The Well-Being Impacts Associated with College in Prison: A Comparison of Incarcerated and Non-Incarcerated Students Who Identify as Women

Abstract

Benefits of college in prison (CIP) programs often center on reduced recidivism and other post-release adjustment outcomes. This article argues for broader recognition of positive outcomes to include well-being, coping, and academic engagement for incarcerated students, noting the specific challenges for female-identifying inmates with backgrounds that often include high levels of abuse and violence. Using longitudinal data from female-identifying students in a CIP program and comparison data from a matched non-incarcerated sample, improvements in these outcome measures are demonstrated as incarcerated students remain in the program. Importantly, their measures of well-being, coping, and academic engagement were significantly better than the matched non-incarcerated sample for most measures. Further, the possible underlying mechanism for the findings is discussed, linking them to program characteristics that emphasize the development of community, agency, and personalized—in-person education—qualities that are in alignment with trauma-informed practices.

Keywords

college-in-prison well-being gender

Introduction

Among the multiple benefits associated with college in prison (CIP) programs, advocates often cite lower recidivism rates as a means of justifying such educational expenditures to critics who voice concern over the cost-prohibitive nature of higher education for non-incarcerated members of society (Harnish, 2019). That is, connecting educational attainment to reductions in recidivism highlights the benefit to society-at-large in the form of lower crime rates or incarceration costs that require public funding. Scholars speculate that the association between higher education and lower recidivism—a relationship now well documented in the extant literature—is mediated via many mechanisms, including higher rates of post-release employment or higher wages and salaries (Cho & Tyler, 2013; Davis et al., 2013; Duwe & Clark, 2014; Erisman & Contardo, 2005; Kim & Clark, 2013; Vacca, 2004).

Acknowledging that the question of who is “deserving” of affordable education is fraught with many complexities, we argue that there are many other benefits associated with higher education besides the actually earned degree that may also mediate the relationship with lower recidivism. For example, enhanced self-esteem and coping skills (from accomplishing difficult tasks), specific academic skills and abilities (e.g., argument development and writing), meaningful social connections (from contacts with students and faculty), and “soft skills” (e.g., time management and stress management) are frequently cited as benefits of higher education (Morales, 2012; Pascarella & Blaich, 2013; Seifert et al., 2008). As a set, these types of experiences may contribute to an overall improved sense of well-being, coping, and academic engagement for all students; however, we believe these may be arguably more impactful for those who are incarcerated given that there are fewer opportunities to learn such skills in prison.

With such a strong focus on reducing recidivism rates, however, coupled with the methodological challenges of making causal inferences between education and well-being, these broader outcomes have not received much attention in previous research. The present study proposes to add to the growing body of literature that calls for scholars to attend to these broader outcomes (e.g., Gould, 2018; McCorkel & DeFina, 2019). Namely, we examine the connection between educational participation and well-being, coping, and academic engagement for incarcerated students. Further, our analyses provide several examples of innovative methodological approaches as a means of offering a stronger basis for making connections between participation in higher education and these outcomes.

Gender and Mental Health of Incarcerated Students

Adding to the importance of studying well-being, coping, and academic engagement, the mental health of incarcerated persons is, unsurprisingly, extremely poor and documented in the extant literature. As compared to the population more generally, incarcerated individuals are more likely to have been the victims of abuse and violent crime, and to have suffered from conditions such as substance use disorder and depression (Treatment Advocacy Center, 2016). Mental health and well-being disparities between the incarcerated and non-incarcerated populations are especially acute for incarcerated women. Indeed, the vast majority of women in prison have been victims of violence prior to their incarceration—including domestic violence, rape, sexual assault, and child abuse. Seventy-nine percent of women in federal and state prisons reported physical abuse, and over 60% reported past sexual abuse (Lapidus et al., 2005). Women in prison are three to four times more likely than male prisoners to have experienced abuse, whether as a child or adult (Lapidus et al., 2005). Such violence has been noted for decades: in the early 1990s, Snell reported an estimated 56% of those women in prison who experienced abuse reported that their abuse included rape, and another 13% reported an attempted rape (Snell, 1994). In New York, a 1999 study of female prisoners at Bedford Hills Correctional facility found that over 80% of those prisoners had a childhood history of physical and sexual abuse, and more than 90% had experienced physical or sexual assault during their lifetime (Women in Prison Project, 2002). Studies have demonstrated that violence also correlates with an increased likelihood of addiction and substance abuse (Allard, 2002). Thus, when trauma and violence are already present, students may have less confidence, coping skills, and self-esteem related to this violence and a constellation of connected issues.

In light of these realities, we believe that the broader well-being benefits are important in their own right for currently incarcerated students. However, we also believe that an improved sense of personal well-being, coping, and academic engagement may play an important role in post-release life, particularly for incarcerees who have experienced forms of violence and trauma. Furthermore, participation with others in a community-based educational program in which they share a supportive identity as a college student may be particularly important in the development of increased well-being (Baranger et al., 2018; Evans et al., 2018). Not only may these benefits contribute to successful post-release readjustment in many arenas beyond employment, but they may also have positive potential “ripple effects” on the families and communities to which students return after their release.

Current Study

Given the heavy focus on recidivism reduction as a means for justifying in-college prison programs, researchers have not typically studied the variables related to well-being in this population. When they have, scholars have often used anecdotal evidence or qualitative data, making it difficult to gauge the impact of the experience. There are also methodological challenges, such as finding measures whose item content applies to an incarcerated person or context and determining a type of comparison group to gauge the impact of the educational experience. This study aims to address these two issues. Namely, using multiple, quantitative indicators of personal well-being, coping, and academic engagement whose content fits both incarcerated and non-incarcerated students, we compare CIP students: (1) over time; (2) at different stages of degree completion to each other, and (3) to a comparison group of non-incarcerated students to measure the strength of the program's impact. We argue that this approach demonstrates several useful methodological options for researchers in this field of inquiry and helps clarify a number of the important benefits associated with CIP programs.

Method

CIP Program Description

In the present study, we examine the well-being, coping, and academic engagement impacts among female-identifying students who are enrolled in the Freedom Education Project Puget Sound (FEPPS) located at the Washington Corrections Center for Women. Initiated by student interest and founded in 2011, this program began as a noncredit-bearing educational opportunity that developed to offer students the ability to earn their AA degree. Volunteer faculty from area colleges and universities teach small, in-person courses that are modeled after a traditional liberal arts educational approach—one that includes critical inquiry, close reading, intensive writing, classroom participation, and high degrees of interaction between students and faculty. A student-based Advisory Council helps to guide the direction of the program and supports the related out-of-classroom activities (e.g., film series and study hall hours). Admission to the program uses a “cohort model,” where groups of students begin and are oriented to the program on a yearly basis. Collectively, these features of FEPPS result in a program characterized by a high degree of rigor and collegiality.

Participants and Procedures

In December, 2019, we collected quantitative survey data from 58 incarcerated students enrolled in FEPPS who were at varying stages of degree completion. As noted above, students enrolled in this program took in-person courses that possessed the characteristics of a traditional liberal arts education, including a heavy emphasis on writing, reading original texts, argument development, critical analysis, class participation, small class sizes, high degrees of interaction between students and faculty, and personalized attention to their work.

As part of a semester-end, all-student meeting, we explained the aims of the program evaluation, and invited all students to participate in the survey designed to measure various indices of academic and personal well-being. We clarified that they would not receive any compensation for their participation which was voluntary, and we provided an alternative activity to those who opted not to complete the survey. In advance of this meeting, we attended classes to explain the nature and aims of the evaluation project in an effort to encourage robust meeting attendance. Of approximately 125 students who were eligible, 60 attended the meeting. Of these, all but two students chose to complete the survey (mean age = 37.3 SD = 9.7).

Approximately 16 months later in April, 2021, we returned to this same group of FEPPS-enrolled students and asked them to participate a second time by completing the same survey of academic and personal well-being. Due to the COVID-19 pandemic, we were unable to engage in the earlier procedures of survey administration. Instead, we corresponded with students via the FEPPS correspondence paper distribution and return system. As before, we highlighted the survey's voluntary nature and assured complete participation confidentiality. Of the approximately 125 surveys distributed, we obtained 37 completed surveys; 17 of the 37 surveys came from students who had also participated at Time 1. In advance of all data collection, we obtained approval from the Department of Corrections, the Washington State Institutional Review Board, and our own university's IRB.

To provide a comparison sample, in the fall of 2019, we collected data via MTurk from 515 non-incarcerated participants who responded to our survey “HIT” (i.e., the name given by Amazon to work opportunities), utilizing the same survey questions related to academic and personal well-being as with the incarcerated student respondents. In an effort to obtain data that matched our sample of incarcerated students, we requested that only female-identifying persons who had not earned a four-year college degree participate in the study. We also explained that they would be paid $0.75 for their participation, that the survey would take approximately 10 minutes, and that the questions would relate to academic priorities, adjustment to change, coping style, family support, and feelings about their future. Upon receipt of the data, we deleted responses from those who indicated that they had earned a degree higher than an associate's degree on either of the two educational background screening questions (n = 93), or who indicated a gender other than female on either of the two gender-identifying questions (n = 65). We also deleted responses from those who answered the survey in under three minutes (n = 90), a time approximately one-third the length we estimated the average respondent would need to complete the survey. This resulted in useable comparison data from 337 respondents (mean age = 37.3, SD = 12.2). We obtained approval to conduct this data collection phase from our university's IRB.

Materials

The measures reported in this article were part of a longer survey that also included background demographic questions (e.g., education complete, future educational plans) and, in the case of the incarcerated sample, questions related to their college course experiences. The 11 measures of personal well-being, coping, and academic engagement (see Table 1) consisted of multi-item Likert-type statements where respondents indicated their answer using a 5-point scale anchored from definitely false (1) to definitely true (5), or in the case of the general advocacy-behavior scale, never (1) to more than 10 times (5). The items were drawn from existing published scales in the literature, but shortened in order to fit the available time interval for data collection and modified in order to fit the needs and characteristics of an incarcerated sample.

Table 1.

Description of Personal Well-Being, Coping, and Academic-Engagement Measures.

Construct	Variable	Focus of measure	Example item (number of items)	Reliability	Source
Personal well-being	General self-esteem	Overall feelings of one's worth	I have a positive opinion of myself. (5)	0.82	Debowska et al. (2017) three items (modified directions) Horon et al. (2018)—2 items
	Family belongingness	Amount of connection, support from family members.	In our family, we really help and support each other. (6)	0.88	Horon et al. (2018)—2 items Fok et al. (2014)
	General belongingness	Degree to which fits in, has connections with others.	People tend to accept me as I am. (6)	0.79	Horon et al. (2018)—1 item Malone et al. (2012)—2 items Alonso-Tapia et al. (2017)—2 items
	Hope in the future	Degree to which one has positive beliefs about the future	The future appears bright to me. (4)	0.79	Shumway et al. (2014)—2 items Written for study—2 items.
Coping-related	Embracing change	One's willingness to try new behaviors, experiences.	I enjoy finding new ways of working out problems. (4)	0.64	Alonso-Tapia et al. (2017)—1 item Greco and Roger (2001)—3 items
	General advocacy-confidence	Belief regarding one's ability to cope in situation requiring interpersonal interactions.	I can work with other people to solve problems. (8)	0.81	Hawley et al. (2016)
	General advocacy-behavior	Number of times in the past year one has engaged in healthy types of interactions to meet needs. (modified time frame from three months to one year)	In the past year, how many times have you worked together with someone else to solve a problem in your life? (5)	0.79	Selected items from Hawley et al. (2016)
	Locus of control	Belief regarding whether outcomes are a function of luck or effort.	When a problem arises, I often think that there is little I can do about it. (4)	0.61	Gregg et al. (2017)—1 item Wang and Su (2013)—2 items Alonso-Tapia et al. (2017)—1 item
	Proactive coping	One's capacity to enact solutions to problems.	I reach out to others in times of need. (7)	0.72	Gregg et al. (2017)—1 item Shumway et al. (2014)—3 items Alonso-Tapia et al. (2017)—3 items
Academic engagement	Subject-focused engagement	Degree to which one is engaged and satisfied with math, science, and writing.	I find learning science interesting. (7)	0.66	Gregg et al. (2017)—5 items Curtis and Trice (2013) 1 item written for study—1 item
Academic engagement	Academic prioritization	Degree to which school is an important focus	I can easily be talked out of studying. (3)	0.58	Curtis and Trice (2013)

We developed paper surveys for participants who were part of the incarcerated sample; MTurk respondents were provided with a link that took them to a Qualtrics electronic survey. Upon receipt of the data, we entered (incarcerated sample) or downloaded (MTurk) the responses into SPSS. As needed, items were reverse-coded and then summed to arrive at total scores for each participant.

Results

To examine the impact of FEPPS enrollment on the outcome measures, we ran a number of analyses, both within the incarcerated sample as well as between the incarcerated and MTurk comparison sample.

Comparison Among Incarcerated Students

Turning first to analyses that involved only the incarcerated sample at Time 1, we examined the relationship between program completion measures and the total scores for the measures related to personal well-being, coping, and academic engagement. Correlations between number of academic credits earned in the college-in-the-prison program were significant for measures of academic prioritization (r = .27, p < .05) and subject-specific self-esteem (r = .30, p < .05). Graduates of the program (i.e., those who had earned their associate degree, n = 14) compared to new and continuing students (n = 44) revealed higher scores on all measures; however, most failed to reach standard levels of statistical significance. Effect sizes of these differences ranged from 1% to 8%.

Due to the small sample size of only 17 FEPPS-enrolled participants who completed both Time 1 and Time 2 surveys, we were limited both in our analytical options and their statistical power. However, we did conduct a number of simple comparisons in an effort to describe general trends in the data. Comparing program graduates (n = 5) to new and continuing students (n = 12), we found higher reports of well-being, coping, and academic engagement for all measures at Time 2 for program graduates, significantly so for the measures of General Self-Esteem, t (15) = 2.16, p < .05 and Hope for the Future, t (15) = 2.81, p < .05.

Unfortunately, in addition to the small sample size, over-time comparisons were also complicated by the substantial impact of the COVID-19 pandemic. Specifically, visitors were no longer permitted to the prison, movement within the prison was substantially curtailed, and nearly all volunteer-based educational, recreational, training, and therapeutic programming was paused in an effort to reduce transmission of the virus. Although FEPPS courses continued during this time, in-person delivery of these courses was no longer permitted and, subsequently, shifted to paper-based correspondence.

Under such circumstances, we anticipated some loss in students’ reports of well-being, coping, and academic engagement. However, examination of Time 1 to Time 2 means using paired-sample t-tests revealed only a significant decline in Family Belongingness, t (16) = 2.70, p < .05. Although this finding is unsurprising given the inability of students to see family members during the 14 months preceding their Time 2 survey responses, we were heartened to see the relative stability of the other well-being, coping, and academic engagement measures over time, discussed below.

Comparisons Between Incarcerated and Non-Incarcerated Students

Next, we ran a number of analyses that compared well-being, coping, and academic engagement between the Time 1 incarcerated and comparison (MTurk) samples. Moreover, for analyses that involved such comparisons, we divided the MTurk sample by levels of educational experience and by the previous incarceration in order to better understand the ways in which incarcerated students may have benefitted from the educational program.

Table 2 presents the means, F-values, and post-hoc comparisons divided by their levels of earned education. (Even though mean age for the two samples was identical, we controlled for age in all analyses, as this systematically varied within the comparison sample as a function of educational level.) As shown, the incarcerated sample's scores were significantly greater than the comparison group's mean scores for most measures. Of specific note, post-hoc comparisons illustrate that mean scores for general self-esteem, hope in the future, willingness to embrace change, general advocacy-behavior, proactive coping, and academic prioritization demonstrated the same pattern; namely, that in all cases, the incarcerated sample reported significantly greater levels than did the non-incarcerated sample. That is, for a given level of educational attainment, the incarcerated subsample reported greater levels of well-being than did a non-incarcerated sample at that same educational level. The measure of locus of control reflected a similar pattern, although, in addition, we found significant differences between the MTurk sample that had versus those not completed some college or technical school as compared to those who had graduated from high school. Subject-specific self-esteem was the single variable measured whose pattern revealed a different trend. In this case, the incarcerated sample who had graduated with an associate's degree reported significantly higher levels than all other groups.

Table 2.

Scale Means (SEs), F-Values, and Post-hoc Comparisons Between the Incarcerated and Comparison Samples: as a Function of Completed Educational Level.

	Personal well-being				Coping-related					Academic engagement
Comparison groups	General self-esteem	Family belongingness	General belongingness	Hope in future	Embrace change	General advocacy-confidence	General advocacy behavior	LOC	Proactive coping	Subject-focused SE	AcadPrioritz
High school or below—Mturk (1) n = 94	19.32 (0.49)	23.59 (0.63)	23.47 (0.55)	15.26 (0.40)	13.57 (0.34)	30.70 (0.64)	15.48 (0.45)	13.26 (0.32)	24.43 (0.53)	20.25 (0.53)	9.34 (0.26)
Some college or tech school—Mturk (2) n = 165	19.83 (0.37)	22.88 (0.48)	23.40 (0.42)	15.30 (0.30)	14.19 (0.25)	32.68 (0.48)	16.24 (0.35)	14.28 (0.24)	24.76 (0.40)	22.72 (0.40)	9.28 (0.20)
AA degree—Mturk (3) n = 74	19.41 (0.55)	22.96 (0.71)	23.45 (0.62)	15.66 (0.45)	14.25 (0.38)	32.64 (0.72)	15.89 (0.51)	14.44 (0.36)	24.81 (0.60)	23.31 (0.60)	9.67 (0.30)
Some college—FEPPS (4) n = 44	22.19 (0.72)	23.10 (0.92)	24.84 (0.82)	18.57 (0.59)	15.75 (0.50)	32.27 (0.95)	19.61 (0.67)	17.20 (0.48)	27.79 (0.78)	21.61 (0.77)	11.65 (0.39)
AA degree—FEPPS (5) n = 14	22.97 (1.27)	25.13 (1.64)	26.90 (1.42)	18.88 (1.04)	17.14 (0.87)	34.21 (1.66)	20.52 (1.18)	17.99 (0.84)	30.63 (1.47)	26.28 (1.37)	12.52 (0.68)
F-value (df = 4, 385)	4.5	0.585	1.95	8.8	6.01	2.07	10.09	16.16	7.12	6.94	12.01
Significance level	0.001	ns	ns	0.0001	0.0001	ns	0.0001	0.0001	0.0001	0.0001	0.0001
Post-hoc tests	1, 2, 3, < 4, 5	NA	NA	1, 2, 3, < 4, 5	1, 2, 3, < 4, 5	NA	1, 2, 3, < 4, 5	1 < 2, 3, 4, 5; 2, 3 < 4, 5	1, 2, 3, < 4, 5	1 < 2, 3, 5; 1, 2, 3, 4, < 5	1, 2, 3, < 4, 5
Effect size (adj)	5.90%	1.00%	3.40%	7.40%	5.30%	3.60%	8.40%	15.40%	7.90%	6.20%	12.30%

Interestingly, within the MTurk sample, a small subset of respondents reported that they had been incarcerated at some point in the past while they were presently not incarcerated. We, therefore, divided the MTurk comparison sample into those who formerly had been incarcerated (n = 20) versus had never been incarcerated (n = 311), and compared these two groups to the presently incarcerated sample. Table 3 provides these findings. Turning to the post hoc tests, we found that the currently incarcerated sample was significantly higher than both the never and formerly incarcerated respondents for willingness to embrace change, general advocacy-behavior, locus of control, proactive coping, and academic prioritization. For the measures of general self-esteem, general belongingness, and hope in the future we found that the incarcerated sample was significantly higher than those never incarcerated who, in turn, were significantly higher than those who had been previously incarcerated.

Table 3.

Scale Means (Standard Error), F-Values, and Post-hoc Comparisons:as a Function of Incarceration Status.

	Personal well-being				Coping-related					Academic engagement
Comparison groups	General self-esteem	Family belonging	General belongingness	Hope in future	Embrace change	General advocacy-confidence	General advocacy-behavior	LOC	Proactive coping	Subject-focused SE	AcadPrioritz
MTurk—never incarcerated (0) n = 311	19.74 (0.27)	23.36 (0.35)	23.71 (0.30)	15.56 (0.22)	14.03 (0.19)	32.31 (0.35)	15.96 (0.25)	14.07 (0.18)	24.83 (0.29)	22.09 (0.30)	9.38 (0.15)
MTurk—previously incarcerated (1) n = 20	17.43 (1.06)	19.58 (1.36)	20.07 (1.17)	12.65 (0.86)	13.73 (0.73)	29.52 (1.39)	15.86 (0.99)	13.76 (0.71)	22.84 (1.14)	23.13 (1.24)	9.58 (0.59)
FEPPS presently incarcerated (2) n = 58)	22.39 (0.62)	23.59 (0.80)	25.36 (0.70)	18.65 (0.51)	16.10 (0.43)	32.76 (0.82)	19.83 (0.58)	17.40 (0.42)	28.41 (0.68)	22.74 (0.69)	11.88 (0.34)
F-value (df = 2, 387)	10.76	3.76	7.64	23.12	10.02	2.13	18.98	27.29	14.01	0.65	22.69
Significance level	0.0001	0.024	0.001	0.0001	0.0001	ns	0.0001	0.0001	0.0001	ns	0.0001
Post-hocs	2 > 0 > 1	1 < 0, 2	2 > 0 > 1	2 > 0 > 1	2 > 0, 1	NA	2 > 0, 1	2 > 0, 1	2 > 0, 1	NA	2 > 0, 1
Effect size (adj)	7.20%	2.80%	5.70%	10.20%	4.90%	NA	8.40%	13.80%	8.20%	NA	12.20%

Discussion

As noted at the outset of this article, the extant literature, often using anecdotal or qualitative claims, has frequently connected the benefits and utility of CIP programs to reduced recidivism (Cho & Tyler, 2013; Davis et al., 2013; Duwe & Clark, 2014; Erisman & Contardo, 2005; Gould, 2018; Kim & Clark, 2013; Vacca, 2004). We believe this outcome has merit; however, in this article, we sought to expand this evidence to include more systematic measurement and forms of comparisons as ways to offer additional perspectives for understanding CIP program impact. Namely, we demonstrated several methods for achieving such comparisons: (1) within a given sample, comparing students at varying levels of degree completion; (2) overtime comparison of the same students; and (3) comparison of an incarcerated and non-incarcerated sample, matched for different types of experiences (e.g., educational level, incarceration). Without question, each form of comparison had shortcomings and limitations; however, the combined efforts of these approaches begin to paint a more complete picture—one that suggests that high-quality CIP opportunities confer multiple benefits to students in the areas of well-being, coping, and academic engagement. Such evidence is critical. Although CIP program continuation is oftentimes threatened; the COVID-19 pandemic has had an additional, deleterious effect on higher education in prisons. As volunteer-powered programs are slowly reinstated following the pandemic's end, it may be additionally important to demonstrate program benefits with data that are robust and generalizable.

Our findings illustrate such benefits. Within the incarcerated sample, we found modest improvements as students progressed through the program and completed more courses. Although many of these changes failed to reach statistical significance, this was likely due to lower statistical power given the relatively small sample of FEPPS students (making within-group comparisons difficult) and our inability to measure students’ well-being, coping, and academic engagement before they entered the program (and thereby failing to capture initial improvements). Nonetheless, the patterns obtained were consistent across all variables and demonstrated a steady pattern of improvement as students stayed in the program.

This pattern was further confirmed by comparing graduates to non-graduates at Time 2 and examining the overtime changes from Time 1 to Time 2 in the small subset of 17 participants who responded to both surveys. Noted earlier, we found only a decline in Family Belongingness moving from Time 1 to Time 2, although we had anticipated declines in nearly all measures of well-being, coping, and academic engagement due to the profound effects of the COVID-19 pandemic on the day-to-day life of incarcerated students in general and their academic experiences in particular. Failure to find much decrease may stem from a lack of statistical power; however, a simple inspection of mean scores showed nearly no changes in total scores from Time 1 to Time 2. It is also possible that students derived a sense of well-being and capacity to cope from other sources, not just their educational experiences. However, we believe that it is equally plausible that students’ well-being, coping, and academic engagement were maintained with even the limited contacts with FEPPS staff and coursework, especially if the changes students experienced occurred earlier in their academic history and were more permanent. That is, the relative stability of these measures overtime, even in the midst of a profoundly difficult pandemic year, may signal that the educational impacts occurred earlier in the student's educational development and were resilient and robust.

The analyses that compare FEPPS and MTurk (non-incarcerated) participants lend further credence to our assertion that participation in CIP confers important and meaningful impacts. Turning first to Table 2, across nearly all variables when the comparisons are statistically significant, FEPPS students (Groups 4 and 5) report higher levels of personal well-being, coping, and academic engagement as compared to the MTurk comparison sample, regardless of education completed. More specifically, comparing only FEPPS and MTurk respondents at similar educational levels demonstrates that FEPPS students report higher levels of these same variables (analyses not shown). For example, comparisons of only AA graduates find that FEPPS graduates are significantly greater than MTurk AA graduates on the measures of general self-esteem, general belonging, hope in future, embracing change, general advocacy-behavior, locus of control, proactive coping, subject-specific self-esteem, and academic prioritization. Likewise, we find that comparisons of those who have “some college” FEPPS students are significantly greater than MTurk participants on the measures of general self-esteem, hope in future, embracing change, general advocacy-behavior, locus of control, proactive coping, and academic prioritization.

Turning to Table 3, we also found that respondents in the FEPPS sample reported better academic and personal well-being outcomes than respondents in the formerly or never- incarcerated comparison group (Table 3). Especially striking, we found that measures of well-being and coping were notably better for those who are presently incarcerated (FEPPS students) as compared to those MTurk respondents who were formerly incarcerated. Whether that speaks to the educational benefits for FEPPS students as compared to the education formerly incarcerated women received outside of prison, or to the poor support available to formerly incarcerated women living outside of prison, cannot be determined from our data.

Although this MTurk comparison group was not a control group per se, the fact that currently incarcerated students reported levels of well-being, coping, and academic engagement that were either equal to or better than students at similar educational levels (Table 2) or those who were formerly incarcerated (Table 3) lends credence to the assertion that education in the prison confers benefits in these areas. Our interpretation of how or why this is the case is, however, limited. It may be that such a benefit exists for students who receive in-person education characterized by small class sizes and rigorous standards. It is also the case that we did not measure the educational experiences of our MTurk comparison sample. Assuming that they reflected the typical population, it is likely that these respondents’ experiences comprised both in-person and online coursework, large and small class sizes, and varying degrees of rigor. Future research that seeks to use such forms of comparison would be well-served by collecting such data so that educational program characteristics could be examined and compared more closely. Nevertheless, given that much of the extant literature demonstrates significantly poorer mental health and well-being among incarcerated persons, especially women, our findings are encouraging.

Gender, Higher Education, and Well-Being

As noted earlier, incarcerated women are prevalently survivors of violence, a long-standing social issue that contributes to the complexity of the overall picture. Research from over two decades ago by the Human Rights Watch and the U.S. Department of Justice indicated that nearly half of all incarcerated women had been physically or sexually abused prior to their imprisonment, and over one-third had been abused prior to the age of 18—roughly twice the percentage among the general population of American women (Harlow, 1999). More recently, the Compassion Prison Project (2021) reported that incarcerated women experienced three times the rate of physical and sexual abuse when they were children as compared to the general population. In fact, sexual abuse is one of the primary predictors of girls’ entry into the juvenile justice system, a fact which the Human Rights Project for Girls and the Georgetown Law Center on Poverty and Inequality recently framed as “the sexual abuse to prison pipeline” (Saar et al., 2015). With its focus on community, choice, and support for individuals, higher education in prison programs may provide what some trauma-informed practitioners consider as the core values of trauma-informed services in prison: safety, choice, collaboration, and empowerment. (Kubiak et al., 2017). The fact that the present study found higher levels of self-esteem may be the result of the choice, agency, and collaboration they experience in the FEPPS program. Students can elect their courses; they participate in an environment where disagreement and debate are valued, and they must collaborate and support one another to succeed. We believe that students’ successful completion of challenging courses and degree attainment is a tangible way to address some of the effects of trauma and violence and strengthen the mental health of students. These experiences may offer promising directions for researchers seeking to link education to soft skills that also address the traumatic pathways that lead to incarceration and the trauma that students experience in prison.

Study Limitations and Future Directions

We acknowledge that response bias and self-selection into the program may account for our findings. Even as we emphasized that the study's results would have no bearing on the program or their enrollment, it is possible the FEPPS students may have been motivated to report that they experienced better outcomes than they actually had because they wanted to ensure that the program itself would not be perceived poorly. We cannot rule out this interpretation entirely. However, the type of in-class experiences and education opportunities FEPPS students received did provide them with skills aimed at changing their way of understanding and interacting with others and their environment. That is, they practiced “hard” (e.g., using evidence to construct arguments) and “soft” (time management) skills directly connected to their capacities to advocate for themselves and navigate complex situations, and they experienced success with academic challenges and had the support of peers and faculty while doing so.

In addition, we acknowledge that the measures were heavily modified for use in this study. To this concern, we note that internal consistency reliability estimates for each of the measures, as well as individual item-to-total correlations, were acceptable and nearly identical between the incarcerated and non-incarcerated respondents. This provides some evidence for both homogeneity of item content within each measure and similar perception of item content by participants in both the incarcerated and non-incarcerated samples. (Limited sample size for our incarcerated group precluded more advanced analyses of factor structures.) Future research aimed at providing additional psychometric evidence would benefit the field.

Last, the disruption of the educational programming and unprecedented challenge of daily living, restricted interactions, and generalized fear stemming from the COVID-19 pandemic negatively impacted data collection and, perhaps, students’ responses. We anticipate collecting data from students again in the future once the COVID-19 impacts are less and in-person educational programming has resumed.

Even with these limitations, however, we believe that the use of common, quantitative measures and efforts to include a comparison sample stand to bolster the evidence for CIP's positive impacts. We also recommend that scholars continue to consider the broad range of benefits and find ways to link them to important types of involvements such as work, family life, and societal participation, both inside and outside of prison. Although we believe that these well-being-related outcomes are important in their own right, we suspect they play an important mediating role in post-release adjustment. We encourage future scholarship that examines these connections.

Footnotes

Acknowledgements and Credits

All data were collected with the approval of the Department of Corrections in additional to the Washington State and University of Puget Sound institutional review boards.

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. We thank the Olson Foundation for funding the data collection reported in this article.

Author Biographies

Sarah Y. Moore, PhD, is a professor and chair in the department of psychology, University of Puget Sound. She has taught courses as part of the Freedom Education Project Puget Sound (FEPPS) Program and also conducts research in the area of employee well-being.

Tanya Erzen, PhD, is the academic and faculty director of FEPPS and an associate research professor of religion, University of Puget Sound. She is a member of the Steering Committee for the National Organization for Higher Education in Prison.

References

Allard

(2002). Life sentences: Denying welfare benefits to women convicted of drug offences. The Sentencing Project. https://www.opensocietyfoundations.org/publications/life-sentences-denying-welfare-benefits-women-convicted-drug-offenses

Alonso-Tapia

Garrido-Hernansaiz

Rodriguez-Ray

Ruiz

Nieto

(2017). Personal factors underlying resilience: Development and validation of the Resiliency Questionnaire for Adults. International Journal of Mental Health Promotion, 19(2), 104–117. https://doi.org/10.1080/14623730.2017.1297248

Baranger

Rousseau

Mastrorilli

M. E.

Matesanz

(2018). Doing time wisely: The social and personal benefits of higher education in prison. The Prison Journal, 98(4), 490–513. https://doi.org/10.1177/0032885518776380

Cho

R. M.

Tyler

J. H.

(2013). Does prison-based adult basic education improve post-release outcomes for male prisoners in Florida? Crime & Delinquency, 59(7), 975–1005. https://doi.org/10.1177/0011128710389588

Compassion Prison project.org (2021, March 14). Childhood trauma statistics. Compassion Prison Project. Retrieved January 2, 2022, from https://compassionprisonproject.org/childhood-trauma-statistics/

Curtis

N. A.

Trice

A. D.

(2013). A revision of the academic locus of control scale for college students. Perceptual and Motor Skills, 116(3), 817–829. https://doi.org/10.2466/08.03.PMS.116.3.817-829

Davis

L. M.

Bozick

Steele

J. L.

Saunders

Miles

J. N. V.

(2013). Evaluating the effectiveness of correctional education. RAND Corporation; Bureau of Justice Assistance, U.S. Department of Justice. https://www.rand.org/content/dam/rand/pubs/research_reports/RR200/RR266/RAND_RR266.sum.pdf

Debowska

Boduszek

Sherretts

(2017). Self-esteem in adult prison population: The development and validation of the self-esteem measure for prisoners (SEM-P). Deviant Behavior, 38(11), 1240–1251. https://doi.org/10.1080/01639625.2016.1246034

Duwe

Clark

(2014). The effects of prison-based educational programming on recidivism and employment. The Prison Journal, 94(4), 454–478. https://doi.org/10.1177/0032885514548009

10.

Erisman

Contardo

J. B.

(2005). Learning to reduce recidivism: A 50-state analysis of postsecondary correctional education policy. Institute for Higher Education Policy.

11.

Evans

D. N.

Pelletier

Szkola

(2018). Education in prison and the self-stigma: Empowerment continuum. Crime & Delinquency, 64(2), 255–280. https://doi.org/10.1177/0011128717714973

12.

Fok

C. C. T.

Allen

Henry

, & People Awakening Team, University of Alaska Fairbanks. (2014). The brief family relationship scale: A brief measure of the relationship dimension in family functioning. Assessment, 21(1), 67–72. https://doi.org/10.1177/1073191111425856

13.

Gould

M. R.

(2018). Rethinking our metrics: Research in the field of higher education in prison. The Prison Journal, 98(4), 387–404. https://doi.org/10.1177/0032885518776375

14.

Greco

Roger

(2001). Coping with uncertainty: The construction and validation of a new measure. Personality and Individual Differences, 31(4), 519–534. https://doi.org/10.1016/S0191-8869(00)00156-2

15.

Gregg

Galyardt

Wolfe

Moon

Todd

(2017). Virtual mentoring and persistence in STEM for students with disabilities. Career Development and Transition for Exceptional Individuals, 40(4), 205–214. https://doi.org/10.1177/2165143416651717

16.

Harlow

(1999). Prior abuse reported by inmates and probationers. Bureau of Justice Statistics. Retrieved from https://bjs.ojp.gov/library/publications/prior-abuse-reported-inmates-and-probationers

17.

Harnish

(2019). Philosophical implications of taxpayer funding for prison education. Critical Education, 10(5), 1–14. https://doi.org/10.14288/ce.v10i5.186298

18.

Hawley

Gerber

Pretz

Morey

Whiteneck

(2016). Initial validation of personal self-advocacy measures for individuals with acquired brain injury. Rehabilitation Psychology, 61(3), 308–316. https://doi.org/10.1037/rep0000093

19.

Horon

Williams

S. N.

McManus

Roberts

(2018). The Culture and Protective Suicide Scale for Incarcerated Persons (CAPSSIP): A measure for evaluating suicide risk and protection within correctional populations. Psychological Services, 15(1), 45–55. https://doi.org/10.1037/ser0000197

20.

Kim

R. H.

Clark

(2013). The effect of prison-based college education programs on recidivism: Propensity score matching approach. Journal of Criminal Justice, 41(3), 196–204. https://doi.org/10.1016/j.jcrimjus.2013.03.001

21.

Kubiak

S. P.

Covington

S. S.

Hillier

(2017). Trauma-informed corrections. In Springer

Roberts

(Eds.), Social work in juvenile and criminal justice systems (4th ed., pp. 1–28). Charles C. Thomas.

22.

Lapidus

Luthra

Verma

Small

Allard

Levingston

(2005). Caught in the net: The impact of drug policies on women and families. American Civil Liberties Union.

23.

Malone

G. P.

Pillow

D. R.

Osman

(2012). The General Belongingness Scale (GBS): Assessing achieved belongingness. Personality and Individual Differences, 52(3), 311–316.

24.

McCorkel

DeFina

(2019). Beyond recidivism: Identifying the liberatory possibilities of prison higher education. Critical Education, 10(3), 1–19. http://ojs.library.ubc.ca/index.php/criticaled/article/view/186316

25.

Morales

E. E.

(2012). Learning as liberation: How liberal arts education transforms first-generation/low socio-economic college students. Journal of College Student Retention: Research, Theory, and Practice, 13(4), 499–518. https://doi.org/10.2190/CS.13.4.e

26.

Pascarella

E. T.

Blaich

(2013). Lessons from the Wabash national study of liberal arts education. Change: The Magazine of Higher Learning, 45(2), 6–15. https://doi.org/10.1080/00091383.2013.764257

27.

Saar

Epstein

Rosenthal

Vafa

(2015). The sexual abuse to prison pipeline: The girls’ story. Center on Poverty and Inequality. Georgetown Law Center on Poverty and Inequality. https://rights4girls.org/wp-content/uploads/r4g/2015/02/2015_COP_sexual-abuse_report_final.pdf

28.

Seifert

T. A.

Goodman

K. M.

Lindsay

Jorgensen

J. D.

Wolniak

G. C.

Pascarella

E. T.

Blaich

(2008). The effects of liberal arts experiences on liberal arts outcomes. Research in Higher Education, 49(2), 107–125. https://doi.org/10.1007/s11162-007-9070-7

29.

Shumway

S. T.

Dakin

J. B.

Jordan

S. A. S.

Kimball

T. G.

Harris

K. S.

Bradshaw

S. D.

(2014). The development of the Hope and Coping in Recovery Measure (HCRM). Journal of Groups in Addiction and Recovery, 9(4), 280–293. https://doi.org/10.1080/1556035X.2014.969059

30.

Snell

T. L.

(1994). Women in prison, survey of state prison inmates, 1991. Bureau of Justice Statistics. https://bjs.ojp.gov/content/pub/pdf/SOSPI91.PDF

31.

Treatment Advocacy Center (2016, September). Serious Mental Illness (SMI) prevalence in jails and prisons. Office of Research and Public Affairs. https://www.treatmentadvocacycenter.org/storage/documents/backgrounders/smi-in-jails-and-prisons.pdf

32.

Vacca

J. S.

(2004). Educated prisoners are less likely to return to prison. Journal of Correctional Education, 55(4), 297–305. https://www-jstor-org-443.web.bisu.edu.cn/stable/23292095

33.

Wang

(2013). Longitudinal factor structure of general self-concept and locus of control among high school students. Journal of Psychoeducational Assessment, 31(6), 554–565. https://doi.org/10.1177/0734282913481651

34.

Women in Prison Project of the Correctional Association of New York (2002). Women in prison fact sheet. https://www.prisonpolicy.org/scans/Fact_Sheets_2002.pdf