The Association of Ethnic Pride With Health and Social Outcomes Among Young Black and Latino Men After Release From Jail

Abstract

The goal of this study was to understand whether ethnic pride among young, incarcerated Black and Latino men was associated with successful community reentry. We interviewed 397 Black and Latino men 16 to 18 years old in a New York City jail and then again 1 year after their release to determine the relationship between participants' sense of ethnic pride during incarceration, and substance use, violence, recidivism, and education/ employment after release from jail. Participants with higher ethnic pride scores were less likely to engage in illegal activities and be reincarcerated. Ethnic pride was also associated with feeling safe in gangs and positive attitudes toward avoiding violence in situations of conflict. Ethnic pride was not associated with substance use, education, or engagement in community-based organizations post release. This study demonstrated that ethnic pride might be a source of strength that young men of color can harness for successful community reentry after release from jail.

Introduction

The United States has the highest rate of juvenile incarceration in the developed world, with stark gender, race, and class disparities when it comes to who gets incarcerated. More than two million juveniles are arrested every year. In 2006, 85% of all incarcerated juveniles were male (Child Trends, 2010); 70% of these young offenders were males of color (Ramaswamy & Freudenberg, 2012), even though men of color make up only 22% of the total U.S. population. In 2006, the detention rates were more than 5 times higher among Black youth and 2 times higher among Latino youth when compared with White individuals of the same age group (Hartney & Vuong, 2009), and the majority of those young Black and Latino men came from low-income families (Daniels, Crum, Ramaswamy, & Freudenberg, 2009).

The majority of the youth who spend time in criminal justice facilities in the United States are confined for non-serious crimes. Only 26% of all incarcerated youth were convicted for crimes defined as violent index offenses— murder, rape, robbery, or aggravated assault. Using incarceration as a method of punishment for minor offenses is inappropriate and ineffective, as it turns these young men into repeat offenders who get stuck in a vicious cycle of involvement with the criminal justice system, with recidivism rates as high as 55% (Daniels et al., 2009; U.S. Department of Justice, 2006).

The incarceration of juveniles in the United States and related criminal justice policies and practices are associated with a number of negative social and health consequences, including unemployment, discrimination, and high school dropout (Daniels et al., 2009). One study showed that having a criminal record reduced the likelihood of being called back for a job by 50% (Pager, 2003). Minority status and a history of incarceration only intensify the stigmatization of these individuals. For example, the effect of incarceration on future employment prospects for Black offenders is up to 40% worse than White offenders (Pager, 2003). Incarceration has a negative impact on offenders' education as well because young people miss school while they spend time in jail. These days are counted as unexcused absences and school administrators often use this as an excuse to expel adjudicated youth (Nellis, 2011). Unemployment and dropping out of school perpetuate likely run-ins with the criminal justice system (Ramaswamy & Freudenberg, 2012). Without education and employment, ex-offenders often return to criminal activity to make a living.

Incarcerated individuals also have higher rates of infectious diseases, chronic illnesses, mental health, and drug problems compared with people in the general population (Daniels et al., 2009; Golzari, Hunt, & Anoshiravani, 2006; Joseph-DiCaprio et al., 2000). Many young people have poorer health status even before they enter the criminal justice system; however, the incarceration event itself can also increase the risk of developing disease or poor health, given the prevalence of individuals with infectious disease in jail, combined with engagement in risky behavior within the facility (Romero et al., 2007). Juvenile offenders might also experience physical or psychological harm in the form of physical and sexual assaults (from staff or from other offenders), and excessive use of isolation and restraints (Mendel, 2011). When young people leave jails—the majority of whom will do so—they often return to poor, disadvantaged communities void of health care infrastructure to begin with. Young people leaving jails are often too old to receive government medical benefits (Ramaswamy & Freudenberg, 2012) and may experience difficulty gaining access to medical care in the community to treat existing health conditions (Golzari et al., 2006).

So what are the resources that can help young men to overcome the negative social and health consequences of incarceration? Beyond school, employment, and connections to health care, for example, this study sought to investigate other personal resources that the young men could draw on to improve their reentry experiences. We have argued here that minority status and race can be a disadvantage for young men who get picked up by the justice system. While research across a number of disciplines focuses on the negative aspects of race and criminal justice involvement, some research within public health acknowledges that an individual's racial and ethnic pride may also be a resource to help cope with the challenges of social inequality. Young men of color with criminal justice histories often have so little to draw on in terms of resources that could facilitate success and avoidance of ongoing criminal justice involvement that we sought to explore one source of resilience that young men could potentially harness over the life course: the pride associated with their ethnic identity.

Ethnic identity refers to the relationship between racial/ethnic minority individuals and their broader racial/ethnic group. This psychological attachment is based on shared cultural background and history. Researchers, including J. S. Phinney, define ethnic identity in terms of positive feelings toward the individual's group. According to Phinney's theory, ethnic identity is a dynamic and multidimensional concept characterized by a person's (a) sense of affirmation and belonging and (b) identity achievement, and engagement in ethnic behaviors (Roberts et al., 1999). In this study, we were primarily concerned with identifying how positive attitudes toward an ethnic group can affect outcomes among our target population of incarcerated young men. In understanding this relationship, behavioral interventions that build on racial and ethnic pride may serve as novel, easy-to-implement, and cost-effective interventions to improve outcomes for incarcerated young people.

Ethnic identity can be a central feature of identity development, especially for young people from some racial or ethnic backgrounds (Phinney & Alipuria, 1990). Racial or ethnic minority adolescents report that it is important for them to learn about their ethnicity, culture, and traditions because these are important aspects of self-definition. Learning about their ethnicity, culture, traditions, and achieving a stable sense of self is an accomplishment associated with increased self-esteem and improved psychological well-being (Phinney & Chavira, 1992; Umana-Taylor, Gonzales-Backen, & Guimond, 2009). Ethnic pride, in particular, is positively associated with reduced mental illness. Pride is a coping mechanism that can reduce the depressive symptoms caused by racial discrimination (Mandara, Gaylord-Harden, Richards, & Ragsdale, 2009; Mossakowski, 2003). The feeling of belonging that comes from identifying oneself as part of a larger group acts as a buffer against the feeling of isolation that results from discrimination (Mossakowski, 2003).

Higher levels of ethnic pride are also associated with more concrete results, such as positive outcomes in education. Miller (1999) showed that racial identity promotes resilience, particularly in Black adolescent males. Academic achievement is an indicator of resilience, and children with a strong sense of racial/ethnic pride (positive perceptions of their own group) have better academic outcomes (higher grade point averages; Altschul, Oyserman, & Bybee, 2006; Miller, 1999; Smith, Atkins, & Connell, 2003).

Ethnic pride is also associated with reduced substance use and other asocial behaviors. American Indian adolescents with strong ethnic pride had better grades and stronger anti-drug norms than those with lower levels of pride (Kulis, Napoli, & Marsiglia, 2001). Similar trends were seen in a group of pre-adolescent Mexican Americans in Southwest United States (Marsiglia, Kulis, Hecht, & Sills, 2004).

A higher sense of ethnic pride is also associated with fewer problematic externalized behaviors, such as delinquency, aggression, and anti-social behavior (Shrake & Rhee, 2004). One study showed that a violence intervention program focused on improving ethnic pride in a group of Hispanic American teens saw a significant decrease in violence as the ethnic pride levels increased (Enriquez, Kelly, Cheng, Hunter, & Mendez, 2012). Last, a study examining Hawaiian teens showed that those who reported higher levels of ethnic pride experienced and witnessed fewer incidents of violence in their lives (Austin, 2004).

While racial and ethnic pride are important factors that shape the experiences of young people of color, there is limited information on the role of ethnic pride in the lives of young men who become incarcerated, a disproportionate number of whom are racial/ethnic minorities. Ethnic pride may be one of the few personal resources that poor, young men of color can draw on in resisting the stigma associated with their criminal justice histories and marginalization based on race, class, and gender. Therefore, the purpose of this study was to determine if there was an association between ethnic pride and positive outcomes for young men over time, in this case, after release from jail. We hypothesized that ethnic pride would have the same benefits for our youth as it does in the general population: (a) reduced engagement in illegal activity and lower rates of recidivism for this sample, (b) positive attitudes toward avoiding violence and gangs, (c) lower rates of substance dependence and daily marijuana use, and (d) increased enrollment in schools or employment programs.

To test these hypotheses, we analyzed data that were originally collected as part of a substance use, reincarceration, and HIV risk reduction intervention, the Returning Educated African-American and Latino Men to Enriched Neighborhoods (REAL MEN) study. Ethnic pride could serve as a personal resource that young men could use to protect against negative social and health outcomes during the transition from jail to community. If we are able to demonstrate that ethnic pride can indeed be a resource, future intervention efforts might build on the relationship between ethnic pride, social, and health outcomes.

Method

Participants

This analysis was based on data from the REAL MEN intervention study, conducted between 2004 and 2007, where we interviewed 552 young men ranging from 16 to 18 years of age at the New York City Department of Corrections' Rikers Island Detention Center (Daniels et al., 2009; Freudenberg et al., 2010; Valera, Epperson, Daniels, Ramaswamy, & Freudenberg, 2009). We conducted follow-up interviews with 72% of the sample, 397 young men, 1 year after their release from jail. All participants completed an informed consent process approved by the Hunter College and New York City Department of Health and Mental Hygiene Institutional Review Board prior to enrolling in the study.

REAL MEN Intervention

REAL MEN was a 30-hour intervention program that began in jail and ended in the community (Daniels et al., 2009; Freudenberg et al., 2010; Valera et al., 2009). Participants were randomized to either receive some workshops in the jail and a regular jail discharge plan (control) or the 30-hour REAL MEN intervention, which included workshops in jail, a discharge plan, and continued workshops and case management in the community at a non-profit, partner organization. The intervention group received a series of six workshops designed to address substance use, reincarceration risk, and HIV risk in ways that were relevant to the young men most likely to be incarcerated in New York City. Workshops addressed topics including masculinity, relationships, and the risk of HIV/sexually transmitted infections (STIs); economic class, drug involvement and incarceration; and the impact of civil rights leaders on the lived experience of race, racism, and racial or ethnic pride. All the workshops on racial and ethnic pride were designed to help the young men identify their strengths, protect their health, and reduce future criminal justice involvement (Daniels et al., 2009; Freudenberg et al., 2010; Valera et al., 2009). The specific details of the REAL MEN program, including participant enrollment, randomization, and main outcomes of the intervention, have been described elsewhere (Daniels et al., 2009; Freudenberg et al., 2010).

Data Collection

Survey data were collected at the time of incarceration (Time 1) and 1 year after release from jail (Time 2) by project staff. At Time 1, the mean age of participants was 17.99 (SD = 0.71). One year after release from jail, the mean age was 19.60 (SD = 0.93), suggesting that on average, the young men spent 7 months in jail before release. Five hundred fifty-two young men completed the survey at Time 1, while 397 of these young men returned to complete the Time 2 questionnaire (72% follow up). Through these questionnaires, the team collected Time 1 and Time 2 information relating to participants' social support networks, knowledge of HIV, health and psychological history, substance use and treatment history, sexual risk behavior, criminal justice history, experience with violence and relationships, and ethnic pride.

Variables

Socio-demographic factors

Participants' socio-demographic information was collected at both Time 1 and Time 2, but only the Time 1 factors were used as independent variables in this study because we were interested in young men's baseline levels of ethnic pride before the experience of incarceration and release. We recorded the young men's age, race, employment status, and whether they had earned their high school diploma. We also assessed their living situation (whether they lived with family or not) and their history of homelessness. The purpose of the study was to determine whether a higher baseline (Time 1) ethnic pride level was associated with positive outcomes after release from jail (Time 2). Previous studies have shown that socio-demographic factors affect the outcomes measured in this study (Townsend & Smith, 2010), so we used those Time 1 factors as controls in our analysis to make sure these factors did not confound the effects attributed to Time 1 levels of ethnic pride.

Multi-Group Ethnic Identity Measure (MEIM): Main explanatory variable

The MEIM was included in the REAL MEN survey at Time 1 and administered orally by project staff. This scale was designed to assess ethnic identity by measuring self-identity/self-labeling, affirmation and belonging, ethnic identity achievement, and ethnic behaviors. It was first developed by J. S. Phinney, and it is the only measure that examines ethnic identity as a general phenomenon that can be reliably used across different ethnic groups. It should be noted that this questionnaire is an ethnic identity measure. However, we justified using it as a measure of ethnic pride in our study because Phinney identified “purely positive feelings and attitudes towards one's ethnic group” as part of ethnic identity (Valk & Karu, 2001, p. 586). Second, our intervention focused specifically on exploring and building ethnic pride, and in designing the study, investigators decided that the MEIM best captured elements of ethnic pride and was able to measure the same. Each question in this 12-item questionnaire was rated on a 4-point scale from 1 (strongly disagree) to 4 (strongly agree). The overall scores for each item, factor, and complete questionnaire can be determined by calculating the mean of scale scores. A higher score on the MEIM corresponds to strong ethnic identity, and hence a high sense of ethnic pride (Phinney, 2004; Roberts et al., 1999). The MEIM can be divided into two factors that can be analyzed separately to provide more detailed information about (a) participants' sense of belonging and positive feelings toward their ethnic group and (b) ethnic identity exploration and a sense of achievement.

The MEIM was designed for adolescents and young adults from diverse groups, and therefore, it is well suited for our study (our participants were mostly young Black and Latino men between 16 and 18 years of age). It has been used in a number of projects involving a wide range of ethnic groups and has been shown to be reliable—studies reported a coefficient alpha score of .81 with high school students and .90 with college students (Phinney, 2004; Roberts et al., 1999). In this study, we measured our respondents' baseline level of ethnic pride by assessing their sense of belonging in their ethnic group, as well as their commitment to this group before incarceration.

Dependent variables

Eight dependent variables were selected because of their association with successful community reentry. These outcomes fell under four domains: (a) Criminal justice involvement, (b) Attitudes toward violence, (c) Substance use, and (d) Constructive outcomes. Table 1 includes the names, definitions, and descriptive statistics for the eight dependent variables.

Table 1.

Descriptions of Outcomes After Release From Jail.

Outcome	Definition	Number (%)
Criminal justice involvement
1. Reincarceration	Reincarcerated in the past 12 months	184 (46.9)
2. Engaged in illegal activities	Never (0 times)	127 (32.5)
	Rarely (1-9 times)	98 (25.1)
	Sometimes (10-25 times)	50 (12.8)
	Often (26-49 times)	27 (6.91)
	Very often (>50 times)	89 (22.8)
Attitudes toward violence
1. Avoiding violencea	5-item questionnaire used to measure attitudes toward violence and its acceptability in situations of conflict	Mean = 2.691
2. Feeling safe in a gang	The idea that belonging to a gang makes kids feel safe	217 (55.5)
Substance use
1. Substance dependence	Dependent on drugs or alcohol in the past year, based on DSM-IV criteria	66 (16.6)
2. Daily marijuana use	Smoked marijuana on a daily basis in the past 90 days	121 (31.0)
Constructive outcomes
1. Enrolled in educational programs	Currently enrolled in and attending, or graduated from, educational or vocational program in the past 12 months	226 (57.4)
2. Went to community-based organization	Participated in community-based organizations after release from jail	137 (34.5)

Note. DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994).

^aAttitudes toward violence: 1 to 4 where 1 = strongly agreed with using violence to stop conflict and 4 = strongly disagreed with using violence to stop conflict.

Statistical Analyses

Prior to initiating the outcome analyses, we studied the baseline characteristics of all the variables in the data. We also examined the normality of the continuous variables (ethnic pride score and attitudes toward violence) and found that the normality assumption did not hold for these variables. Then we carried out non-parametric tests (Hollander & Wolfe, 1999) to study whether ethnic pride scores were significantly different across the various levels of each post-release outcome. We used the Mann–Whitney test to examine whether ethnic pride scores were significantly different across the two levels of dichotomous outcome variables (drug dependence, daily marijuana use, attitudes toward gangs, reincarceration, involvement in educational programs, and engagement in community-based organizations), and also the Kruskal–Wallis test to examine whether the ethnic pride scores were different across more than two levels of the variable “engagement in illegal activity” that had more than two response categories. Last, we used the Spearman's rank correlation test to assess the relationship between ethnic pride and the continuous outcome variable, attitudes toward violence.

With a bivariate analysis, we were able to determine the significant differences in ethnic pride scores with different categories of response for some dependent variables. To learn more about the nature of this relationship, and to quantify the impact of various simultaneous influences on each dependent variable, we then carried out multiple logistic regression analysis (Agresti, 2002) for each of the categorical dependent variables. We fitted separate models for each dependent variable (engagement in problem behaviors, reincarceration, attitudes toward gangs, substance dependence, daily marijuana use, enrollment in educational programs, and engagement in community-based organizations), using the ethnic pride score as a covariate and adjusting for age, race, education and employment history, family background, and homelessness. We also conducted linear regression analysis for the continuous dependent variable, attitudes toward violence. Because the distribution of the continuous response variable, avoidance of violence, was non-normal, we transformed (square-rooted) the data before fitting the linear regression model. Although the goal of the present study was not to evaluate the REAL MEN intervention, when analyzing data here, we accounted for the effect of the program (by controlling for it in our analysis) as it explicitly engaged the young men in discussions about the relationship between gender, race, class, and health (Schulz, Freudenberg, & Daniels, 2006). We also examined the presence of an interaction effect between the ethnic pride score and intervention for all the models.

Results

Of the 552 young men who were recruited at the beginning of the study, 397 individuals completed the follow-up interview. The present analysis included only individuals who completed both interviews (N = 397). All participants in the study were men with a mean age (at intake) of 17.47 (range = 16-18 years; Table 2). The majority of the individuals were either Black (55.9%) or Latino (38.1%). Most participants (93.2%) reported that they did not have a high school diploma, and 77.3% reported that they were unemployed before incarceration. More than 21% were not living with family prior to incarceration and at least 8.3% of the participants reported a history of homelessness.

Table 2.

Description of Socio-Demographic Factors Prior to Incarceration.

Factors	Number (%)
Age	Mean = 17.47
16	43 (10.8)
17	141 (35.5)
18	213 (53.7)
Race
Black	221 (55.8)
Latino	151 (38.1)
Other	16 (4.03)
Biracial	4 (1.01)
White	4 (1.01)
Received high school diploma (or equivalent)	27 (6.80)
Employed full-time, part-time, or on and off	90 (22.6)
Living with family	312 (78.6)
Homeless	33 (8.31)

Ethnic Pride Scores

Ethic pride was the main predictor variable in the analysis. The average of the 12 categorical responses on ethnic pride for this population was 2.924. To give context, the average ethnic pride scores among Blacks in one study of a high school and college-based sample were 3.04 and 3.46, respectively. For Latinos, the scores were 2.91 for high schoolers and 3.07 for college students (Phinney, 1992). We also investigated the association of the two factors of the ethnic pride scores (Roberts et al., 1999): (a) exploration (ethnic identity search) and (b) affirmation (sense of belonging and positive feelings) with the outcomes of interest. The average score for the exploration factor was 2.678 and the average for the affirmation factor was 3.100.

Criminal Justice Involvement and Engagement in Illegal Activity

One hundred and eighty-four individuals (46.9%) who participated in REAL MEN returned to jail for some amount of time in the year following their release. Up to 68% of the population (264 individuals) had participated in illegal activities in the year after their incarceration—22.8% reported that they were involved in activities that they knew would get them in legal trouble if caught “very often” or “more than 50 times in the past year.”

Attitudes Toward Violence and Gangs

When we assessed participants' attitudes toward violence after release from jail, we found that the average score was 2.691 on a 4-point scale (1 = strongly agreed with using hitting/fighting to stop conflict, and 4 = strongly disagreed with using violence to stop conflict). More than half of this group (55.5%) reported that belonging to a gang made them feel safe.

Substance Use

We found that up to 16.6% of the participants could be characterized as drug or alcohol dependent in the year after release from jail. Almost one third of the group (31.0%) reported daily marijuana use.

Constructive Outcomes

Last, when assessing constructive outcomes, we found that 57.4% of the participants were enrolled in and attending educational/vocational programs or had graduated from a program between Time 1 and Time 2. One hundred and thirty-seven participants (34.5%) went to a community-based organization after release from jail for reentry support and 75.2% (N = 103) of this subset attended the community-based organization for more than 30 days.

Association of Ethnic Pride With Outcomes

Our non-parametric analyses showed that the ethnic pride scores were significantly different across the levels of outcome variables: reincarceration, engagement in illegal activity, feeling safe in gangs, and avoiding violence. As these methods do not adjust for possible confounding covariates, we carried out multiple logistic regression for all outcome variables, except attitudes toward violence, which was continuous. For this variable, we carried out multiple linear regression.

Columns 2 to 8 of Table 3 represent the results of the multiple logistic regression analyses and column 9 represents the multiple linear regression analysis results. Both of the regression analyses were adjusted for the following socio-demographic factors: age, race, family status, homelessness, educational attainment, employment status, and intervention status. We found that engagement in high levels of illegal activities (50 times in 1 year or higher), reincarceration, feeling safe in gangs, and avoiding violence in situations of conflict were associated with ethnic pride. For every 1-unit increase in ethnic pride, (a) the odds of engaging in high levels of illegal activity (50 times in a year or higher) decreased by 65% (95% confidence interval [CI] for the reduction was between 17% to 85%), (b) the odds of getting reincarcerated decreased by 55% (95% CI for the reduction was 15% to 76%), and (c) the odds of participants reporting that gangs made them feel safe increased by 112% (95% CI for the increase was 15% to 301%). The multiple linear regression analysis showed that ethnic pride and attitudes toward avoiding violence in situations of conflict were positively associated. In other words, individuals with higher levels of ethnic pride tended to favor avoiding violence. For a one-unit increase in ethnic pride score, the participant's score on attitudes toward avoiding violence (in square root scale) increased by 0.04 (95% CI = [0.003, 0.07]). However, the statistical significance of the relationship was weak (as shown by the lower limit of the CI, which is close to 0).

Table 3.

Multiple Regression Results.

	Categorical outcome variables							Continuous outcome variable
	Criminal justice involvement		Attitudes toward violence	Substance use		Constructive outcomes		Attitudes toward violence
	Reincarceration OR (95% CI)	Engaged in illegal activities (very often) OR (95% CI)	Feeling safe in a gang OR (95% CI)	Substance dependence OR (95% CI)	Daily marijuana use OR (95% CI)	Enrolled in educational programs OR (95% CI)	Engaged in community- based organizations OR (95% CI)	Avoiding violence Coefficient (95% CI)
Primary predictor
Ethnic pride	0.45** [0.24, 0.85]	0.35* [0.15, 0.83]	2.12* [1.15,4.01]	0.69 [0.30, 1.55]	0.56 [0.28, 1.09]	1.31 [0.71, 2.47]	1.02 [0.51, 1.98]	0.04* [0, 0.07]
Mediating variables
Age
17	1.50 [0.73, 3.11]	1.57 [0.54, 4.55]	1.25 [0.61, 2.57]	1.68 [0.58, 6.14]	2.44 [0.99, 6.92]	0.48 [0.20, 1.06]	1.26 [0.58, 2.79]	0.03 [−0.01, 0.07]
18	1.12 [0.55, 2.30]	2.44 [0.86, 6.93]	1.35 [0.66, 2.74]	2.67 [0.97, 9.52]	3.41** [1.43, 9.54]	0.26** [0.11, 0.56]	0.89 [0.42, 1.96]	0.05* [0.01, 0.08]
Race
Black	0.53 [0.21, 1.30]	0.27* [0.08, 0.92]	1.06 [0.43, 2.57]	1.59 [0.50, 7.13]	1.28 [0.49, 3.81]	0.50 [0.17, 1.31]	1.90 [0.68, 6.25]	0.04 [−0.002, 0.09]
Hispanic	0.72 [0.28, 1.81]	0.42 [0.12, 1.46]	0.66 [0.26, 1.62]	1.30 [0.39, 5.91]	1.39 [0.52, 4.18]	0.31* [0.11,0.82]	1.90 [0.66, 6.32]	0.02 [−0.02, 0.07]
High school diploma	0.32* [0.1 1,0.81]	0.55 [0.17, 1.78]	1.07 [0.46, 2.61]	0.70 [0.16, 2.21]	0.38 [0.11, 1.05]	1.09 [0.45, 2.72]	0.24* [0.07, 0.68]	−0.01 [−0.06,0.03]
Employment	0.92 [0.55, 1.52]	0.70 [0.35, 1.42]	0.81 [0.49, 1.34]	0.95 [0.48, 1.80]	1.05 [0.61, 1.78]	1.20 [0.72, 2.02]	1.03 [0.58, 1.79]	−0.01 [−0.04, 0.02]
Living with family	0.71 [0.42, 1.20]	0.57 [0.28, 1.18]	0.96 [0.57, 1.61]	0.71 [0.37, 1.39]	0.80 [0.47, 1.40]	1.88* [1.12, 3.20]	1.27 [0.72, 2.28]	−0.02 [−0.04, 0.01]
Homelessness	1.17 [0.53, 2.57]	0.38 [0.10, 1.41]	1.25 [0.58, 2.79]	0.55 [0.15, 1.52]	1.14 [0.50, 2.50]	1.78 [0.81, 4.09]	1.11 [0.46, 2.53]	0.001 [−0.04, 0.04]

Note. For attitudes toward violence, we fitted linear regression models and presented the coefficients. For all the other variables, we ran logistic regression models and presented ORs. For the engagement in illegal activity outcome, we only presented analysis for the “very often (>50 times)” category because only this category showed statistically significant outcomes. Boldfaced odds ratios indicate statistically significant associations. OR = odds ratio.

^*p < .05.

^**p < .01.

We did not find any significant association between either of the following dependent variables: substance use (drug dependence and daily marijuana use) and constructive outcomes (enrollment in educational programs and participation in community-based organizations). We also investigated if the outcome variables were influenced by the interactive effect of ethnic pride and the intervention by including an interaction term in the model. Such interaction effects were significant in none of the models.

We also carried out multiple regression analysis using the two factors of ethnic pride instead of using global (single factor) ethnic pride score. Factor 2 (sense of affirmation, commitment, and belonging) was strongly associated with feeling safe in gangs. For every one-unit increase in their Factor 2 score, the odds of participants' reporting that gangs make them feel safe increased by 123.9% (95% CI for the increase was 27% to 302%). We did not find any significant association between Factor 1 and Factor 2 for any of the other variables.

Discussion

In this study, we found that baseline ethnic pride was associated with the following community reentry outcomes: (a) decrease in likelihood of engaging in illegal activities, (b) decrease in likelihood of reincarceration, (c) higher reporting of feeling safe in gangs, and (d) higher reporting of avoiding violence in situations of conflict. Substance use (drug dependence and daily marijuana use) and positive outcomes (engagement in educational programming or community-based organizations) were not significantly associated with ethnic pride.

As noted earlier, our results showed that a strong sense of ethnic pride was associated with decreased engagement in illegal activity after release from jail and subsequent reincarceration. This was consistent with previous findings (Shrake & Rhee, 2004). One explanation for this result was that low self-esteem might lead to psychological problems that can manifest as aggression and delinquency (Donnellan, Trzesniewski, Robins, Moffitt, & Caspi, 2005). Ethnic pride can protect against these risk factors by promoting high self-esteem (Umana-Taylor et al., 2009). The decreased rates of reincarceration can be explained by the trend seen with illegal activities. If individuals with high levels of ethnic pride are less likely to engage in illegal activity after release from jail, then they are also less likely to return to jail.

We also found that high overall ethnic pride and an individual's sense of affirmation and group-belonging were both strongly associated with attitudes about gangs, specifically the feeling of safety if a person belonged to a gang. These findings were consistent with theories relating to gangs and collective identity (Brotherton & Barrios, 2004; Feixa, 2005). Gangs often incorporate cultural pride into the group's identity. They serve as a social space in which members' heritage is explicitly recognized and becomes part of gang rituals (Brotherton & Barrios, 2004). In addition, the gangs' collective identity reflects individual member's process of exploration and affirmation (Brotherton & Barrios, 2004). There is some evidence that gang membership can also make members feel safe in everyday situations when members can look out for each other, in the absence of economic and other types of social capital that might protect young people (Conquergood, 1994). The counter argument, however, is that gangs can also serve as a space for the legitimization of violence (Yablonsky, 1997), which was not exactly the domain we measured, yet could potentially be related to ongoing street organization membership.

Our last finding regarding attitudes toward violence in situations of conflict was consistent with previous studies (Shrake & Rhee, 2004; Soriano, Rivera, Williams, Daley, & Reznik, 2004). These studies showed that ethnic pride was associated with a decrease in violent behaviors among young men. We found—though marginal in its significance—that as ethnic pride scores increased, so did attitudes favoring avoidance of violence in situations of conflict. It is unclear if our measure of attitudes would translate to behaviors, as measured in other studies (Austin, 2004; Enriquez et al., 2012), but our findings point to a similar trend and provide further support for the investment in behavioral interventions that build ethnic pride to reduce violence among marginalized groups (Enriquez et al., 2012).

We found that substance use and constructive outcomes after incarceration (educational involvement and participation in community-based organizations) were not associated with ethnic pride. This is inconsistent with results presented in previous studies, where researchers found that ethnic pride was associated with stronger anti-drug norms, less drug use, and better school performance (Altschul et al., 2006; Kulis et al., 2001; Marsiglia et al., 2004). Based on our contrary findings, we concluded that substance use and constructive outcomes in our sample were more strongly influenced by other factors than by young men's sense of ethnic pride. For example, in other studies with this sample, drug and alcohol use were related to a clustering of other risk behaviors (Valera et al., 2009), while engagement in constructive outcomes, like employment, were associated with post-jail housing stability (Ramaswamy & Freudenberg, 2012). These findings suggest that the pathways to success for young men leaving jails are complex and layered.

Strengths, Limitations, and Areas for Future Study

The primary strength of this study is its empirical contribution to critical race research and public health policy. First, this study showed the relationship between ethnic pride and factors that indicate successful community reentry after incarceration. This is the first article to our knowledge that shows how ethnic pride may be associated with outcomes after release from jail, which highlights important areas for intervention.

This study also had some limitations. First, this was a secondary data analysis, so the questions included in the analysis were limited to surveys and interviews conducted as part of an HIV risk reduction program. Second, we did not examine changes in ethnic pride over time and the factors that may have accounted for that change. The data in the study were self-reported and subject to recall bias. The sample for this study was recruited only from one jail with 95% of inmates who were Black and Latino, which may have affected generalizability of results. We also did not collect data about the heterogeneity of Latino participants, though we know that only 9% of the sample was foreign born, indicating that Latinos in the sample were not largely members of new immigrant groups.

Future studies might examine the ways in which ethnic heterogeneity among Latinos affects ethnic pride and longer-term health and social outcomes for these groups. Futures studies might also examine the ways in which the incarceration experience affects levels of ethnic pride. Does incarceration leave these young men feeling disconnected from society and lead to a decrease in ethnic pride, or does it result in increased ethnic pride as individuals pledge their allegiance to gangs or other organizations within correctional facilities? What impact do these changes have on health and social outcomes? Such questions could be answered using longitudinal and mixed-methods approaches. They might better inform the scholarship on race, gender, and criminal justice involvement, as well as public health intervention efforts.

Conclusion

We found that among a group of Black and Latino young men, ethnic pride as measured during incarceration, was associated with decreased rates of incarceration and engagement in illegal activity 1 year after release from jail. Higher reports of ethnic pride were also associated with the feeling of safety in gangs and endorsement of anti-violence attitudes after release. The most important and tangible variable here, in terms of reentry success for these marginalized young men, is recidivism. Each incarceration of a young man in New York City costs taxpayers an average of US$168,000 per year, in addition to significant social, economic, and emotional costs to the incarcerated individual (New York City Independent Budget Office, 2013; Pager, 2003). Furthermore, the cycle of incarceration, release, and reincarceration disrupts not only individuals' chances for success, but also family stability, and reduces community cohesion (Clear, 2007). If short-term, jail-based interventions can effectively address the protective factor of ethnic pride—building young men, rather than breaking them—individuals and society might stand to benefit.

Biographies

Satyasree Upadhyayula is an internal medicine resident at Saint Louis University Hospital. She authored this manuscript while she was in a joint MD/MPH program at University of Kansas School of Medicine.

Megha Ramaswamy is an Assistant Professor of Public Health and Preventive Medicine and University of Kansas School of Medicine. Her current work addresses the social context of sexual and reproductive health risk among women in the justice system.

Prabhakar Chalise is a Research Assistant Professor of Biostatistics at University of Kansas School of Medicine. His work includes statistical genetics and genomics, survival analysis, experimental design, clinical trials, multivariate analysis, and computational statistics.

Jessie Daniels is Professor of Public Health at The CUNY School of Public Health. Her research interests include racism, incarceration, digital media and health.

Nicholas Freudenberg is a Distinguished Professor of Public Health at The CUNY School of Public Health. He was the Principal Investigator of the REAL MEN study. His interests include public health policy, urban health, incarceration and health, and the health impact of corporate practices.