The Effects of Emotional Literacy Programs on Inmates’ Empathy,Mindfulness,and Self-Regulation: A Random-Effect Meta-Analysis

Abstract

Applying a random-effect meta-analysis, the current study examines previous empirical findings about three main purposes of Emotional Literacy (EL) correctional programs: improving empathy, mindfulness, and self-regulation among inmate populations. A total of 22 programs in 15 studies in Germany, Portugal, the Netherlands, and the US are analyzed. Results suggest that EL programs significantly improve levels of empathy, mindfulness, and self-regulation, although a possibility of publication bias is observed. Based on these findings, this study concludes that EL programs retain a potential as an effective rehabilitation strategy and should be sought and implemented in U.S. correctional settings.

Keywords

emotional literacy program empathy mindfulness self-regulation rehabilitation meta-analysis

Introduction

Mass incarceration has become increasingly controversial in the United States because it incurs not only massive prison expenditures but also facilitates far-reaching consequences that weaken whole communities (Clear et al., 2015). To address this controversial issue, correctional institutions have initiated various prison programs—counseling, cognitive-behavioral treatment, education, and recreation—that can help inmates adapt to prison life and develop opportunities for re-entry preparation (Clear et al. 2015). Some programs utilize psychological approaches designed to curb criminal behaviors and provide substance abuse counseling or anger management, while others involve education or vocational training for entry employment (Clear et al., 2015). At the same time, the nation’s inmate population increased consistently from the 1980s to the 2010s. According to the Bureau of Justice Statistics (2018), the total number of US federal inmates is estimated at 189,192, while the total number of inmates held in state facilities numbered 1.32 million at the end of 2016. Blevins et al. (2010) argue that a prison environment includes many adverse conditions such as overcrowding, boredom, victimizations, and episodic violence, all of which compromise effective correctional programming. Moreover, because correctional policy emphasizes the control and punishment of the offender, a commitment to rehabilitative objectives and treatment programs is not a priority, and, therefore, they are typically under-resourced (Pitts et al., 2014).

With mounting political and economic pressures stemming from mass incarceration and the recent gradual shift toward decarceration, the need to prepare inmates for reentry through meaningful correctional programs has become even more critical in the United States (Cochran et al., 2014). Emotional Literacy (EL) correctional programs arespotlighted here, as there is some empirical evidence that they can impact recidivism. According to Scott et al., (2015) study, previous research indicates that EL programs in the correctional systems of the United States, the United Kingdom, Germany, and the Netherlands produced positive outcomes. These EL programs are designed to decrease the likelihood of aggressive or other counterproductive social behaviors. The focus of this approach is to provide cognitive and emotional skills that develop and improve capacities for self-regulation, mindfulness, and empathy. These skillsets should better prepare offenders to avoid emotional triggers that lead to further criminality, helping offenders to create a less volatile prison atmosphere in general.

While numerous studies suggest the effectiveness of EL programs on recidivism, few studies are conducted to review EL programs systematically. Thus, the purpose of this study is to conduct a systematic review of EL programs and to identify promising formulations of this program approach. This article begins with an overview of EL programs that have been implemented in four countries, including Germany, Portugal, the Netherlands, and the US. A review of the relevant literature on EL programs is also presented, followed by a discussion of the methodology used for the current study. A total of 22 programs in 15 empirical studies are examined using a random-effects meta-analysis. Finally, we discuss the findings from our meta-analysis of the selected studies, along with policy implications for EL correctional programs in correctional settings.

Literature Review

The Prison Environment

A prison environment encompasses a variety of negative factors that include inadequate living spaces, a variety of deprivations, exposure to mental, physical and occasional sexual exploitations, tribalistic violence, and degrees of isolation (Blevins et al., 2010; Johnson, 1995; Sykes, 1958). All of these factors can contribute to inmate stress, fear, anxiety, and stress-related health problems such as depression, and aggressive behaviors (Blevins et al., 2010; Pitts et al., 2014). Overcrowding often leads to inmate overstimulation, which heightens arousal levels and reactive behaviors. When cortical arousal is heightened in tandem with reactive behaviors, incarcerated persons are more likely to act aggressively and violently (Blevins et al., 2010). More inmate-on-inmate and inmate-on-staff assaults, as well as correction officer-on-inmate violence, occur in overcrowded prisons (Blevins et al., 2010). Lack of resources also adds to volatility in correctional facilities that are over capacity. Specifically, prisons struggle to provide inmates with meaningful work or other rehabilitative programs to reduce stress and improve readiness for post-incarceration life (Pitts et al., 2014). Therefore, if a prison is overcrowded, lacks adequate correctional programming, and fails to provide constructive activities, success in inmates’ community reintegration is further jeopardized. Emotional Literacy programs may be a promising approach to support inmate adaptation to the day-to-day stressors of incarceration.

Overview of EL Programs

EL was first defined by Steiner and Perry (1997) as a combination of the abilities: (a) to understand the emotions of others; (b) to better understand one’s own emotions; (c) to have the ability to empathize with others and; (d) to express emotions more appropriately and efficiently. With that in mind, three major factors are commonly categorized in Emotional Literacy programs: (1) empathy, (2) mindfulness, and (3) self-regulation. According to Agnew’s (2009) General Strain Theory, inmates are confronted with negative stimuli in prison that can lead to negative emotions, causing them to act out aggressively and inappropriately (Blevins et al., 2010). Specifically, EL programs can teach participants how to act non-aggressively in challenging circumstances and how to manage emotions appropriately (Camilleri et al., 2012; Casarjian et al., 2007).

The primary objective of EL correctional programs is to provide offenders with self-awareness training and the ability to address areas that mask subconscious pains and other emotional and cognitive blind spots. Offenders who learn self-control of their emotions and better understand the emotions of others will respond more appropriately during incarceration (Casarjian et al., 2007; Ferszt et al., 2009). EL works by allowing an inmate to learn appropriate strategies to deal with negative emotions (Ferszt et al., 2009). Introducing inmates to new methods of processing and releasing negative emotions in a non-aggressive manner, particularly in a prison setting, is critically important and could lead to reductions in retaliatory violence.

The concept of Emotional Literacy as a program model is more familiar in the field of education. Resources for educators who teach Emotional Literacy to students focus upon a variety of areas such as self-awareness, emotional resilience, and self-regulation. According to our review, most EL programs address three central areas: (1) Self-regulation, (2) Mindfulness, and (3) Empathy.

Self-regulation is defined as an individual’s ability to control their own emotions (Knight & Modi, 2014). When an inmate has increased levels of self-regulation, they are more likely to engage in prosocial behaviors. Inmates who lack self-regulation are more likely to be impulsive, fearful, hostile, and display aggressive behaviors (Casarjian et al., 2007; Komarovskaya et al. 2007). Komarovskaya et al. (2007) found higher rates of violent acts committed by female inmates with less ability to manage impulsive tendencies in compared to counterparts who exercised more self-discipline. These researchers applied the Barratt Impulsiveness Scale (BIS) to determine levels of impulsivity. The higher the score on the 44-item self-reported scale, the greater the likelihood of impulsive behavior. Thirty-six percent of study participants were classified as severely impulsive.

An individual’s ability to develop understanding and resilience, especially toward negative stimuli, defines mindfulness (Heppner et al., 2008; Vujanovic et al., 2010). Mindfulness training includes a variety of practices like breathing exercises (Leonard et al., 2013). While the individual focuses on their emotions, they are instructed by a facilitator to remain focused on their breathing and to be open to drifting thoughts and emotions during the exercise without judgment (Leonard et al., 2013). In addition to breathing exercises, other examples of mindfulness training include yoga and meditation (Ferszt et al., 2009; Middleton et al., 2019). Overall, studies examining the effect of diverse mindfulness programs have revealed positive program effects (e.g., Dewey et al., 2019; Leonard et al., 2013; Middleton et al., 2019; Rousseau, et al., 2019).

Mindfulness training is usually paired with other forms of treatment interventions. For example, stress reduction methods, as well as cognitive behavioral therapy, often accompany EL programs (Leonard et al., 2013). Mindfulness-based Cognitive Behavioral Therapy (MBCBT) programs combine Mindfulness Training (MT) with exercises that help a person learn to regulate their emotions and behavior (Leonard et al., 2013). Leonard et al. (2013) found that the CBT/MT intervention significantly attenuated degradation in attentional control, whereas the group that received no CBT/MF was significantly worse on attentional control testing after a five-week duration. The other example is the combination of art programs with mindfulness training (Dewey et al., 2019). In their report of this programming with 75 Wyoming state female inmates, Dewey et al. (2019) documented that arts-based mindfulness education exerted positive and long-lasting impact on incarcerated people.

Empathy is reflected by an individual’s ability to understand and resonate with another’s emotions (Lockwood et al., 2014). It is separated into two types, cognitive and affective empathy. Cognitive empathy is an individual’s ability to understand and identify the emotions of others; affective empathy is an awareness of the emotions of another person (Lockwood et al., 2014). Individuals who are able to understand their emotions and the emotions of others are more likely to behave in a prosocial manner (Lockwood et al., 2014). Connections between lower empathy levels and increased delinquent behaviors have been found in both juvenile and adult studies (Knight & Modi, 2014; Salas-Wright et al., 2012). Therefore, increasing an inmate’s empathy level may decrease propensity toward overly aggressive behavior. Van der Helm et al. (2012) reported on a program soley focused on improving juveniles’ empathy levels in a Netherlands juvenile correctional facility. The study yielded significant results between cognitive empathy and carceral settings that facilitated either supportive or repressive environments (Van der Helm et al., 2012). In other words, in a supportive correctional environment, the inmate gains in cognitive empathy, whereas a repressed environment diminishes the inmate’s ability to be empathetic.

Applications of EL Programs

Bullying is increasingly recognized as a form of delinquent behavior. It can take place both physically and verbally. A study in the UK used pre- and post-testing on an EL school system intervention program specifically targeting students who bully others. Previous research illustrates that bullies differ from peers who conduct prosocial behaviors because they are unable to acknowledge the extent to which their behaviors affect the emotions and feelings of others (Knowler & Frederickson, 2013). Study results indicate that the intervention group significantly decreased their bullying behaviors compared to the comparison group. This EL program focused on (1) developing self-awareness, (2) learning self-regulation, (3) enhancing empathy, and (4) improving social skills (Knowler & Frederickson, 2013). Participants who engaged in 12 weekly sessions were able to develop self-awareness by identifying strengths, feelings, emotions, and appropriate behaviors (Knowler & Frederickson, 2013).

Although this EL program was not heldin a correctional setting, its fundamental components havebeen implemented in prison settings elsewhere. Houses of Healing is a 12-week EL program that made its first appearance in a variety of prisons in the Northeastern part of the United States. It teaches inmates similar practices to those in the UK program. A greater capacity for self-awareness, stress-management, and self-regulation provides inmates with alternate ways to respond to prison life challenges (Casarjian et al., 2007). By providing inmates with EL training in small-group settings supported with a counselor to facilitate the practice of these new skills via exercises with other inmates, participants were better able to understand the emotions of others and more effectively regulate their own emotions in stressful situations through role playing and group discussions (Casarjian et al., 2007).

Power Source is another EL program that is designed particularly for youth detention facilities. It combines cognitive behavioral therapy with mindfulness training to reduce aggressive behaviors among adolescent inmates (Leonard et al., 2013). Power Source emphasizes positive role models and peers choices, placement in situations where violence or offending regulations are avoided and will not occur, and development of skills to alter circumstances when maladaptive behavior ensues (Leonard et al., 2013). Evaluation of the Power Source model has shown effectiveness in limiting deprivation in attentional performance for adolescent inmates (Leonard et al., 2013).

Distinguishing EL Programs from Conflict-Resolution Programs

Conflict resolution, including strategies to achieve peaceful dispute resolution or mediation, is well represented in workplaces and educational settings, as well as the larger geopolitical context. Such programs that are implemented in prison train inmates and prison staff to manage and resolve conflicts effectively (Love, 1994). EL programs, on the other hand, enable inmates to avoid the occurrence of conflicts or to deescalate situations that may become problematic. Conflict resolution involves multiple parties and does not hone in on the specific individual triggers that may facilitate conflicts.

Moreover, conflict resolution training does not provide the skills necessary for emotional self-regulation. In contrast, EL teaches inmates that understanding their emotions defuses a trigger for impulsive and poor behavior, with the result that they may reduce their aggressiveness and solve conflicts on their own without aid from a third party. The Pennsylvania Department of Corrections describes six steps in the conflict resolution program often employed in its prison system: (1) defining the problem, (2) brainstorming possible solutions, (3) having each party eliminate unacceptable solutions, (4) selecting a solution or a combination of solutions, (5) implementing the solution(s), and (6) setting time for evaluation (Love, 1994). Program goals involve improving listening and problem-solving skills, clarifying values, mediation, and anger control (Love, 1994). Some of these goals appear to align with those found in EL, but there are differences. As noted above, EL seeks to provide participants with practices that decrease their likelihood of being in conflict at the outset, thus preventing the need to resolve something that did not develop. Conflict resolution skills are important, but an inmate who is involved in an EL program can learn to control and facilitate their emotions appropriately, thereby reducing the likelihood of escalating tensions.

While there are a growing number of studies that recognize the positive effects of EL initiatives on a variety of carceral populations, most of these studies describe these effects as “soft conditions” (e.g., improved self-esteem and interpersonal skills) that are a critical nexus for the reduction of recidivism (Cheliotis, 2014; Cheliotis & Jordanoska, 2016; Davey et al., 2014; Miner-Romanoff, 2016; Rosenbaum, 2019). There are, nonetheless, too few studies that systematically explore the efficacy of EL programs among correctional populations and its potential to impact recidivism. The current study aims to address this deficiency by investigating the utility of Emotional Literacy programs for incarcerated persons through the meta-analysis.

Method

Search Strategies

Studies were collected by searching for peer-reviewed articles from the following databases: Criminal Justice Collection, Criminology: a Sage Full-Text Collection, SocIndex with Full Text, and EBSCO Host. In addition, a search of the Google Scholar database, a manual review of relevant journals and reference lists of relevant articles were conducted to identify articles published through early 2017. As this study focuses on evaluating Emotional Literacy (EL) programs that take place in prison systems, the search was conducted combining the following search terms: Emotional Literacy program, intervention, treatment, therapy, training, empathy, self-regulation, mindfulness, prison, and incarceration. From this process, a total of 15 peer-reviewed journal articles were identified and introduced to the current study. As studies that were published in peer-reviewed journals can be a non-random sample of those that were actually conducted, the current research examines publication bias to determine if there exist any selection biases in the effect size estimates.

Selection of Studies

Several criteria were utilized to identify studies for the sample data and to decline others. Studies were judged relevant if they: (1) included emotional or psychological treatment programs for inmates and; (2) included data relevant to either empathy, mindfulness, or self-regulation factors associating with treatment programs for aggressive behaviors. Likewise, studies were excluded: (1) when the full text of the study was not available; (2) the study reported insufficient information to conduct an analysis of the effect sizes, and additional information could not be acquired from the author; (3) the study reported no results for either of the three outcome variables and; (4) the data overlapped either partly or completely with the data from other studies included in the current meta-analysis. No geographical, cultural, or age restriction was applied to data selection. Research that used intergroup comparison designs, intergroup change designs, or randomized controlled trials were used to determine the controlled effect sizes for intergroup comparisons at the post-treatment phase. For those with multiple outcome variables related to Emotional Literacy programs, each variable was independently included. Only the findings from the analysis of the total sample were included for the studies analyzing both total sample and sub-samples within the same population. We do not impose any restriction on participant conditions and publication date as these factors do not meaningfully influence the study of Emotional Literacy programs.

To analyze the effect of EL programs more comprehensively, we included all the variables which were related to three outcome measures of EL programs: empathy, mindfulness, and self-regulation. For the identification of mindfulness outcomes, studies analyzing spirituality measures were included. Self-regulation variables also included self-control measures as well as a measure for hostility as it indicates the opposite of self-regulation (Samuelson et al., 2007). Empathy outcomes were not comprised of any additional dimensions. Follow-up data were included if the above variables were available.

We rated the quality of each study design according to the Scientific Methods Scale (SMS) criteria developed by Farrington et al. (2002). Each study was classified independently in accordance with five levels from one (the weakest scientific method) to five (the highest scientific method); a study quality score was then used as a moderator (Farrington et al., 2002). The detailed study selection process is illustrated in Figure 1.

Figure 1.

Flow diagram of the review and input study selection process.

Data Items

Information was extracted from each study according to: (1) characteristics of the study including publication year, number of participants, type of outcome variables, type of assessment tools, and follow-up period in months; (2) characteristics of the treatment including type of treatment, target population, and hours of intervention applied and; (3) characteristics of participants including average age, percentage of female participants, and cultural background. If the length of treatment is not reported in hours or cannot be calculated in hours, we considered it as missing information. A full list of the studies included in the meta-analysis and their descriptive information are found in Table 1.

Table 1.

List of Studies and Their Descriptive Information.

Author (year)	N total (tNto/tNt1)¹ (cNt0/cNt1)²	Treatment type	Hours of intervention	Average age (% female)	Follow-up period	Population	Region	Assessment	SMS
Barnert et al. (2014)	29 (16/16) (13/13)	MBA	22	16.3 (0)	None	Adolescent males	US	Mindfulness (MAAS-A), self-regulation (HSR)	2
Casarjian et al. (2007)	69 (69/69) (0/0)	HHP	24	Not reported (0)	None	Males	US	Hostility (CMHS), mindfulness (PSI)	2
Evans-Chase (2013)	27 (29/14) (30/13)	MMT	8	18 (0)	None	Youth	US	Self-regulation (SAS, ICS)	4
Ferszt et al. (2009)	18 (21/7) (15/11)	HHP	Not reported	34 (100)	None	Women	US	Spirituality (SPS)	3
Harris and Landreth (1997)	22 (12/12) (10/10)	FT	20	32 (100)	None	Mothers	US	Empathy (MEACIS)	3
Helm et al. (2012)	55 (59/55) (0/0)	PGCI	Not reported	17.4 (0)	None	Delinquent boys	Netherlands	CE, AE (BES)	1
Himelstein (2011)	48 (60/48) (0/0)	MBSAT-IY	12	16.3 (0)	None	Youth	US	Self-regulation (HSR)	2
Himelstein et al. (2011)	31 (32/31) (0/0)	MBA-A	10	16.75 (0)	None	Adolescents	US	Mindfulness (MAAS), Self-regulation (HSR)	2
Himelstein et al. (2015)	27 (27/27) (0/0)	MBSAT-M	18	16.45 (0)	None	Adolescents	US	Mindfulness (MAAS)	4
Landreth and Lobaugh (1998)	32 (16/16) (16/16)	FT	15	30.94 (0)	None	Fathers	US	PAC (PPAS)	4
Le and Proulx (2015)	33 (33/33) (0/0)	MBA-A	10	Not reported (24)	None	Youth	Hawaii	Mindfulness (CAMM), self-regulation (HSR)	2
Pechorro et al. (2015)	221 (221/221) (0/0)	None	None	16.75 (0)	None	Juvenile offenders	Portugal	Empathy (BES)	1
Samuelson et al. (2007)	1350 (1486/1050) (467/300)	MBSR	20	Not reported (22)	None	Inmates	US	Hostility (CMHS)	2
Vannoy and Hoyt (2004)	31 (16/16) (15/15)	AT	18	35(0)	None	Adult males	US	EC (IRI)	4
Woessner and Schwedler (2014)	185 (185/185) (0/0)	STT	Not reported	31.8 (0)	48.4	Sexual and violent offenders	Germany	Empathy (E-scale), EI (FPI-R), LSC (LSCS)	2

Note. MBA = mind body awareness project; MAAS-A = mindfulness attention awareness scale, adolescent version; HSR = healthy self-regulation scale; MBSAT-M = mindfulness-based substance abuse treatment-a modified version; HHP = house of healing program; CMHS = cook and Medley hostility scale; PSI = psychometric spirituality inventory; MMT = mindfulness meditation treatment; SAS = suppression of aggression subscale; ICS = impulse control subscale; SPS = spiritual perspective scale; FT = filial therapy; MEACIS = the measurement of empathy in adult–child interaction; PGCI = prison group climate instrument; CE = cognitive empathy; AE = affective empathy; BES = basic empathy scale; MBSAT-IY = mindfulness-based substance abuse treatment for incarcerated youth; MBSR = mindfulness-based stress reduction; MBA-A = mind body awareness project, adapted version; MAAS = mindfulness attention awareness scale; PAC = parental acceptance of child; PPAS = porter parental acceptance scale; CAMM = child and adolescent mindfulness measure; AT = anger therapy; EC = empathic concern; IRI = interpersonal reactivity index; STT = social-therapeutic treatment; EI = emotional instability; FPI-R = Freiburg personality inventory-revised edition; LSC = low self-control; LSCS = low self-control scale.

Number of subjects at pre- and post-test in the treatment condition.

Number of subjects at pre- and post-test in the control condition.

Analytical Strategy

A meta-analysis was conducted by using the computer program Comprehensive Meta-Analysis, Version 3 (Borenstein et al., 2009). Separate effect sizes for empathy, mindfulness, and self-regulation outcomes were calculated by using intragroup designs, intergroup designs, and a combination of both designs. Pearson correlation coefficient was used to compute effect size when available. In the other studies, the effect size was calculated by using the provided means, standard deviations, and other statistics, including F, p, and t. In the intragroup study design, some studies did not report the correlation between pre- and post-treatments which are necessary to estimate pre-post effect size. In that case, we used a conservative estimate of r = 0.70 according to the recommendation by Rosenthal (1993).

For all studies, Hedges’ g was computed to calculate effect sizes. The equation of Hedges’ g is given as:

g = d (1 - \frac{3}{4 (n^{t} + n^{c}) - 9})

(1)

where, d denotes Cohen’s d, and $n^{t}$ and $n^{c}$ indicates the sample sizes of treatment and control groups correspondingly. As shown at equation (1), Hedges’ g controls for the possible bias of Cohen’s d when the sample size is not big enough (see Hedges & Olkin, 1984). Cohen (1988) suggests that the magnitude of Hedges’ g of 0.20 might be considered small, 0.50 considered medium, and 0.80 considered large. In addition, a 95% confidence interval and related p values were calculated for each Hedges’ g value.

To obtain the mean effect size of selected studies, we pooled individual effect size estimates across studies by performing a random-effect model rather than a fixed-effects model, given that the effect size parameters were not identical (Hedges & Vevea, 1998). The $I^{2}$ index and the Q statistic were examined to assess the heterogeneity of individual effect sizes among the selected studies in each group. $I^{2}$ measures the proportion of variations due to differences among studies, and the degree of $I^{2}$ can be interpreted using Higgins et al.’s (2003) recommendation for low (0.25), moderate (0.50), and high (0.75) (Higgins et al., 2003). A significant value of Q means that estimated effect sizes were dissimilar.

As a check against publication bias, we computed the fail-safe N value (Rosenthal, 1993) to demonstrate how many non-significant studies would be needed to change significant findings into non-significant ones. In meta-analyses, this estimation has been introduced to address the “file drawer problem” (Winkler et al., 2013). The results of the meta-analysis are deemed to be robust against publication bias if the fail-safe N value is over 5 K + 10 (Winkler et al., 2013).

Findings

Characteristics of Studies

Among 15 studies selected for the meta-analysis, some are found to examine the connections between factors of EL and aggressive behaviors while others are more involved in the inspection of specific treatment programs. More specifically, 13 of the 15 articles provide evaluations of treatment programs in prison, with the remaining two articles consisting of analyses that examined the effect of EL programs on aggressive behaviors when implemented in a carceral setting. As stated previously, studies were divided into three groups according to the outcome variables each assessed: Empathy; Mindfulness; and Self-regulation. If a study contains multiple outcome variables that are eligible to be included in corresponding groups, the analytical result from each variable was introduced to the corresponding group meta-analysis.

Six studies are found to employ an experimental design, with a control group, and four of these are randomized trials; eight others did not have a control group as a treatment condition. The total number of inmates included in the meta-analysis was 2,178, with 1,800 in treatment and 378 in control groups. The number of female inmates was 345 (15.84%), and the average age of participants ranged from 16.30 to 35 (M = 23.47, SD = 7.91). Eleven different types of treatments were applied, and the numbers of intervention hours reported in 11 ranged from 8 to 24 (M = 16.09, SD = 5.16). Follow-up data were not reported in most except for one (48.40 months). The majority were conducted in the mainland of the US (n = 11); others examined EL programs in Germany (n = 1), Portugal (n = 1), the Netherlands (n = 1), and Hawaii (n = 1). The total number of outcome variables included in the current study was 22: seven outcome variables were included in the Empathy group, seven in Mindfulness, and eight in Self-regulation. The SMS scores for each study are provided in Table 1, and ranged from 1 to 4 points (M = 2.64, S = 0.97).

Results from Meta-Analyses

First, we gained significant results from the Q-test of heterogeneity in effect size estimates for Empathy (Q(6) = 63.16, p < .000), Mindfulness (Q(6) = 21.92, p = .001), and Self-Regulation (Q(7) = 24.34, p = .001). This finding suggests that there exists a significant amount of heterogeneity across the effect size estimates of studies in each group and supports our use of random-effect meta-analysis rather than a fixed-effect approach. The estimations of $I^{2}$ statistics for three groups also show a large amount of inconsistency (Empathy $I^{2}$ = 90.50%, Mindfulness $I^{2}$ = 72.63%, Self-Regulation $I^{2}$ = 71.24%), which also provides the consistent results with those of Q-tests.

Secondly, the results from the Hedges’ g value estimations for selected studies, the 95% confidence intervals (CI), standard errors (SE), and p-values of intergroup comparisons or intragroup changes, as well as forest plots of effect size estimates, are shown in Table 2. As for the studies of Empathy, three out of seven studies have presented significant positive effects of EL programs on the degree of inmates’ empathy statuses, such as Harris and Landreth (1997) (g = 2.084, p < .001), Landreth and Lobaugh (1998) (g = 1.559, p < .001), and Pechorro et al. (2015) (g = 0.896, p < .001). The other four studies have reported positive changes; however, they are not statistically significant. The random-effect meta-analysis of these seven studies has presented the overall significant positive effect of the EL program on Empathy (g = 0.655, p = .006).

Table 2.

Effect Size (Hedges’ g) Estimation for Empathy, Mindfulness, and Self-regulation.

Authors (publication year)	g	SE	95% CI	p
Empathy
Harris and Landreth (1997)	2.084	0.518	[1.068, 3.099]	<.001**
Helm et al. (2012) (1)	0.553	0.284	[−0.004, 1.109]	0.052
Helm et al. (2012) (2)	0.032	0.273	[−0.504, 0.567]	0.908
Landreth and Lobaugh (1998)	1.559	0.396	[0.783, 2.335]	<.001**
Pechorro et al. (2015)	0.896	0.148	[0.606, 1.186]	<.001**
Vannoy and Hoyt (2004)	0.022	0.350	[−0.664, 0.708]	.951
Woessner and Schwedler (2014)	0.060	0.148	[−0.230, 0.349]	.686
All studies	0.655	0.240	[0.184, 1.125]	.006**
Mindfulness
Barnert et al. (2014)	0.187	0.182	[−0.170, 0.545]	.305
Casarjian et al. (2007)	0.552	0.167	[0.225, 0.879]	<.001**
Ferszt et al. (2009)	0.004	0.460	[−0.899, 0.906]	.994
Himelstein et al. (2011)	0.256	0.178	[−0.093, 0.606]	.150
Himelstein et al. (2015)	0.149	0.166	[−0.177, 0.475]	.371
Le and Proulx (2015)	0.113	0.171	[−0.221, 0.447]	.507
Woessner and Schwedler (2014)	0.336	0.130	[0.081, 0.591]	.010**
All studies	0.270	0.066	[0.142, 0.399]	<.001**
Self-regulation
Barnert et al. (2014)	0.290	0.185	[−0.072, 0.652]	.117
Casarjian et al. (2007)	0.271	0.121	[0.033, 0.509]	.025*
Evans-Chase (2013)	0.019	0.374	[−0.713, 0.751]	.960
Himelstein et al. (2011)	0.828	0.204	[0.428, 1.229]	<.001**
Himelstein (2011)	0.155	0.143	[−0.125, 0.435]	.279
Le and Proulx (2015)	0.030	0.170	[−0.303, 0.363]	.859
Samuelson et al. (2007)	0.127	0.033	[0.063, 0.191]	<.001**
Woessner and Schwedler (2014)	0.265	0.129	[0.013, 0.518]	.039*
All studies	0.230	0.068	[0.097, 0.363]	.001**

Note. The table displays estimated effect sizes (Hedges’ g), standard errors, the 95% confidence intervals, and the significance of changes in empathy, mindfulness, and self-regulation from before to after EL treatment of inmates.

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

All seven studies on Mindfulness have also presented the positive effects of EL programs; however, only two of them are statistically significant such as Casarjian et al. (2007) (g = 0.552, p = .001) and Woessner and Schwedler (2014) (g = 0.336, p = .01). As the sample sizes of the other five non-significant studies are relatively small, nonetheless, their weights in overall effect size are minimal, and the overall effect size of seven studies has shown the significant positive effect of EL program on Mindfulness (g = 0.270, p < .001). The analysis of Self-regulation have revealed that four out of eight studies present significant positive effect such as Casarjian et al. (2007) (g = 0.271, p = .025), Himelstein et al. (2011) (g = 0.828, p < .001), Samuelson et al. (2007) (g = 0.127, p < .001), and Woessner and Schwedler (2014) (g = 0.265, p = .039). The overall effect size of Self-regulation studies is estimated at 0.230, which is significant at the 0.01 significance level (p = .001).

To examine if our significant findings are robust against publication bias, we have estimated the classic fail-safe N statistics, and the results are presented in Table 3. The fail-safe N statistics of Empathy and Self-regulation are larger than those of thresholds suggested by Winkler et al. (2013). They support that our findings are robust against the file-drawer problem occurring from a tendency to publish positive results but not to publish negative or non-confirmatory results. The fail-safe N statistic of Mindfulness, however, is found to be smaller than the threshold value and suggest that our finding about the effect of the EL program on Mindfulness should be revisited with more representative samples.

Table 3.

Classic Fail-safe N Statistics Estimation for Empathy, Mindfulness, and Self-regulation.

	Empathy	Mindfulness	Self-regulation
Threshold	45	45	50
Fail-safe N	65	19	52

Discussion

This study aims to assess and compare the effectiveness of Emotional Literacy carceral-based programs for empathy, mindfulness, and self-regulation. Notably, we examined 15 studies that include 14 treatment conditions and 2,178 inmates of diverse ages and genders with a random-effect meta-analysis. Although the result should be regarded as preliminary due to the limited number of studies, the findings illustrate that EL prison program effects are considerable and support our speculation that participating in one would increase inmate empathy, mindfulness, and self-regulation. The random-effect meta-analysis has revealed that the overall positive effect of EL programs on these three conditions are statistically significant, while the possibility of publication bias is inspected for its effect on mindfulness.

More specifically, the studies with a larger sample size such as Samuelson et al. (2007), Pechorro et al. (2015), and Woessner and Schwedler (2014) have reported that EL programs retain a significant positive impact on prosocial changes in incarcerated violent offenders. As these studies are given more weight in our random-effect meta-analyses, they also have mainly driven our findings. On the contrary, most of the studies with smaller sample sizes have failed to present a significant effect of EL programs, and their effects on our meta-analyses are minimal. While all the studies in our meta-analyses have employed EL programs, their specific approaches are distinctive from each other. While some programs are rather focused on social-therapeutic treatments (i.e., Woessner & Schwedler, 2014), others have applied mindfulness-based stress reduction program (i.e., Casarjian et al., 2007; Samuelson et al., 2007) or Mind Body Awareness self-regulation programming (i.e., Himelstein, 2011). After reviewing the components of EL programs in these studies, we have found the following factors which should contribute to the enlargement of the effectiveness of future EL programs:

Inclusion of essential EL program components

The first noteworthy factor is that a program that incorporates all three aspects of EL programs (i.e., empathy, self-regulation, and mindfulness) is more effective on prosocial change of inmates. For example, although the Houses of Healing program could enlarge the empathy level of incarcerated persons, it could be strengthened by incorporating other components to improve the psychological status of inmates (Casarjian et al., 2007). Such components could include participant group work with victimology experts, exposure to media about experiences of crime victims, the use of case scenarios that focus on crime victims, or meetings with actual crime victims (Lummer & Hagemann, 2014). Similarly, the Power Source program does not include program components for empathy. However, through adding other components, the Power Source program should increase not only levels of mindfulness but also those of empathy and self-regulation (Leonard et al., 2013). By including all the components, more comprehensive EL programs should provide inmates with increased skills in all three program dimensions and aid them in adopting alternative non-violent behaviors that may impact current and post-release adjustment.

Lengthof program duration

Treatment period should also be sufficient in length. For example, the Houses of Healing program, which is found to be effective on both mindfulness and self-regulation, runs for 12-weeks, with each weekly session lasting 2 hours (Casarjian et al., 2007). Furthermore, the other successful program, a therapeutic program implemented in Germany as described by Woessner and Schwedler (2014), runs for 13 months, the longest period in the programs we analyzed. While the average length of time inmates spent in this program is not discussed by the researchers, its successful effect should be based on a sufficient period of program time allocation. We speculate that approximately 5 hours of training a week for 6 to 12 months should be sufficient for EL programs to be effective on inmates’ prosocial behaviors.

Facilitator-led small groups in a supportive environment

We have also found that the EL programs employing a small group model and conducted in a supportive correctional environment are more effective. After evaluating the therapeutic intervention program in Germany, for instance, Woessner and Schwedler (2014) have identified two core components of successful intervention: (1) working in small groups and, (2) leadership by a facilitator who is able to establish a positive relationship with program participants. These factors enable every member to participate and engage fully in group dynamics. Overall, it is reported that participants have performed better in programs where the environment is appropriately responsive, and where there is mutual respect and trust between the participants and the group facilitator (Woessner & Schwedler, 2014).

Specifying the goals of program activities

To make EL programs more effective, it is recommended that the goals of each practice to increase inmates’ skills in self-regulation, mindfulness, and empathy be specified. For example, strengthening self-regulation skills necessitates the following practices: selecting positive peers, managing stress, transforming anger and resentment constructively, developing an awareness of consequences of negative emotions, and practicing methods of acting rationally and non-aggressively in challenging situations (Ferszt et al., 2009; Leonard et al., 2013). Through clarifying the goals of these activities, facilitators can more effectively assist program participants in working together and being involved in diverse activities positively. Another example can be found from the self-regulation practices—such as mindfulness of breathing, yoga, meditation, and control attention (Ferszt et al., 2009). Specifying the goals of these mindfulness exercises should prepare participants (especially those in a prison-setting) to be attentive in their human interactions and allow negative thoughts and emotions to pass rather than act on them aggressively or impulsively.

Policy Implications

The current study has found that the implementation of EL programs in carceral settings is effective in promoting inmates’ prosocial behaviors and improving adverse prison conditions. First, the successful implementation of these programs, especially in male prison settings, could provide several benefits that include less occurrences of distress and apprehension, contentiousness, and violence. Additionally, EL programs can increase the understanding of anger and hostility among inmates and decrease the level ofhypermasculinity and exhibition of “toughness.” These benefits can transform the atmosphere of prison and even reform prison subcultures. Second, the use of EL programs can reduce the need for solitary confinement for those who are demonstrating poor behavior on a heightened level—or shorten stays in security housing units. For example, the California Department of Corrections and Rehabilitation has employed the EL training to reduce the number of inmates who are currently housed in solitary confinement. They note that EL program participants better understand how past experiences, misinterpretations of emotions, and bad choices have led to their current circumstances (Casarjian et al., 2007). Finally, upon reentry, ex-offenders with better self-regulation skills are more likely to retain a higher level of prosocial behaviors and less likely to recidivate. This benefit of EL programs is also expected to contribute to the relief of prison overcrowding.

Limitations

Despite the meaningful findings of this research, several limitations shouldbe noted for future applications. First, we included studies with EL-related factors into our meta-analysis if they have employed similar approaches to the major factors in Emotional Literacy programs. In addition, we have also included other treatments or interventions which are designed to impact three EL factors. These approaches can be debatable because some treatments/interventions overlap with other types of rehabilitation programs. We suggest that future studies should control for approaches of other rehabilitation programs and estimate the effect of EL programs more precisely.

Second, the publication bias tests have revealed that our finding about the effect of EL programs on mindfulness could have a file-drawer problem, indicating our results are vulnerable to the tendency of publishing only positive results but not publishing negative or non-confirmatory results. This result needs the attention of future studies, as five of seven studies on mindfulness have reported the non-significant effect of EL programs on mindfulness. While two studies with larger sample sizes have induced the overall significant effect in our meta-analysis, our findings should be reexamined with future studies with more cases.

Conclusion

Although research on EL programs is in the beginning stages, the potential of these correctional programs to improve offender prosocial behaviors in significant and measurable ways should not be ignored. Rather, our findings suggest that there should be a greater effort to implement EL programs in U.S. correctional settings and elsewhere. Just as restorative justice models have had an increased footprint in corrections, Emotional Literacy programs should also be explored systematically as a viable approach to addressing inmate cognitive and emotional behaviors and choices through careful planning, implementation, and evaluation. Emotional Literacy has proven to be an effective approach in educational settings, and the results of our assessments of its effectiveness in correctional facilities also hold promise. The current study contributes to the further understanding of this lesser-known correctional treatment program and anticipates additional interest in EL programs as a result.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Author Biographies

Carolyn Petrosino, PhD, is professor emerita of criminal justice at Bridgewater State University. Her publication record includes journal publications, book chapters, and several books in the areas of institutional corrections, criminal justice policy, and hate crimes.

Kyung-Shick Choi, PhD, is professor of practice and director of the Cybercrime Investigations and Cybersecurity graduate program at Boston University. His research focus in on the intersection of human behavior and technology.

Sinyong Choi is a doctoral student in Criminal Justice at the University of Nevada, Las Vegas. His current research involves the study of cryptocurrency gambling, illegal online sports betting, fear of crime in social media, darknet market, and program evaluation.

Sinchul Back is an assistant professor at the University of Scranton. He obtained his PhD in international crime and justice at Florida International University and received his M.S. in criminal justice at Bridgewater State University.

Seong-Min Park, PhD, is an assistant professor of criminal justice at the University of Nevada, Las Vegas. His research focuses on examining crime patterns and addressing related methodological issues, and his works have been published in multiple peer-reviewed journals.

References

Agnew

(2009). General strain theory: Current status and directions for further research. In Cullen

F. T.

Wright

J. P.

Blevins

K. R.

(Eds.), Taking stock: The status of criminological theory (pp. 101–123). Transaction.

Barnert

E. S.

Himelstein

Herbert

Garcia-Romeu

Chamberlain

L. J.

(2014). Exploring an intensive meditation intervention for incarcerated youth. Child and Adolescent Mental Health, 19(1), 69–73.

Blevins

Listwan

Cullen

Jonson

(2010). A general strain theory of prison violence and misconduct: An integrated model of inmate behavior. Journal of Contemporary Criminal Justice, 26(2), 148–166.

Borenstein

Hedges

L. V.

J. P.

Rothstein

H. R.

(2009). Introduction to meta-analysis. Wiley.

Bureau of Justice Statistics. (2018). Prisoners in 2016. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. https://www.bjs.gov/content/pub/pdf/p16.pdf

Camilleri

Caruana

Falzon

Muscat

(2012). The promotion of emotional literacy through personal and social development: The Maltese experience. Pastoral Care in Education, 30(1), 19–37.

Casarjian

Philips

Wolman

(2007). An emotional literacy intervention with incarcerated individuals. American Journal of Forensic Psychology, 25(3), 43–63.

Cheliotis

(2014). The arts of desistance: Evaluation of the Koestler trust arts mentoring programme for former prisoners. http://www.artsevidence.org.uk/evaluations/arts-desistance-evaluation-koestler-trust-arts-men/

Cheliotis

Jordanoska

(2016). The arts of desistance: Assessing the role of arts-based programmes in reducing reoffending. The Howard Journal of Criminal Justice, 55(1–2), 25–41.

10.

Clear

T. R.

Cole

G. F.

Reisig

M. D.

Petrosino

(2015). American corrections in brief. Cengage Learning.

11.

Cochran

J. C.

Mears

D. P.

Bales

W. D.

(2014). Assessing the effectiveness of correctional sanctions. Journal of Quantitative Criminology, 30(2), 317–347.

12.

Cohen

(1988). Statistical power analysis for the behavioral sciences. Lawrence Earlbaum Associates.

13.

Davey

Day

Balfour

(2014). Performing desistance: How might theories of desistance from crime help us understand the possibilities of prison theatre? International Journal of Offender Therapy and Comparative Criminology, 59, 798–809.

14.

Dewey

Muthig

A. J.

Colter

R. S.

Brock

(2019). Art and mindfulness behind bars: Examples from Wyoming prison-based courses on memoir and stoic philosophy. The Prison Journal, 99(4S), 61S–83S.

15.

Evans-Chase

(2013). Internet-based mindfulness meditation and self-regulation: A randomized trial with juvenile justice involved youth. Journal of Juvenile Justice, 3(1), 63–79.

16.

Farrington

D. P.

Gottfredson

D. C.

Sherman

L. W.

Welsh

B. C.

(2002). The Maryland scientific methods scale. In Sherman

L. W.

MacKenzie

L. D.

Farrington

D. P.

Welsh

B. C.

(Eds), Evidence-based crime prevention (pp. 13–21). Routledge.

17.

Ferszt

Salgado

DeFedele

Leveillee

(2009). Houses of healing a group intervention for grieving women in prison. The Prison Journal, 89(1), 46–64.

18.

Fleischer

(2010). Developing emotional literacy: Transition planning for youth at risk. Reclaiming Children and Youth, 19(1), 50–53.

19.

Gendreau

French

S. A.

Gionet

(2004). What works (what doesn’t work): The principles of effective correctional treatment. Journal of Community Corrections, 13(3), 4–70.

20.

Harris

Z. L.

Landreth

G. L.

(1997). Filial therapy with incarcerated mothers: A five week model. International Journal of Play Therapy, 6(2), 53–37.

21.

Hedges

L. V.

Olkin

(1984). Nonparametric estimators of effect size in meta-analysis. Psychological Bulletin, 96, 573–580.

22.

Hedges

L. V.

Vevea

J. L.

(1998). Fixed- and random-effect models in meta-analysis. Psychological Methods, 3, 486–504.

23.

Helm

G. H. P.

Stams

G. J. J.

Stel

J.C.

Langen

M. A. M.

Laan

P. H.

(2012). Group climate and empathy in a sample of incarcerated boys. International Journal of Offender Therapy and Comparative Criminology, 56(8), 1149–1160.

24.

Heppner

Kernis

Lakey

Campbell

Goldman

Davis

Cascio

(2008). Mindfulness as a means of reducing aggressive behavior: Dispositional and situational evidence. Aggressive Behavior, 34, 486–496.

25.

Higgins

J. P.

Thompson

S. G.

Deeks

J. J.

Altman

D. G.

(2003). Measuring inconsistency in meta-analyses. British Medical Journal, 327(7414), 557–560. https://doi.org/10.1136/bmj.327.7414.557

26.

Himelson

(2011). Mindfulness-based substance abuse treatment for incarcerated youth: A mixed method Pilot study. International Journal of Transpersonal Studies, 30(1–2), 1–10.

27.

Himelstein

Hastings

Shapiro

Heery

(2011). Mindfulness training for self-regulation and stress with incarcerated youth: A pilot study. Probation Journal, 59(2), 151–165.

28.

Himelstein

Saul

Garcia-Romeu

(2015). Does mindfulness meditation increase effectiveness of substance abuse treatment with incarcerated youth? A pilot randomized controlled trial. Mindfulness, 6(6), 1472–1480.

29.

Johnson

(1995). Hard time: Understanding and reforming the prison. Wadsworth

30.

Knight

Modi

(2014). The use of Emotional Literacy in work with sexual offenders. Probation Journal, 61(2), 132–147.

31.

Knowler

Frederickson

(2013). Effects of an Emotional Literacy intervention for students identified with bullying behaviour. Educational Psychology, 33(7), 862–883. https://doi.org/10.1080/01443410.2013.785052

32.

Komarovskaya

Loper

Warren

(2007). The role of impulsivity in antisocial and violent behavior and personality disorders among incarcerated women. Criminal Justice and Behaviour, 34(11), 1499–1515.

33.

Landreth

G. L.

Lobaugh

A. F.

(1998). Filial therapy with incarcerated fathers: Effects on parental acceptance of child, parental stress, and child adjustment. Journal of Counseling & Development, 76(2), 157–165.

34.

T. N.

Proulx

(2015). Feasibility of mindfulness-based intervention for incarcerated mixed-ethnic Native Hawaiian/Pacific Islander youth. Asian American Journal of Psychology, 6(2), 181.

35.

Leonard

Jha

Casarjian

Goolsarran

Garcia

Cleland

Gwadz

Massey

(2013). Mindfulness training improves attentional task performance in incarcerated youth: A group randomized controlled intervention trial. Frontiers in Psychology, 4, 792.

36.

Lockwood

Seara-Cardoso

Viding

(2014). Emotion regulation moderates the association between empathy and prosocial behavior. PLoS One, 9(5), 1–6.

37.

Love

(1994). Program curbs prison violence through conflict resolution. Corrections Today, 56(5), 144–146.

38.

Lummer

Hagemann

(2014). Victim empathy within prison walls: Experiences from pilot projects in Schleswig-Holstein. Ljetopis Socijalnog Rada, 22(1), 37–60. https://doi.org/10.3935/ljsr.v22i1.84

39.

Middleton

Harris

D. A.

Ackerman

A. R.

(2019). A mixed-methods process evaluation of the art of yoga project for girls in custody. The Prison Journal, 99(4S), 38S–60S.

40.

Miner-Romanoff

(2016). Voices from inside: The power of art to transform and restore. Journal of Correctional Education, 67, 58–74.

41.

Pechorro

Ray

J. V.

Salas-Wright

C. P.

Maroco

Goncalves

R. A.

(2015). Adaptation of the Basic Empathy Scale among a Portuguese sample of incarcerated juvenile offenders. Psychology, Crime & Law, 21(7), 699–714.

42.

Pitts

Griffin

Johnson

(2014). Contemporary prison overcrowding: Short-term fixes to a perpetual problem. Contemporary Justice Review, 17(1), 124–139.

43.

Rhine

E. E.

Mawhorr

T. L.

Parks

E. C.

(2006). Implementation: The bane of effective correctional programs. Criminology and Public Policy, 5(2), 347–358.

44.

Rosenbaum

J. L.

(2019) Art and mindfulness behind bars. The Prison Journal, 99(4S), 3S–13S.

45.

Rosenthal

(1993). Meta-analytic procedures for social research. Sage Publications.

46.

Rousseau

Long

Jackson

Jurgensen

(2019). Empowering through embodied awareness: Evaluation of a peer-facilitated trauma-informed mindfulness curriculum in a woman’s prison. The Prison Journal, 99(4S), 14S–37S.

47.

Salas-Wright

Olate

Vaughn

(2012). Assessing empathy in Salvadoranhigh-risk and gang-involved adolescents and young adults: A Spanish validation of the basic empathy scale. International Journal of Offender Therapy and Comparative Criminology, 57(11), 1393–1416.

48.

Samuelson

Carmody

Kabat-Zinn

Bratt

M. A.

(2007). Mindfulness-based stress reduction in Massachusetts correctional facilities. The Prison Journal, 87(2), 254–268.

49.

Scott

Back

Choi

(2015). Emotional Literacy program and its effectiveness-potential policy implication in South Korea. Korean Criminal Psychology Review, 11, 274–293.

50.

Steiner

Perry

(1997). Achieving Emotional Literacy: A program to increase your emotional intelligence. Avon.

51.

Sykes

G. M.

(1958). The society of captives. Princeton University Press.

52.

Van der Helm

G. H. P.

Wissink

I. B.

De Jongh

Stams

G. J. J. M.

(2012). Measuring treatment motivation in secure juvenile facilities. International Journal of Offender Therapy and Comparative Criminology, 57(8), 996–1008.

53.

Vannoy

S. D.

Hoyt

W. T.

(2004). Evaluation of an anger therapy intervention for incarcerated adult males. Journal of Offender Rehabilitation, 39(2), 39–57.

54.

Vujanovic

Bonn-Miller

Bernstein

McKee

Zvolensky

(2010). Incremental validity of mindfulness skills in relation to emotional dysregulation among a young adult community sample. Cognitive Behaviour Therapy, 39(3), 203–213.

55.

Winkler

Dörsing

Rief

Shen

Glombiewski

J. A.

(2013). Treatment of internet addiction: A meta-analysis. Clinical Psychology Review, 33(2), 317–329.

56.

Woessner

Schwedler

(2014). Correctional treatment of sexual and violent offenders: Therapeutic change, prison climate, and recidivism. Criminal Justice and Behavior, 41(7), 862–879.