Abstract
Purpose:
Emerging evidence indicates that incarcerated populations’ perceptions of dehumanization by officers are prevalent, yet measures of it are few, and to our knowledge, no self-report measure of dehumanization from officers exists. To fill this gap, we have developed the Perceived Dehumanization from Officers Scale (PDOS), which is designed as a brief measure to assess perception of officer treatment as dehumanizing.
Methods:
In this article, we provide preliminary evidence from two studies examining the reliability and validity of the PDOS. In Study 1, a jail sample (n = 411), we analyzed the exploratory factor structure, internal consistency, and discriminant validity (in relation to Procedural Justice) of the PDOS. Additionally, using a cross-sectional Ordinary Least Squares (OLS) regression analysis, we related independent variables with the PDOS, the dependent variable. In Study 2, a prison sample (n = 2,993), we confirmed the findings from Study 1.
Results:
The PDOS appears to be a psychometrically sound measure of perceived dehumanization from officers with strong association between perceptions of procedural justice and perceived dehumanization from officers.
Conclusions:
The PDOS provides opportunity for future research, intervention through rehumanization efforts, and signals the important officer treatment. Importantly We close by discussing implications of these studies, limitations, and future research directions to further develop and test the PDOS.
Keywords: Dehumanization, Procedural Justice, Incarceration
Introduction
Dehumanization involves an abnegation of the qualities essential to being human (Haslam, 2006). This can be experienced as degradation, aggression, denial, or intellectual derision. Dehumanization may occur under conditions of conflict or extreme negative evaluation; however, these are not the only types of precipitants (Haslam, 2006). It may also be internalized through daily cognitive processes and experienced as a social phenomenon.
Dehumanization is highly prevalent within U.S. prisons and jails (Stinglhamber et al., 2022; Vieraitis et al., 2018). People incarcerated in the U.S. live separate from society and are stripped of many fundamental rights, and privileges (Deska et al., 2018; Kelman, 1973; Sykes, 1958). Prisons and jails are fundamentally organized to “protect the community against what are felt to be intentional dangers to it, with the welfare of the persons thus sequestered not the immediate issue” (Goffman, 1961 p. 4–5). In addition, as Goffman (1961) noted, entrance into prisons and jails represents “prima facie evidence that one must be the kind of person the institution was set up to handle… automatic identification of the inmate is not merely name-calling; it is at the center of a basic means of social control” (p. 84). That is, prisons and jails are disconnected from society to protect the public from the dangerous people housed in them, staffed by people who may well have preconceived judgements about the type of person who would be in such a place, and there may be an institutional emphasis on “othering” inmates as a means of establishing social distance in a subordinate-superordinate relationship between prisoner and staff. These fundamental characteristics of prisons and jails provide the type of environment which permits (and may even encourage) the treatment of people in a manner that in most any other context would be deemed unjustifiable. In other words, the correctional officer-prisoner power dynamic provides the basic framework for a relationship that presents frequent opportunity for the dehumanization of others.
Within the penal system, dehumanization literature has called for more humanistic treatment by officers (Guenther, 2012; Kelso, 2014). Empirically, dehumanization’s impact extends beyond confinement into courtrooms—participants determined incarcerated people with dehumanizing descriptions (compared to control incarcerated people) should receive harsher punishments and judged them to have reduced mental sophistication (Deska et al., 2018; 2020). Within carceral institutions, officer’s experiences of organizational dehumanization (i.e., feeling less than human as a staff member) have been cross-sectionally associated with perpetuating depersonalization and dehumanization on to incarcerated persons (Stinglhamber et al., 2022). Therefore, dehumanization within and around carceral environments can carry very real penalizations to incarcerated people.
While dehumanization is likely to have important internalized psychological and stigmatizing impacts on people in prison (cf. ANONYMISED), it may also impact behavioral outcomes such as prison rule compliance. For example, other psychological constructs such as procedural justice, on which higher levels indicate an individual views the procedural system as fair and just, have been linked to higher levels of public compliance with police authority (Mazerolle et al., 2013; Tyler, 2006a). Moreover, higher levels of procedural justice in correctional facilities have been found to promote obedience to correctional rules (Alward & Baker, 2019; Baker et al., 2019; Reisig & Mesko, 2009). Given that prior research within the context of procedural justice has demonstrated a consistent relationship between how people perceive they are treated by authority figures and their obedience to rules, it may also be the case that people’s perceptions of how they are treated within the context of dehumanization have similar consequences.
Given the importance of conducting additional research on the role of dehumanization for the long-term psychological and behavioral welfare of people incarcerated in prisons and jail, a measure of incarcerated people’s perceptions of dehumanization is needed. The goal of this study is to develop such a measure. Below we discuss the origins of the questions included on The Perceived Dehumanization from Officers Scale (PDOS) and test the measure’s reliability and validity using two studies, one of people incarcerated in jail and a second of people incarcerated in prison.
Item Development
The seminal work of Bandura and colleagues (1975) aimed to understand how responsibility and dehumanization impacted people’s willingness to administer pain to others. The study had three groups: a dehumanized condition, a humanistic one, and a neutral group. The dehumanized group of people were described to participants as “animalistic” and a “rotten bunch.” Consistently, the highest mean pain shock was given to the “dehumanized” recipients, both when responsibility was diffused by a group and in the condition in which there was no diffusion of any sort (i.e., individual responsibility). When performers were humanized, participants more readily disapproved the use of physical punishment (Bandura et al., 1975). Two of the items adapted for the current studies were developed from the Bandura et al. work’s use of “animalistic,” and the opposite of the humanized group being described as “perceptive,” resulting in the use of “stupid.”
The third item used for this study extends from the key loss of identity that accompanies the experience of dehumanization (e.g., Reimann & Zimbardo, 2011; Zimbardo, 1969; Zimbardo et al., 1999). Specifically, it is commonplace for people in prison and jails to be referred to by their assigned correctional number rather than by name. The final item in the proposed measure of dehumanization aims to capture this experience by asking participants if they feel as though correctional officers view people incarcerated in jails/prisons simply as numbers. Those who feel this is true may experience of loss of identity and a greater sense of dehumanization. These three items, while initially inspired from Bandura and colleagues, align with the theoretical integrative review of dehumanization, which proposed both animalistic and mechanistic subsets, offered by Haslam in 2006 (Haslam, 2006) which has been previously adapted for work in carceral facilities (Stinglhamber et al., 2022).
We used a three-item measure to maximize the likelihood of implementation within correctional facilities. Other studies have successfully defended a three-item measure (cf. Ribeiro et al., 2011; Stanley et al., 2021). Additionally, the confirmatory factor analysis with three indicators results in a model that is just identified; therefore, this measure still meets essential statistical thresholds.
Study 1
The intention of Study 1 was to exploratorily examine the factor structure, internal consistency, and discriminant validity (in relation to Procedural Justice) of the Perceived Dehumanization from Officers Scale (PDOS). Furthermore, Study 1 will include a cross-sectional Ordinary Least Squares (OLS) regression analysis of independent variables within a jail sample (n = 411) and their association with officer dehumanization, the dependent variable, using demographic control variables selected based on widespread use in criminology research (ANONYMIZED, 2022). We hypothesize that the Procedural Justice scale will have a negative correlation with scores on the PDOS. Additionally, we expect this correlation’s magnitude to be moderate, as both are related to perceptions of treatment within the jail context, suggesting they are related, yet distinct, constructs.
Methods
Setting
Between March of 2018 and March of 2019, the third author collected self-report data in two suburban jails located in Texas and Florida. Eligibility requirements for the study were as follows: participants were required to be 18 years of age at the time of data collection, have a fourth-grade English literacy level, and not be currently housed in mental health units or administrative segregation. Researchers visited the eligible housing units and provided both a verbal and written description of the procedure, purpose, consent, and the voluntary and anonymous nature of participation. Consent was deemed granted by taking the survey and all participants were able to keep a consent form. The entirety of the study involved self-report questionnaires on paper. To minimize potential coercion, only research personnel were involved in survey administration. There were no incentives for participation.
A total of 548 individuals voluntarily consented to participation and met the eligibility criteria; this response rate aligned with prior work in jails with self-report instruments (e.g., Fox et al., 2010; Trotter et al., 2018). Due to the sensitive nature of some questions, an on-staff psychologist was available at both facilities for debriefing. All components of the study, survey, and consent were approved by the affiliated university’s Institutional Review Board, state department of corrections ethics review board, and the state director of the department of corrections. Due to confidentiality of the state department of corrections ethics boards, the study analysis code for this study is available, but study materials are not.
Participants
The final analytic sample, after dropping some cases for missing data, is 411. We used The Statistical Package for the Social Sciences (SPSS version 25) to initially clean the data. All missing values were assessed using a Little’s MCAR test, which suggested that missing data were missing at random (MCAR), χ2 (2) = 2.140, p = .343. Additionally, exploratory tests for missing at random (MAR) were applied (cf. Garson, 2015), supporting multiple imputation (MI) to address missing data. Descriptively, this sample is 72% male, with a mean age of 35. The majority of the sample self-identified as White (42%). The average incarceration time was 133 days (for current incarceration, maximum values were capped at 365 days). The majority of the sample reported being held for a nonviolent crime (78%). See Table 1 for a full description.
Table 1.
Jail Sample Descriptive Statistics
| Variables | M | SD | Skew | Kurtosis |
|---|---|---|---|---|
|
| ||||
| PDOS1 | 3.119 | .989 | −.891 | 2.700 |
| PDOS2 | 3.032 | .988 | −.685 | 2.369 |
| PDOS3 | 2.966 | 1.026 | −.543 | 2.067 |
| PJ1 | 2.114 | .990 | .359 | 1.983 |
| PJ2 | 2.024 | .972 | .461 | 2.060 |
| PJ3 | 2.582 | .940 | −.192 | 2.156 |
| Age | 35.304 | 10.542 | .633 | 2.798 |
| Sex | .723 | .448 | −.995 | 1.989 |
| Violent Offense | .221 | .416 | 1.342 | 2.801 |
| Days Incarc. | 132.874 | 124.89 | .853 | 2.304 |
| White | .418 | .494 | .330 | 1.109 |
NOTES: Days Incarcerated was truncated at 365+ days.
ABBREVIATIONS: PDOS = Perceived Dehumanization from Officers; PJ = Procedural Justice; Days Incarc. = Days Incarcerated.
Measures
Dehumanization.
We used the three item PDOS to assess perception of dehumanization. These items asked respondents to answer on a four-point Likert-type scale of agreement (ranging from strongly disagree to strongly agree): “Most officers view inmates as just numbers,” “Most officers think inmates are stupid,” and “Most officers treat inmates like they’re animals.” Reliability and validity will be addressed below.
Perceived Procedural Justice.
We used three items from the Procedural Justice (PJ) scale to assess discriminant validity of the PDOS. These items asked respondents to answer on a 4-point Likert-type scale of agreement (ranging from never to almost always): “Officers at this facility treat inmates fairly and with respect,” “When there’s a problem officers give you the chance to tell your side of the story,” and “How often do officers treat inmates fairly?” (Alward et al., 2021). The PJ scale has shown strong reliability and validity in previous research (Baker & Gau, 2018; Gau, 2015; Tyler, 2006b) and was included to assess the construct validity of the PDOS. In this sample, the measure also demonstrated good internal consistency (α = .78).
Demographic Variables.
We also used a variety of demographic variables as independent variables. The survey included demographic questions about age (continuous), gender (women/other vs. male), race/ethnicity (White vs. other), time in facility (in days), and current reason for incarceration (recoded to reflect violent [e.g., assault, sexual assault, sexual battery, robbery, homicide] versus nonviolent offense [e.g., property, traffic, drug, or violation of parole]).
Results and Discussion
Internal Consistency
Study 1 means and standard deviations are included in Table 1. To evaluate scale reliability, we examined inter-item correlations between the PDOS items presented in Table 2. Given the strong inter-item correlations, it is not surprising that the coefficient alpha in this sample was reasonably robust (α = .83).
Table 2.
Jail Sample Factor Loadings and Inter-item correlations (n = 411)
| Factor Loadings | PDOS1 | PDOS2 | PDOS3 |
|---|---|---|---|
|
| |||
| PDOS 1 .796 | 1 | ||
| PDOS 2 .855 | .712 | 1 | |
| PDOS 3 .852 | .708 | .776 | 1 |
ABBREVIATIONS: PDOS = Perceived Dehumanization from Officers.
NOTES: All Correlations significant p < .001.
Exploratory Factor Analysis
Next, we used a classic factor-analytic approach (cf. Gorsuch, 1983) of initial exploratory factor analysis (EFA) with Oblique/Promax Rotation. Due to an a priori decision (i.e., three items included in the PDOS), we only examined a unifactorial solution to ensure model identification. As seen in Table 2, the items all significantly loaded onto a single factor (PDOS 1 = .796, PDOS 2 = .855, PDOS 3 = .852, all ps <.001) with excellent fit, measuring the latent construct of perceived dehumanization from officers.
Preliminary Validity
To examine the discriminant validity of the PDOS in relation to PJ, to determine if they are, in fact, separate and distinct, we conducted a second EFA. Here, we subjected the individual items from the dehumanization scale (e.g., PDOS) and the procedural justice scale (e.g., PJ) to a joint, two-factor analysis extraction rotated to “simple structure.” We used Oblique/Promax Rotation. In the rotated model, see Table 3, two distinct factors emerged with weak cross loadings (cf. a common threshold of .40 is considered adequate). Therefore, this provides supporting evidence for the discriminant validity of the PDOS in comparison to PJ (Factor 1: PDOS 1 = .817, PDOS 2 = .866, PDOS 3 = .808, Factor 2: PJ 1 = .785, PDOS 2 = .729, PDOS 3 = .564). Model fit statistics (Akaike’s information criterion, AIC = 27.349; Bayesian information criterion, BIC = 71.554) signaled quality model fit for the two-factor model (cf. one-factor model AIC = 289.439; BIC = 313.551).
Table 3.
Discriminant Validity of Dehumanization and Procedural Justice using Exploratory Factor Analysis (n = 411)
| Unrotated Factor Model | Rotated Factor Model | |||
|---|---|---|---|---|
|
| ||||
| Variable | Factor 1 Loadings | Factor 2 Loadings | Factor 1 Loadings | Factor 2 Loadings |
|
| ||||
| PDOS1 | .720 | .342 | .817 | .043 |
| PDOS2 | .777 | .351 | .866 | .028 |
| PDOS3 | .834 | .242 | .808 | −.113 |
| PJ1 | −.615 | .486 | .001 | .785 |
| PJ2 | −.562 | .456 | .011 | .729 |
| PJ3 | −.530 | .313 | −.095 | .564 |
|
| ||||
| Model Selection Criteria for Number of Factors | ||||
| One Factor | Two Factor | |||
| AIC | 289.439 | 27.349 | ||
| BIC | 313.551 | 71.554 | ||
ABBREVIATIONS: PDOS: Perceived Dehumanization from Officers; PJ: Procedural Justice AIC = Akaike's information criterion; BIC = Bayesian information criterion.
NOTES: Oblique/Promax Rotation Used.
Regression onto Independent Variables
Finally, we ran an Ordinary Least Squares (OLS) regression of dehumanization onto each of our independent variables including age, sex, violence of offence, days incarcerated, race, and perceptions of procedural justice. Results are detailed in Table 4 and suggest that incarcerated people who identify as younger (β = −0.17, SE = .004, p =.01), have spent more time incarcerated (β = 0.14, SE = .000, p =.01), and perceive the procedural justice system to be inequitable (β = −.42, SE = .054, p =.001) are more likely to report dehumanization from officers. As perceptions of procedural justice and dehumanization from officers loaded onto two distinct factors in the EFA, see Table 3, as well as were negatively associated with one another from the OLS regression (i.e., zero-order correlation of r1,2 = −.415, p <.001), they are associated, yet unique, constructs.
Table 4.
Jail Sample Bivariate OLS Regression of Dehumanization on each Independent Variable
| Variables | b | SE | B | R2 |
|---|---|---|---|---|
|
|
||||
| Procedural Justice | −.468*** | .054 | −.415 | .170 |
| Age | −.015** | .004 | −.169 | .029 |
| Sex | .013 | .101 | .007 | .000 |
| Violent Offense | .039 | .102 | .018 | .000 |
| Days Incarcerated | .001** | .000 | .137 | .019 |
| White | −.144 | .089 | −.078 | .006 |
ABBREVIATIONS: b = unstandardized coefficient, SE = robust standard error; B = standardized coefficient.
p < .01;
p < .001
Discussion
Structural analyses of the PDOS in this sample indicate that all items significantly load on to a single factor, which suggests that the items are good indicators of the latent construct of dehumanization. Furthermore, regarding validity, findings from Study 1 provide preliminary discriminant validity, here regarding procedural justice, of the PDOS as a measure of perceived dehumanization. Cross-sectionally, reported dehumanization is also significantly related to incarcerated people who are characterized as younger, those who have spent more time within the system, and individuals who do not believe that process of procedural justice is fair.
Study 1 provided useful preliminary data; however, this investigation, like all research, has limitations. Because the data were collected from a jail sample, we were curious as to how this would fare in a larger, more regulated sample within the penal system, such as a prison. Additionally, exploration of a confirmatory factor analysis with additional measurement validity would strengthen the PDOS as an index of perceived dehumanization from officers. As such, in Study 2, the psychometric properties, factor-analytic structure, and correlates of the PDOS were examined in a large prison sample.
Study 2
Methods
Setting
In 2015, the second author collected self-report measures from incarcerated individuals across five separate prison facilities in a southern state. These facilities ranged from minimum to maximum security. Eligibility requirements required all interested people incarcerated to be at least 18 years of age at the time of data collection, English speaking, and not be currently housed in mental health units or administrative segregation. All participation was voluntary, and participants were provided with verbal and written descriptions of study procedures, purpose, and consent. Surveys were self-report questionnaires administered with paper and pencil. To avoid coercion, among other procedures, correctional officers did not distribute or collect the surveys. All components of the study, survey, and consent were approved by the affiliated university’s Institutional Review Board, state department of corrections ethics review board, and the state director of the department of corrections. The final sample size (n = 3,531) was composed of 59% of all eligible participants (in line with prior work in prisons with self-report surveys, e.g., Baker & Gau, 2018; Mitchell et al., 2020). To see the full description of data collection procedures, please see Alward et al. (2021). The final analytic sample, after dropping some cases for missing data, using Maximum Likelihood Estimates, is 2,993. Due to confidentiality of the state department of corrections ethics boards, the study analysis code for this study is available, but study materials are not.
Participants
The sample is 69% male, with a mean age of 38.8. A plurality of the sample self-identified as White (37%). The average incarceration time was 11.3 years (for current incarceration). A plurality of the sample reported being held for a nonviolent crime (41.6%). See Table 5 for a full description.
Table 5.
Prison Sample Descriptive Statistics of Independent Variables
| Variables | M | SD | Skew | Kurtosis |
|---|---|---|---|---|
|
| ||||
| PDOS1 | 3.246 | .761 | −.783 | 3.17 |
| PDOS2 | 3.170 | .804 | −.664 | 2.776 |
| PDOS3 | 2.956 | .866 | −.323 | 2.199 |
| PJ1 | 1.79 | .805 | .641 | 2.506 |
| PJ2 | 1.959 | .859 | .370 | 2.094 |
| PJ3 | 2.428 | .761 | −.152 | 2.567 |
| Age | 38.805 | 11.304 | .615 | 2.821 |
| Sex | .688 | .463 | −.811 | 1.658 |
| Violent Offense | .584 | .493 | −.340 | 1.115 |
| Years Incarcerated | 11.294 | 8.586 | 1.188 | 4.420 |
| White | .367 | .482 | .554 | 1.307 |
ABBREVIATIONS: PDOS = Perceived Dehumanization from Officers; PJ = Procedural Justice.
Measures
Dehumanization.
Refer to Study 1 for a detailed description of the PDOS measure.
Procedural Justice.
Refer to Study 1 for a description of the PJ. In this sample, the measure also demonstrated adequate internal consistency (α = .69). In line with the hypothesis noted in Study 1, we anticipate a moderate negative association between the PJ and PDOS, hypothesizing that they are related constructs.
Demographic Variables.
Refer to Study 1 for a detailed description of the demographic questions used in this study.
Results
Study 2 means and standard deviations are included in Table 5.
Internal Consistency
To evaluate scale reliability, we examined inter-item correlations between the PDOS items presented in Table 6. Given the strong inter-item correlations, this resulted in a coefficient alpha that was reasonably good (α = .77).
Table 6.
Prison Sample Confirmatory Factor Loadings and Inter-item Correlations (N = 2993)
| Factor Loadings | PDOS1 | PDOS2 | PDOS3 |
|---|---|---|---|
|
| |||
| PDOS1 .731 | 1 | ||
| PDOS2 .852 | .622 | 1 | |
| PDOS3 .726 | .530 | .618 | 1 |
ABBREVIATIONS: PDOS = Perceived Dehumanization from Officers.
Confirmatory Factor Analysis
Study 2 incorporates a confirmatory factor analysis (CFA) with maximum likelihood estimation (ML) to build upon the analysis presented in Study 1 to examine the structure of the PDOS. As seen in Table 6, the three items all significantly loaded onto a single factor (PDOS 1 = .731, PDOS 2 = .852, PDOS 3 = .726, all p’s < .001). These values suggest the one-factor model provides good fit to the data (coefficient of determination; CD = .831).
Preliminary Validity
To further delineate between Procedural Justice (PJ) and the PDOS scale, i.e., is dehumanization a sub-construct of PJ or distinct, we placed them together into a single model confirmatory factor analysis. When loading onto a single factor, see Table 7, this model shows poor model fit. In the two-factor model, Table 7, two distinct factors emerge with weak cross loadings (c.f. a common threshold of .40 is considered adequate). Therefore, this provides supporting evidence for the discriminant validity of the PDOS in comparison to PJ (Factor 1: PDOS 1 = .723, PDOS 2 = .811, PDOS 3 = .774, Factor 2: PJ 1 = .722, PDOS 2 = .664, PDOS 3 = .631). Model fit statistics, (Comparative Fit Index, CFI = .978; Tilburg Frailty Indicator, TFI = .958; Standard Root Mean Squared Residual, SRMR = .030; Root Mean Square Error of Approximation, RMSEA = .074) all adhered to suggestions of good model fit. Therefore, discriminant validity in the two-factor solution, is established.
Table 7.
Discriminant Validity of Dehumanization and Procedural Justice using Confirmatory Factor Analysis (n = 2993)
| Single Factor Model |
Two Factor Model |
|||||
|---|---|---|---|---|---|---|
| Variable | Factor Loadings | SE | Factor Loadings | SE | Factor Loadings | SE |
|
|
|
|||||
| PDOS1 | .690 | .012 | .723 | .011 | — | — |
| PDOS2 | .763 | .010 | .811 | .010 | — | — |
| PDOS3 | .783 | .009 | .774 | .011 | — | — |
| PJ1 | −.583 | .014 | — | — | .722 | .013 |
| PJ2 | −.546 | .015 | — | — | .664 | .014 |
| PJ3 | −.541 | .015 | — | — | .631 | .015 |
|
| ||||||
| χ2(df) | 662.852(9) | 139.351(8) | ||||
| RMSEA | .156 | .074 | ||||
| AIC | 38090.143 | 37568.643 | ||||
| BIC | 38198.216 | 37682.719 | ||||
| CFI | .889 | .978 | ||||
| TLI | .815 | .958 | ||||
| SRMR | .068 | .030 | ||||
ABBREVIATIONS: PDOS: Perceived Dehumanization from Officers; PJ: Procedural Justice; SE = Standard Error; χ2(df) = Chi-squared (degrees of freedom); RMSEA = Root mean squared error of approximation; AIC = Akaike's information criterion; BIC = Bayesian information criterion; CFI = Comparative fit index, TLI = Tucker-Lewis Index; SRMR = Standardized root mean squared residual.
NOTES: All factor loading are significant at p < .001.
Regression onto Independent Variables
Finally, we conducted a hierarchical linear regression of the PDOS onto independent variables of age, sex, violence of offence, years incarcerated, race, and perceptions of procedural justice. Results are detailed in Table 8 and suggest that incarcerated people who identify as younger (β = −.217, SE = .019, p =.001), do not identify as White (β = −.078, SE = .020, p =.001), identify as male (β = .142, SE = .020, p =.001), perceive the procedural justice system to be inequitable (β = −.707, SE = .016, p =.001), and have spent more time incarcerated (β =−.066, SE = .020, p =.01) are more likely to experience dehumanization from officers. All model fit statistics signal satisfactory fit. Finally, the coefficient of determination, also known as R2, was included to highlight that perceptions of procedural justice (CD = .78) highly impact perceptions of dehumanization from officers, though, they are still distinct from one another.
Table 8.
Prison Sample SEM Regression of Dehumanization on each Independent Variable
| Variables | B | SE | RMSEA | CFI | TLI | SRMR | CD |
|---|---|---|---|---|---|---|---|
|
|
|||||||
| Procedural Justice | −.707*** | .016 | .074 | .978 | .958 | .030 | .780 |
| Age | −.217*** | .019 | .095 | .983 | .950 | .028 | .047 |
| Sex | .142*** | .020 | .022 | .999 | .997 | .008 | .020 |
| Violent Offense | .031 | .020 | .008 | 1.000 | 1.000 | .006 | .001 |
| Years Incarcerated | −.066** | .020 | .060 | .993 | .979 | .019 | .004 |
| White | −.078*** | .020 | .054 | .994 | .983 | .017 | .006 |
ABBREVIATIONS: B = standardized coefficient; SE = robust standard error; RMSEA = Root mean squared error of approximation; CFI = Comparative fit index, TLI = Tucker-Lewis Index; SRMR = Standardized root mean squared residual; CD = Coefficient of Determination.
p < .01;
p < .001
Discussion
Building from Study 1, Study 2 showed a similar structural analysis of the PDOS, further indicating that all items significantly load on to a single factor. With two studies finding similar results this suggests that the PDOS is a strong construct capturing dehumanization from officers. Furthermore, the PDOS is similar, yet distinct, from procedural justice, providing potential important implications for future research. Finally, cross-sectionally, reported dehumanization is also significantly related to younger incarcerated individuals, males, those who do not identify as White, those who have spent more time within the system, and individuals who do not believe that process of procedural justice is fair.
General Discussion
Results from this study suggest that the PDOS is a psychometrically sound measure of perceived dehumanization from officers. Dehumanization may prove to be important; therefore, developing a measure to assess it provides opportunity for potential empirical conceptual relationships such as how dehumanization from officers impacts mental health (e.g., suicide risk). In both studies, there was a significant negative association between perceptions of procedural justice and perceived dehumanization from officers (i.e., Study 1: r = −.41 and Study 2: r = −.71). While the coefficient from Study 1 clearly indicates an association between perceptions of procedural justice and perceived dehumanization, it is not so strong as to emphatically suggest convergent validity between the two concepts. In Study 2, the coefficient is larger and represents fairly clear evidence for convergent validity. We posit that this indicates these constructs, while substantively related, are not identical. As both are measures relating to treatment in prisons, and other studies have linked perceptions of treatment to compliance (i.e., Baker & Gau, 2018; Brunton-Smith & McCarthy, 2016; Ryan & Bergin, 2022), dehumanization from officers may have similar associations. Specifically, procedural justice has been associated with underlying phenomena such as treatment fairness, equity, and officer competence (Brunton-Smith & McCarthy, 2016; Ryan & Bergin, 2022).
As dehumanization was a distinct construct in this paper, separate from procedural justice, it may deserve additional empirical attention in relationship to cooperation, rule compliance, and trust of authority. Future research should model perceived dehumanization from officers and its associations towards one’s likelihood to report rule breaking behavior, as prior work has shown incarcerated people self-report, and guards observe, more extensive rule-breaking than is officially recorded (Hewitt et al., 1984). More humane treatment from guards may reduce observed infractions and increase trust between officers and people incarcerated (Baker et al., 2014).
Additionally, there may be factors unique to dehumanization, such as associations between dehumanization from officers and self-harming behaviors (i.e., suicidality). Future work should longitudinally address the mental health implications of dehumanization from officers on people incarcerated. Furthermore, this relationship should be tested in a bidirectional nature, as officers may experience their own forms of dehumanization due to resource scarcity, job stress, and fear of victimization (Gordon & Baker, 2017). As an example, one study has shown that prison correctional officers who reported experiencing dehumanization from their organizations (i.e., being viewed as tools) resulted in reporting higher levels of depersonalization towards incarcerated persons (Higgins et al., 2022; Stinglhamber et al., 2022). Therefore, humane treatment across the penal system, i.e., jails, prisons, probation, may improve rule adherence, compliance, and mental wellness.
Overall, the findings supplied by both studies in this paper provide evidence of face, construct, and discriminant validity of the PDOS as measure of dehumanization. These findings provide potential utility and usefulness as brief measure, especially in penal settings. The PDOS, therefore, can be given in screenings, routinely, or after incidents within incarceration environments. This provides opportunity for the remedy of dehumanization, rehumanization, through humane officer treatment. Importantly, the PDOS provides a sound measurement tool that may be used to answer future empirical questions; however, such work extends beyond the scope of this project.
Despite the strengths towards the validity of the PDOS in this study, there are still important limitations to be considered. A notable limitation of the current study is both datasets were cross-sectional, therefore we are unable to assess the test-retest reliability of the PDOS. Furthermore, due to the number of initial items, Study 1 was unable to reduce and extract factors. Thus, future studies should continue to generate additional items that may add incremental information to the perceived dehumanization from officers scale. Theory suggests dehumanization may have subfactors (i.e., animalistic and mechanistic; Haslam, 2006). On the one hand, the correlations between items (ranging in Study 1 from r =.71-.78 and in Study 2 from r = .53-.62) may indicate potential future subfactors. On the other hand, this may be due in part to statistical noise (e.g., some may treat animals with more kindness than a devalued human, making animalistic perceptions out to be a positive characteristic). Therefore, a longer measure will further inform this debate within carceral settings.
Additionally, although both studies sample characteristics closely resemble their respective state demographics, the findings here may not be generalizable to other prisons in other state populations due to established legislative variance. Furthermore, these studies assessed the attitudinal evaluations of incarcerated individuals, retrospective self-report, and therefore the perceptions of correctional officers and procedural justice may be limited to these specific institutions. We suggest future research continue to test the validity and reliability of the PDOS. Finally, the Study 1 sample size was reduced due to missing data, future work should test the PDOS in a larger sample jail population.
Nevertheless, this study provides an empirical examination of psychometric properties of the Perceived Dehumanization from Officers Scale. The findings of this study suggest the PDOS is a sound measure of dehumanization, which neighbors but is distinct from procedural justice. This study contributes a measurement that provides potential for new empirical and theoretical work within the correctional setting that future studies may explore.
ACKNOWLEDGEMENTS
Funding Support: This work was in part supported by the Military Suicide Research Consortium (MSRC), an effort supported by the Department of Defense (Grant No. W81XWH-16-20003) and the National Institute of Mental Health (Grant No. 5T32MH093311-12). Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the Military Suicide Research Consortium, the Department of Defense, or the National Institute of Mental Health.
Footnotes
Previous Dissemination: These results have not been previously disseminated, the data are not available, but materials and code are accessible.
DATA AVAILABILITY STATEMENT:
Due to confidentiality of the state department of corrections ethics boards, the study analysis code for both studies is available, but study materials are not. This study was not preregistered.
References
- Abderhalden FP (2022). Environmental and psychological correlates of self-injurious thoughts and behaviors among jail detainees. Corrections: Policy, Practice and Research, 1–24. 10.1080/23774657.2021.2023339 [DOI] [Google Scholar]
- Alward LM, & Baker T (2019). Justice-involved males’ procedural justice perceptions of the police and courts: examining the spill-over effect. Criminal Justice Studies, 34(1), 33–47. 10.1080/1478601X.2019.1706876 [DOI] [Google Scholar]
- Alward LM, Baker T, & Gordon JA (2021). Procedural justice and incarcerated people’s obligation to obey institutional rules: An examination of current, former, and never-gang members. Journal of Criminal Justice, 73, 101757. 10.1016/j.jcrimjus.2020.101757 [DOI] [Google Scholar]
- Baker T, Abderhalden FP, Alward LM, & Bedard LE (2019). Exploring the association between procedural justice in jails and incarcerated people’s commitment to institutional rules. Corrections: Policy, Practice and Research, 6(3), 189–202. 10.1080/23774657.2019.1618224 [DOI] [Google Scholar]
- Baker T, & Gau JM (2018). Female offenders’ perceptions of police procedural justice and their obligation to obey the law. Crime and Delinquency, 64(6), 758–781. 10.1177/0011128717719418 [DOI] [Google Scholar]
- Baker T, Gordon JA, & Taxman FS (2015). A hierarchical analysis of correctional officers’ procedural justice judgments of correctional institutions: Examining the influence of transformational leadership. Justice Quarterly, 32(6), 1037–1063. 10.1080/07418825.2013.877517 [DOI] [Google Scholar]
- Bandura A, Underwood B, & Fromson ME (1975). Disinhibition of aggression through diffusion of responsibility and dehumanization of victims. Journal of Research in Personality, 9(4), 253–269. 10.1016/0092-6566(75)90001-X [DOI] [Google Scholar]
- Brunton-Smith I, & McCarthy DJ (2016). Prison legitimacy and procedural fairness: A multilevel examination of prisoners in England and Wales. Justice Quarterly, 33(6), 1029–1054. 10.1080/07418825.2015.1023215 [DOI] [Google Scholar]
- Deska JC, Almaraz SM, & Hugenberg K (2020). Dehumanizing prisoners: Remaining sentence duration predicts the ascription of mind to prisoners. Personality and Social Psychology Bulletin, 46(11), 1614–1627. 10.1177/0146167220911496 [DOI] [PubMed] [Google Scholar]
- Deska JC, Lloyd EP, & Hugenberg K (2018). Facing humanness: Facial width-to-height ratio predicts ascriptions of humanity. Journal of Personality and Social Psychology, 114(1), 75–94. 10.1037/PSPI0000110 [DOI] [PubMed] [Google Scholar]
- Fox KA, Lane J, & Akers RL (2010). Understanding gang membership and crime victimization among jail inmates: Testing the effects of self-control. 10.1177/0011128710392003, 59(5), 764–787. 10.1177/0011128710392003 [DOI] [Google Scholar]
- Gau JM (2015). Procedural justice, police legitimacy, and legal cynicism: a test for mediation effects. Police Practice and Research, 16(5), 402–415. 10.1080/15614263.2014.927766 [DOI] [Google Scholar]
- Goffman E (1961). Embarrassment and social organization. In Personality and Social Systems. (pp. 541–548). Wiley Inc. 10.1037/11302-050 [DOI] [Google Scholar]
- Gordon J, & Baker T (2017). Examining correctional officers’ fear of victimization by inmates: The influence of fear facilitators and fear inhibitors. Criminal Justice Policy Review, 28(5), 462–487. 10.1177/0887403415589630 [DOI] [Google Scholar]
- Gorsuch RL (1983). Factor analysis (2nd ed.). Psychology Press. [Google Scholar]
- Guenther L (2012). Beyond dehumanization: A post-humanist critique of solitary confinement. Journal for Critical Animal Studies, 10(2), 1948–352. [Google Scholar]
- Haslam N (2006). Dehumanization: An integrative review. Personality and Social Psychology Review, 10(3), 252–264. 10.1207/s15327957pspr1003_4 [DOI] [PubMed] [Google Scholar]
- Hewitt JD, Poole ED, & Regoli RM (1984). Self-reported and observed rule-breaking in prison: A look at disciplinary response. Justice Quarterly, 1(3), 438–447. 10.1080/07418828400088241 [DOI] [Google Scholar]
- Higgins EM, Smith J, & Swartz K (2022). “We keep the nightmares in their cages”: Correctional culture, identity, and the warped badge of honor*. Criminology, 60(3), 429–454. 10.1111/1745-9125.12306 [DOI] [Google Scholar]
- Kelman HG (1973). Violence without moral restraint: Reflections on the dehumanization of victims and victimizers. Journal of Social Issues, 29(4), 25–61. 10.1111/J.1540-4560.1973.TB00102.X [DOI] [Google Scholar]
- Kelso JC (2014). Corrections and sentencing reform: The obstacle posed by dehumanization. McGeorge Law Review, 46, 897–906. [Google Scholar]
- Mazerolle L, Bennett S, Davis J, Sargeant E, & Manning M (2013). Procedural justice and police legitimacy: A systematic review of the research evidence. Journal of Experimental Criminology, 9(3), 245–274. 10.1007/S11292-013-9175-2 [DOI] [Google Scholar]
- Mitchell MM, McCullough K, Wu J, Pyrooz DC, & Decker SH (2020). Survey research with gang and non-gang members in prison: operational lessons from the LoneStar Project. Trends in Organized Crime, 1–29. 10.1007/S12117-018-9331-1 [DOI] [Google Scholar]
- Reimann M, & Zimbardo PG (2011). The dark side of social encounters: Prospects for a neuroscience of human evil. Journal of Neuroscience, Psychology, and Economics, 4(3), 174–180. 10.1037/A0024654 [DOI] [Google Scholar]
- Reisig MD, & Mesko G (2009). Procedural justice, legitimacy, and prisoner misconduct. Psychology, Crime, & Law, 15(1), 41–59. 10.1080/10683160802089768 [DOI] [Google Scholar]
- Ribeiro JD, Bender TW, Selby EA, Hames JL, & Joiner TE (2011). Development and validation of a brief self-report measure of agitation: the Brief Agitation Measure. Journal of Personality Assessment, 93(6), 597–604. 10.1080/00223891.2011.608758 [DOI] [PubMed] [Google Scholar]
- Ryan C, & Bergin M (2022). Procedural Justice and Legitimacy in Prisons: A Review of Extant Empirical Literature. Criminal Justice and Behavior, 49(2), 143–163. 10.1177/00938548211053367 [DOI] [Google Scholar]
- Stanley IH, Hom MA, Christensen K, Keane TM, Marx BP, & Björgvinsson T (2021). Psychometric properties of the Depressive Symptom Index-Suicidality Subscale (DSI-SS) in an adult psychiatric sample. Psychological Assessment, 33(10), 987–997. 10.1037/PAS0001043 [DOI] [PubMed] [Google Scholar]
- Stinglhamber F, Nguyen N, Josse M, & Demoulin S (2022). The development of prison officers’ job satisfaction and its impact on depersonalization of incarcerated persons: The role of organizational dehumanization. Criminal Justice and Behavior, 49(11), 1600–1617. 10.1177/00938548221087182 [DOI] [Google Scholar]
- Sykes GM (1958). The Society of Captives. Princeton University Press. [Google Scholar]
- Trotter RT, Lininger MR, Camplain R, Fofanov VY, Camplain C, & Baldwin JA (2018). A survey of health disparities, social determinants of health, and converging morbidities in a county jail: A cultural-ecological assessment of health conditions in jail populations. International Journal of Environmental Research and Public Health, 15(11). 10.3390/IJERPH15112500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler TR (2006a). Psychological perspectives on legitimacy and legitimation. Annu. Rev. Psychol, 57, 375–400. 10.1146/annurev.psych.57.102904.190038 [DOI] [PubMed] [Google Scholar]
- Tyler TR (2006b). Why People Obey the Law. Why People Obey the Law. 10.1515/9781400828609/HTML [DOI] [Google Scholar]
- Vieraitis LM, Medrano J, & Shuraydi A (2018). “‘That’s a damn good officer any day of the week’: inmates’ perceptions of correctional officers”. Criminal Justice Studies, 31(2), 143–159. 10.1080/1478601X.2017.1420544 [DOI] [Google Scholar]
- Zimbardo PG (1969). The human choice: Individuation, reason, and order versus deindividuation, impulse, and chaos. Nebraska Symposium on Motivation. https://psycnet.apa.org/record/1971-08069-001 [Google Scholar]
- Zimbardo PG, Maslach C, & Haney C (1999). Reflections on the Stanford Prison Experiment: Genesis, transformations, consequences. Obedience to Authority, 207–252. 10.4324/9781410602022-15 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Due to confidentiality of the state department of corrections ethics boards, the study analysis code for both studies is available, but study materials are not. This study was not preregistered.