside prisons: Rates of victimization. Criminal Justice and Behavior 34(5), 588–599.

Wooldredge, J. and Steiner, B. (2014) A bi-level framework for understanding prisoner victimization. Journal of Quantitative Criminology 30(1), 141–162.

Zeng, Z. (2018) Jail inmates in 2016. NCJ, 251210.

[i] The name of the module has been modified to preserve the anonymity of the staff and inmates.

No comments here
Why not start the discussion?