Incarcerated fathers returning home to children and families: introduction to the special issue and a primer on doing research with men in prison.

Men are returning to family life following prison experience in record numbers. The accelerated rates of incarceration in the U.S. are contributing to strained federal and state budgets and disrupted family life in communities across the nation. It has been proposed that successful reentry into family life may positively influence rearrest rates. However, there is only sparse research about the familial aspects of the prison, reentry, and rearrest cycle. This study is a report of a pilot study that examines a methodological attempt to obtain data from men in prison about to be released and their partners. The results show that collecting data from the men was much easier than from their partners/spouses. Additionally, it was found that men's ideas about their relationships with their spouses and children may be unrealistic, ambiguous, and unclear. Suggestions are made about future research and research methods with this unique group.

Keywords: father involvement, incarcerated fathers, father-child contact, fathering identity

**********

The U.S. Department of Health and Human Services (DHHS) has inaugurated an important interagency initiative entitled From Prison to Home: The Effects of Incarceration and Reentry on Children, Families and Communities (National Institute of Corrections, 2002). This initiative began in 2001 and sought to explore effective strategies that could mitigate the negative influences of men/fathers returning home following incarceration. The Prison to Home effort was founded upon two key facts: first, there are increasing numbers of imprisoned mothers and fathers in the U.S.; second, many researchers, policymakers, and practitioners have noted that reentry-based programs are not only difficult to locate and identify but often do not include a family-relationships focus. For example, most prison reentry programs focus exclusively on job placement and drug treatment (Petersilia, 2003).

The papers found in this special issue of Fathering focus on men's return home following incarceration and the family and community context in which that return occurs. Typically men's prison experience and return to family life are viewed either from the larger institutional perspective, usually represented within the criminology and/or demography literature, or from the psychological perspective that focuses on the prisoner's mental well-being, such as incidence of depression, schizophrenia, and/or drug and alcohol addictions. This collection promotes the idea that understanding the nature and texture of familial relationships as the former prisoner returns home is a critical and important feature of understanding how he succeeds in that critical transition (i.e., future parole violations, arrest, and/or reimprisonment).

OVERVIEW OF THIS SPECIAL ISSUE

The high rate of father absence is usually attributed to the high rate of divorce and the birth of children to young, never-married mothers. However, there is increasing recognition that the number of children growing up without their fathers is also the result of the rapid increase in the number of men who are incarcerated in either prisons or jails. In the U.S. it is estimated that 1.5 million children have at least one parent in prison: in 94 percent of these cases, the prisoner is the father (Petersilia, 2003). Additionally, each year about 600,000 men are released from prison, and many will reconnect (or attempt to reconnect) with spouses, former spouses, and children (Travis & Wahl, 2005).

In the United Kingdom (UK), the Home Office estimates that there has been a 47% increase in the prison population since 1993 (Office of National Statistics, 2005). The estimated number of men in prison in the UK now exceeds 50,000. This increase was highlighted by a dramatic jump in the number of prisoners during the months of January to April 2002, when the prison population in the UK increased by more than 4,000. Government officials have suggested that a tougher penal regime has increased the UK's incarcerated population by 28,000 (over the past six years) and required the building of new prisons at an estimated cost of 4.5 billion [pounds sterling] and an additional running cost of 945 million [pounds sterling] (Office of National Statistics, 2005). In short, prisoners cost taxpayers millions of dollars each day. When they leave prison and return to family life, they are often a psychological and financial burden to their families. Some carry infectious diseases to their children and partners, and many are, once again, in trouble with law enforcement within a few months of release.

This linkage between men's prison experience and community and family wellbeing is critical because successful reintegration into family life is associated with lower recidivism (i.e., rearrest and subsequent incarceration) rates Petersilia, 2003). Additionally, successful return to family life may lessen the burden of state welfare programs and reduce the number of children who grow up in reoccurring cycles of poverty.

Therefore, the primary purpose of this collection of articles is to isolate factors that may contribute to successful reentry into family life following incarceration. The articles presented here help us understand the process of fathers' returning home. Each article represents a slightly different view of this important process, and a variety of methods are employed to illuminate the scope of this issue. All but the Roy and Dyson articles are the result of an ongoing research project sponsored by the School of Family Life, College of Family Home and Social Science, and Family Studies Center, all located at Brigham Young University. Data were collected at five different prisons, two in the U.S. and three in the UK. Similar data collection strategies were used in both places, and all data were collected using a student-mentored learning approach. That is, university funds were used to train and support undergraduate and graduate student researchers who collected the data at the prison locations.

First, the present article discusses several key methodological and strategic issues that confronted our research team studying men's attitudes about family life prior to exiting prison and returning to family life. Overall, it was affirmed that researchers could obtain access to men in prison and that these men would speak openly about family issues. Few researchers have tackled the problems of getting prisoners and their families to talk openly about their experiences. It was unclear whether prison men who claimed a child and claimed they were intending to be in contact with that child upon release would disclose private family information. In turn, we did not know whether the partners of these men would agree to participate, if they could be found, or whether they were willing to be reinterviewed some time after the man's release. Therefore, a primary purpose of this project was to develop clear procedures for building a useable and reliable database of this fragile and rather invisible population.

The second article in this series is theoretical. Dyer suggests in his piece that researchers can turn to an identity-symbolic interaction framework to help us understand the transitions men make as they enter prison (as a father) and then negotiate their retransformation following the prison experience.

Third, the Clarke et al. article is an attempt to replicate the project reported in the first article by Day et al. but in the UK. The prison system in the UK has several key differences from prisons in the U.S. For example, the UK has been more progressive in promoting an ideal of rehabilitation rather than punishment. The results of the Clarke et al. study revealed that men in "open prisons" had different expectations and levels of access to their children and partners.

The Bahr, Armstrong, Gibbs, Harris, and Fisher paper focuses on the parole system and its interface with community and familial factors related to successful prisoner reentry. Bahr et al. found that family, friends, and community resources work together to shape men's experiences as they return home.

Arditti, Smock, and Parkman provide a rich look at what they call the "dormant" period of fathering. By dormant, Arditti el al. mean that, for most men in prison, the ability to enact the fathering role is inactive, subdued, and ambiguous. These authors remind us that the stories of these men could help us more fully understand the complicated and delicate issues as experienced by a most vulnerable and fragile population. Researchers are only beginning to record and analyze the complex stories of these families.

Finally, the Roy and Dyson paper shows how men's contact after prison becomes an issue of "strategy." That is, women are typically very careful about allowing men to reenter family life. Men are forced to at least consider the "dual life" approach to post-prison life. This piece points again to the importance of understanding the identity transformation that occurs when fathers enter and then return from prison.

In sum, we know very little about how men are received back into family life. These few articles help us understand more about that important transition into family and community life. It is hoped that these articles will add to the growing literature about how men's return affects the women and children for whom they may have once been responsible.

BACKGROUND

The overall purpose of this paper is to introduce researchers to the issues, challenges, and strategies of conducting research with men in prison. Within that framework, there are four general topics covered. First, we introduce readers to the problem of fathers who are incarcerated. A quick perusal of family science journals reveals a surprising gap in literature about this important social phenomenon. Second, we discuss methodological issues and problems that surround working with this population. One of the reasons why there is so little family-based research in this area is that executing a research project within prison walls is daunting and the usual methodologies are not quite a good fit for this type of research. Third, we present preliminary data about couple and father connection of men who are about to be released and return home or reconnect with a spouse/partner and child. Last, we suggest several topics that can stimulate further research in this area.

THE PROBLEM

Family science is just beginning to recognize the overwhelming nature of the U.S. public policy practice of mass incarceration. Of the many challenges facing American society, one of the weightier is the problem of incarceration (Nurse, 2002; Petersilia, 2003). Never before in U.S. history have so many individuals been convicted and imprisoned. One of the unanticipated results of the "war on crime" and recent mandatory (or determinate) sentencing practices is the sheer numbers of men (and women) who have been arrested and detained in prison (Arditti & McClintock, 2001). In turn, these high arrest rates have fueled the dramatic jump in the numbers of men who are then released from prisons and return to family life. The Department of Justice (DOJ) estimates that 95 percent of the two million prisoners currently incarcerated will eventually be released and return to their families and former communities (Bureau of Justice Statistics, 2002). According to the DOJ, fully 635,000 were released in 2002, or about 1,700 each day.

This phenomenon of "mass imprisonment" is one of the most startling changes in U.S. culture over the past 25 years (Tonry & Petersilia, 1999; Pattillio, Weiman, & Western, 2004). Between 1920 and 1975, the general prison population (both state and federal) represented around .10 of 1 percent of the general population. However, since 1975 that percentage has risen substantially ever year. That percentage now is estimated to be near .70 of 1 percent of the population--a 700 percent rise in the past 50 years and a 400 percent rise over 25 years alone (Beck, Karberg, & Harrison, 2002). Of course, most prison inmates (90 percent) are male, and a very high percentage of those are African Americans. It is estimated that the current prison population of African Americans is seven times that of whites in prison. Shockingly, it is projected that 30 percent of black noncollege men will eventually find their way into prison at some point in their lives (Western & Pettit, 2002).

As Pattillo, Weiman, and Western (2004) show, imprisonment has shifted from a marginal to an ordinary phenomenon. That is, before the 1970s, prison was an extension of the criminal subculture and was directly linked to the crime world outside the prison: imprisonment occurred rarely and was usually restricted to a small number of people who had usually entered the crime subculture early in life. However, in the past 20 years or so, the prison experience has transmuted from an extraordinary and unusual experience to a normalized event, especially for minority men. This normalization has become so pervasive that passage to adulthood for minority youth who have little economic opportunity usually involves crime and prison (Pattillo, Weiman, & Western). According to Garland (2001) the phenomenon of mass imprisonment is so high for some groups that the shock waves and impact (economically, socially, and individually) are felt by the general population in undeniable ways. For example, even as we boast about decreases in social welfare roles and expenditures, the cost of mass incarceration soars. The cost to operate state and federal prisons is overwhelming: the Bureau of Justice Statistics estimates that current spending for prisons in the U.S. topped $50 billion in 2003 with the average prisoner costing taxpayers more than $20,000 per year during incarceration. One of the prison workers in this current study recently quipped that it would be cheaper to send "them all to Stanford."

At a more local and personal level, children of incarcerated fathers live in a high-risk and potentially emotionally disrupting and damaging environment during the period of fathers' incarceration and in the immediate months following their release and return to the family and community environment. More than three million U.S. children are in daily contact with someone, overwhelmingly fathers, who were recently released from prison, and more than 10 million children report ever having direct contact with a parent who has been in prison (Travis & Wahl, 2005).

Children of incarcerated parents have numerous problems related to emotional health, school performance, and general health and well-being. For an excellent review of the scant research conducted about children of inmates, see Hagan and Dinovitzer (1999). This article shows that, even though research about children of prisoners has been of interest since the early 1900s, there are few definitive studies that clearly demonstrate what effect men's incarceration has on children. Among the attendant child behavior problems usually cited are preoccupation with the loss of the parent, depression, separation anxiety, and interpersonal conduct disorder (Fritsch & Burkhead, 1981; Kampfer, 1995; Sack, 1977).

Families face other problems as well. When men return from prison, they may carry and transmit infectious diseases such as hepatitis C, tuberculosis, and HIV/AIDS. Many have been treated for depression and other forms of mental illness and have continuing problems with addictions (Hammett, Roberts, & Kennedy, 2001). Additionally, many seek to resume drug- and/or crime-directed lives, and many will return to substance abuse. While these men may initially be motivated to reconnect with women and children, they may bring with them a history of violence and an orientation to life that does not facilitate mother or child well-being generally. Additionally, there are very few transitional programs that assist men as they try to return home.

The familial aspect of prison life is important to understand because the issues facing couples are complicated and tie directly into the likelihood men will reoffend following release. Contrary to popular belief, most men in prison are incarcerated for a relatively short time (29 months in 1999), most are young (the average age of men in minimum security prisons is about 26), and most will attempt to reconnect with spouses, former spouses, partners, and children following release (Petersilia, 2003).

The issue of arrest and rearrest followed by imprisonment has a dramatic impact on federal and state budgets. These costs are felt not only through direct spending by state and federal corrections departments, but they are also felt in indirect ways through assistance programs (Nelson, Deess, & Allen 1999; Petersilia, 2003). In other words, the research suggests that, as levels of commitment, interpersonal responsibility, and couple stability increase, released prisoners are less likely to be rearrested as quickly or as frequently. Likewise, when the rearrest rates are lowered, there is potential for a significant financial benefit to state and federal assistance programs, decreases in state costs for prison facilities, and reductions in the number of children who grow up in reoccurring cycles of poverty.

Much of the dramatic increase in prison populations and the resulting scenario of thousands of formerly incarcerated men reentering community and family life can be traced to public policies that have spawned harsher and more uniform sentencing practices. For example, since the 1970s there have been significant shifts in sentencing guidelines, criteria for release, and subsequent decreases in state-funded programming for those about to be released (for an overview of the history of this issue, see Petersilia, 2003, and Travis & Wahl, 2005). With the advent of determinate sentencing (codified sentence terms for specific crimes), prisoners are rarely released as a function of merit. Some scholars argue that the practice and high economic costs of determinate sentencing have decreased the incentives and available money for reentry-related program development (Petersilia, 2003; Travis & Wahl, 2005).

Additionally, prison policies have been constructed to deter couple involvement and general family contact. These policy changes include limited contact times (Petersilia, 2003); cost-prohibitive phone charges, often more than $1 per minute (Lynch & Sabol, 2001); and increases in nonlocal incarceration in which inmates are moved to nonproximate locations that are often hundreds of miles from their homes, children, and spouses/partners. Nurse (2002) claims that these policies have been designed to create "deep breaks" from partners/spouses, families, and community life as a way to toughen the public's response to crime (p. 37).

CURRENT PILOT PROJECT

The pilot research project described in this journal issue sought to examine the beliefs and attitudes of incarcerated fathers who were about to be released and return to family life. Our approach was somewhat unusual in that it was family focused. Typically, prisoner reentry is discussed from the perspective of several disciplines within which familial issues are viewed as context. For example, criminologists frequently look at larger causes such as changes in national trends about incarceration and recidivism. Sociologists may examine how institutions respond to changes in political climate and the cultural context within which men experience prison life. Social psychologists and psychology-based clinicians, on the other hand, may look deep into the personal motivations of men in prison or focus on mental-health issues. To the extent family analyses are included in these studies, it is usually incorporated as one of several contextual variables and not as the fundamental social unit linking individuals to their larger communities. Few researchers, however, tackle the familial territory that lies between the institution and community.

First, this pilot research sought answers to several methodological questions: would prison staff allow access by our research team (most of whom were trained undergraduate research assistants)? Would incarcerated men be willing to talk openly about family life? How difficult would it be to conduct a short telephone interview with their identified spouse/partner? Would these partners be willing to talk about the imminent release of the child's father, the well-being of the child, and their relationship with the incarcerated father? How difficult is it to find and interview the participating men several months following release? Our experience with these issues is reported below.

Second, we sought to develop and revise measures that could be used in assessing couple relationships for this unusual couple situation. Most research on couple conflict, satisfaction, and general well-being assumes that the couple has (or could have) some daily interaction. Even when studying nonresident fathers, the measures still attempt to assess frequency of contact and interaction on a regular basis. That basic assumption is invalid when considering the incarceration situation, where there are intentional and unintentional institutional external barriers that make contact extremely difficult between incarcerated men and their partners. Our second activity, therefore, was to create some measure of couple well-being or connectedness that would make sense in this unique environment. Our success with this issue was limited. We experienced significant difficulty in obtaining responses from the spouses/partners in the U.S. sample. However, we did have more success in contacting the partners in the UK sample in the article by Clarke et al. (this issue), and data about couple issues are reported there.

Our next goal was to find out what these men's attitudes and beliefs were about their families and their upcoming release. We had hoped to compare those attitudes and ideas with those of their partners. As will be reported, we were able to interview so few of the women that this activity failed, and we report only preliminary findings about men's feelings about their children, visitation issues, and connectedness to their partners.

Finally, we were hoping to assess in a follow-up interview whether or not the father was able to reconnect with his partner and child and assess the difficulty of that transition. This activity also failed. Following men after prison (at least from our experience) was extremely time consuming and, in some cases, impossible. Such follow-ups far exceeded the resources available for our pilot study. At the end of this article, we suggest alternative methods for future research that may ensure greater success in this regard. As will be reported later, however, we have been able to recontact the prison systems in Oregon and Utah and do have a rough idea of the recidivism rate for the 51 men we initially interviewed prior to their release. The following is a report of each of these activities.

METHODOLOGICAL ISSUES

Conducting research with prison populations is difficult. For this pilot study four of the five prisons (all in the U.S.) were minimum-security prisons where the risk of potential security problems was very minimal. Even so, the first two significant barriers to performing this type of research were securing approval from both the prison systems and from the university Institutional Review Board (IRB). We began by securing the research agreements with both Oregon and Utah departments of correction and simultaneously obtaining approval from the IRB. In our case, these tasks proved to be particularly difficult. Prison populations are a group protected by federal law, and universities and state boards of correction are typically leery about research efforts involving these kinds of groups because of the increased potential for litigation. Our experience with the Oregon prison system was extremely facilitative, and staff were unusually helpful.

A multistep training process was used to train graduate and undergraduate students who administered the interviews within the prison and conducted telephone interviews with the partner women. The training for the students, travel funds, and incentive costs to inmates and spouses were provided by a university-sponsored mentorship grant. A team of 13 students helped to develop the questionnaire, prepare the IRB documents, interview prisoners, and code and analyze data.

Additionally, the Utah prison required an onsite training workshop during which prison staff members were taught about the hazards of prison interviews, laws that can be inadvertently broken, and a host of safety issues. Our interviewers worked in pairs and in an assigned room within the prison. One unexpected problem that may have altered the data collection was that, during some shifts in the Utah prison, guards insisted on being present (albeit at a distance) during the interview. In other situations, we were allowed to be in a conference-style room with the prisoner alone. Interviewers commented that the presence of the guard seemed to dampen responses and conversation. Our experience with this feature of interviewing in prison reminds us that the setting during which interviews are conducted should be controlled as much as possible. To our knowledge, there is no extant research that systematically examines prisoner responses in different settings.

Additionally, the IRB regulations and our sensitivity to the staff needs of the prison limited our ability to actually recruit volunteers in a truly systematic way. Staff printed out lists of potential interviewees who would be released within our designated time flame (three to six weeks from the time of interview). Occasionally, a prisoner who would be a suitable candidate (a non-sex offender about to be released who reported he was a father) was out on work release, ill, or otherwise unable to be interviewed. For the U.S. sample, data collection began in the fall of 2003 and continued through early spring of 2004. The data for the UK sample (reported in Clarke et al. [this issue]) were collected in the summer of 2004.

Sample. There were significant hurdles to overcome in conducting in-prison research (for an overview of sampling within prisons, see research by Pope, Lovell, & Brandl, 2001). Obviously, the prison is a closed system. But within this closed system is a very dynamic and complicated organization. Once the guardian prison officials give approval, one encounters the real world of the staff workers who actually make the prison operate. While the staff may have been informed and asked to cooperate, how they cooperate and facilitate one's research agenda can be another matter. In our situation (Utah, Oregon, and the UK sample reported elsewhere in this collection), we were very fortunate to have a high level of cooperation by the staff. However, that cooperation had its limits. Prisoners often work outside the prison and are not always in the facility. Frequently, prison blocks or even the whole prison can be "locked down" for some unforeseen incident. In the case of our work within the UK prison system, the initial prison from which we were scheduled to collect data was locked down for an entire month beginning the day we were to start data collection.

We had originally hoped we could have more information about who was agreeing to participate and who was not. In the end, we were only marginally confident about our "return rate." We were unclear in some cases what the attending guards said to men as flyers were handed out and men were recruited. In three cases, men came to the interviews and filled out all of the questions, and only then did we find out (from case workers) that their answers to the screening questions were bogus and they, in fact, had no children at all.

At the facility in Utah, we conducted 33 useable interviews over a six-week period in the fall of 2003 and winter of 2004. We estimate that only six eligible inmates refused our request to complete an interview. Eleven others were disqualified because they were sex offenders, had repeated out-of-prison work schedule conflicts, or had no children. Likewise, in the Oregon prison site, 18 interviews were completed, and there were five prisoners who refused. Therefore, we calculate our participation rate at roughly 89 percent, keeping in mind that this number is an estimate of an unusual sampling situation.

The average age of the prisoners surveyed was 29.6. Of the 51 prisoners in the sample, seven indicated they only had "some schooling" but had not graduated from high school or had no GED; 20 had graduated from high school, and 18 indicated they had some college or technical training. Four indicated they had obtained an associate's or two-year technical degree, and two reported having graduated from college. Thirty-two were of Anglo-American, Caucasian heritage, and 10 reported they were of African-American descent. The others were either Mexican American or Native American. All of our sample claimed to be fathers: 28 indicated they had one child with the target mother, 11 said they were the father of two children with the target mother, and 11 indicated they were the father of more than two children.

One father reported having six children. Thirty-seven of the 51 (72%) indicated that this was at least their second prison sentence. These 51 men had been in prison for an average of 25 months with the minimum number of months at 11 and the maximum at 75. Sixteen of the 51 reported they were in prison for some offense directly related to drug charges (selling, manufacturing, and/or possession). Eleven were convicted of burglary, robbery, and/or assault. The rest had been convicted of a variety of other felonies.

Incentives. All participants were given 20-dollar stipends for participating. The stipends were deposited directly into their prison commissary accounts. Incentives are a difficult IRB issue with protected groups. Twenty dollars seems like a nominal amount for completing an extended interview outside of a prison. However, in many cases this small sum represents more than one month's pay for prison work. Therefore, the question arises whether such an amount constitutes coercion. That is, is $20 so much money, relatively speaking, that it may eliminate the prisoner's ability to refuse our recruiting efforts? The two IRBs (Oregon State University and Brigham Young University) did agree that this level of incentive was acceptable. However, in the similar study reported in this issue conducted in the UK, we were unable to offer any incentive. One key question of our project was answered. Inmates about to be released and who met our criteria were willing to respond to our questionnaire and were appreciative of the incentive. The Oregon prison officials suggested alternative incentives might be more appropriate. For example, given the prison diet, having soft drinks and pastries is a substantial incentive without being too expensive.

Interviewer's gender as a bias. We wondered whether inmate men would respond differently to men or women interviewers. We instructed interviewers to take careful notes about this issue. As far as could be discerned, there was no appreciable difference that could be directly attributed to the gender of the interviewer. We attempted to do an informal check of the transcripts of the interviews (looking at length of responses and depth of disclosure) and found no appreciable differences. In only one case did an inmate use inappropriate language and gestures with our female interviewer. At that point the interview was terminated and that participant was eliminated from the study (but he was still given the incentive as per IRB instructions).

It is important to note that it was possible to successfully interview men in prison and they were willing to speak openly and candidly about family issues. Even though they were candid, we could not triangulate on the accuracy of their comments without the reports of other family members. For example, without a crosscheck from the partners/spouses, one is never sure about the veracity of the data collected in these settings. It was our overall assessment that these men were able to respond to their feelings about their families, and on several occasions these reflections were filled with emotions and regret. Team members, however, on several occasions wondered in debriefing sessions about a "halo" effect in these interviews. Since these men felt some of the euphoria and high anxiety with release, we wondered whether their appraisals of the upcoming events were not distorted by the years of absence and the possibility of created fantasies about how release would play out. In the few interviews we did have with partners, it seemed that their stories were quite divergent from the men's. That is, while men believed they were going to have access to their children, women were not so sure about their willingness to have them return to family life and have contact with the children.

We were unprepared for the multiple difficulties of interviewing men following release. We had anticipated that the promise of a follow-up incentive ($25) would be sufficient. We also underestimated the value of having multiple contact avenues. Many of the telephone numbers we were given were not in service; the parties had moved, or the numbers were bogus. We had also anticipated that parole officers would be able to help us with follow-up contacts. This proved to be a difficult issue: while the officers wanted to be helpful, they were constrained by their own heavy case loads and inability to release information without prior consent from agencies and participants. Future studies need to consider these prior consents. This can be somewhat difficult because the connections between the prisons and the parole officers are minimal or nonexistent. If there were a way to involve the parole officer with the inmate during the last month of incarceration, this would be helpful. In both Oregon and Utah, inmates have their first contact with their parole officer about a week after release. On the other hand, our experience in the UK revealed that parole officers work directly within the prison and then transfer their caseload (once a prisoner is released) to another officer working in the town or region where the prisoner lives. With this coordinated effort we had more success in locating and interviewing these men and their spouses/partners following release.

MEASURES

Contact. Couple contact is a key constraint when studying incarcerated fathers (Lynch & Sabol, 2001; Nurse, 2002). We do not usually think about couple interaction in such basic terms, but in this case it becomes essential to do so. Prison researchers often access prison records to ascertain how many times spouses/partners (and other family members) visit. These measures are not sophisticated, nor are they even assessed in multiple-item scale form. Instead, they are measured simply as times visited per month or in some cases over a one-year span. Even though a simple marker, this is a key variable of couple well-being. Continued contact may indicate a higher level of commitment, loyalty, and relationship stability.

In addition, several questions were asked about contact in prison with spouses/partners and children. "How often has your child's mother contacted you by phone (or visits or mail, each asked separately) while you have been in prison" (none at all, 1 or 2 times, less than monthly, about once a month, about once a week)? "How often do you send cards or letters to your spouse/partner" (or child-asked separately, same responses as above)? The quality of the visits was assessed by asking, "Overall, how well would you say the visits go with target mother (or child) when she (or child) comes?" The responses for these questions were scored on a five-point scale ranging from very badly to very well. A chart was then filled out that listed 10 potential problems (transportation, short visits, children don't want to come, waiting-room problems, childcare problems, misbehaving children, scheduling conflicts, and disrespectful prison staff). Participants rated each of these 10 potential problems as either frequently a problem, sometimes a problem, rarely a problem, or never a problem. We also asked whether the prison tried to avoid visiting family members, especially the child's mother and/or the child.

Relationship quality. Of course, simple contact is an incomplete measure of relationship quality, although it is frequently used as a rough proxy for it. Of course, any measure of one's relationship (especially from the view of only one partner) is bound to be subjective in nature. High quality, in our context here, refers to the enduring good traits that form a more positive relationship. That is, we assume that one can subjectively rate a relationship by reflecting upon activities they frequently do together and how they share their life together. In this way researchers have been asking couples for several decades to subjectively rate the quality of their dyadic relationships (see Bowen & Orthner, 1983; Spanier 1976). Unfortunately, these dyadic scales used to assess quality of relationships require couples to respond to common daily activities (such as balancing the checkbook, division of role responsibilities). Obviously, calculating quality of a relationship in which the two partners see each other very sporadically and never in a home context is a unique task. The only approach available for this project was to simply ask these men directly to put a score on how they rated the relationship quality.

The relationship quality questions were: "Before your incarceration, how close to the target child do you think you were?" and "Before your incarceration, how close to the target mother do you think you were?" These questions included a brief description of what was meant by close ("Close can mean that they feel love and affection and that they are interested in what happens to their spouse/partner or child"). They were also asked about the amount of free time they spent with their child (1 = none, 2 = less than 2 hours per week, 3 = 3-10 hours per week, 4 = 11-20 hours per week, 5 = 21-30 hours per week, 6 = 31 hours or more per week). They were also asked whether they thought they were a good father (1 = not very good, 2 = some trouble being a father, 3 = an average father, 4 = a better-than-average father, 5 = a very good father). The same question was then asked to assess how they would rate themselves currently as a father.

Five questions measured the inmates' feelings about their children during incarceration (assessed on a five-point scale ranging from strongly agree [5] to strongly disagree [1]). These included responses to "I feel like my child likes me and wants to be close to me"; "My child rarely does things that make me feel good"; "Sometimes I feel like my child doesn't like me"; "When I do things for my child, I get the feeling that my efforts are not appreciated"; and "My child seems to be harder to care for than most."

Also, several demographic constructs were measured to provide a context for understanding the inmates that were interviewed. These measures included age, race, level of education, number of prior felony convictions, and how long they have been in prison for the current conviction.

RESULTS

Prison men's view of fathering. The 51 men in our sample had some surprising things to say about their identity as a father and how that identity had changed while in prison. Arditti et al. (this issue) provide a qualitative exploration of these issues in depth.

Thirty of these men claimed that they were an average father (or better) before they went to prison. Fifteen indicated that they were better than average or "a very good father." Their response to their evaluation of themselves as a father since being in prison collectively dropped: only 18 said they were an average (or better) father. Thirteen indicated they were a better than average or a very good father. About 26 of these men claimed that their ideas about parenting, disciplining, and their overall approach to interacting with their child had positively changed during their prison experience. Specifically, most of those (20) said that they should have been more connected to their children, given more direction, and been less punitive. Many of these fathers lamented the loss of relationships, changes in how their children perceived them, and how devastating it was to lose the role of being an active parent. Twenty-nine said that even while incarcerated they strongly agreed that their child liked them and wanted to be close to them. Only eight out of the 51 said they disagreed or strongly disagreed with the statement that their child liked them and wanted to be close.

The perceptions of these men's connections to their children were fairly clear for most of the men we interviewed. They saw themselves as pretty good fathers. Prison experience had dampened their ability to be a good father, but overall their children still loved them and wanted them in their lives.

Having said that, other findings seem to contradict that rosy picture. When asked how often they discuss daily issues of the child's life during a visit with the spouse/partner, 41 percent indicated never or rarely did this happen. About 30 percent of the men stated that during visits (with and without children present) there was often or always "an underlying atmosphere of hostility and anger."

Measuring couple quality and interaction. As part of our questionnaire, we set out to find ways of assessing the unique dyadic relationships between men in prison and their partners. The research about these types of relationships is sparse, and our first attempts here should be seen as a preliminary and developmental. First, we found that men were willing and often eager to talk to us about their relationships. However, repeated attempts to speak with the women partners turned out to be less fruitful. The women were often working, deep in childcare issues, uninterested in talking about the absent prisoner, or simply not interested in being interviewed.

We began by examining basic relationship markers prior to prison service. Forty of the men said they were living with the partner before prison with 15 reporting living with the partner for five years or more prior to entering prison. Only six of the prisoners said they had lived for six months or less with the partner prior to prison. Twenty-eight of the men (56%) said they had only one child with that identified partner, and almost half of them said they had another child (at least one child) with another partner that lived in another household. This finding speaks to two notions. Most of these men claimed to have been embedded in serious relationships prior to prison. Additionally, many of them reported multiple partners and multiple children from different partners. This speaks to the complexity found in their lives. They seem to have disparate loyalties and responsibilities.

Surprisingly (compare the following finding with the UK sample reported in the Clarke et al. paper in this collection), 65 percent (33) of our men reported that they had received no visits from the target spouse or partner during their stay in prison. Twenty-three percent indicated they received fairly frequent visits (i.e., once a month or week). These numbers show a kind of bifurcation: either the spouse or partner was supportive and visited regularly, or she was not a part of his life in prison. Prisoners noted that the primary reasons for (in their opinions) visitation problems were the mother's lack of interest, problems with children's care, and claims of harsh and/or disrespectful prison staff.

These men all had access to phone privileges. While the cost for telephone calls is high for both facilities (more than $1 per minute--reverse charges), most said they had contacted family members while in prison. However, 41 percent of these men said they had no or very infrequent (once or twice a year) phone contact with their spouses/partners. In like manner, about half said they never or very infrequently received mail from partners.

We also asked these men about the quality of the relationship they experienced with their partners/spouses before and during incarceration. First it is important to note that prior to prison only five men in our sample claimed they were married. Surprisingly, only about 10 indicated they were cohabiting or dating. Another 20 claimed to be divorced or legally separated upon entering prison. The rest marked that they were friends with the mother of the child, and a few marked "other" and refused to comment. When asked what their relationship with the child's mother was currently, only 21 percent said they were divorced or legally separated. These findings are counterintuitive given that divorce following incarceration usually increases. These men indicated that now the child's mother was more likely to be in the "friend" category (40%) than in the divorced or separated.

Interviewers inquired about couple quality during visits. More than half of our participants indicated there was never or rarely an underlying atmosphere of hostility or anger toward their partner during visits. Keep in mind, however, that few of these women are actually visiting on a regular basis. Probably only partners who care deeply about the prisoner are coming for visits. Prisoners also indicated that they talk regularly about child discipline issues with the child's mother, but only 20 percent indicated that those discussions often or always centered on differences in opinion. Prior to prison, nearly half of these men claimed that their relationship with the partner was either very good or excellent. Unfortunately, we have no reliable way to assess the mother's view of that claim. In like manner, the same number of men (just under half) said their current relationship with the partner was still very good or excellent. Finally, 42 of the 51 men (82%) indicated that their imprisonment had resulted in the creation of serious problems for their family or family members.

It is also apparent that these men have had very few contacts with their children. When asked about visitation and phone or mail contact, 33 reported that they had never received a visit from the child of interest. However, most believed they had a close relationship with the child: 40 reported that before incarceration their relationship with the child was close or very close. Even after years of incarceration and little if any contact, 32 men claimed they were still close or very close to the child. Interestingly, when commenting about their activity level with the child of interest prior to incarceration, about 50 percent of the men said they spent more than 20 hours a week in direct contact with the child in activities like reading, doing homework, or play.

Finally, 18 months after the interview period, a follow-up inquiry was conducted with the Oregon and Utah departments of correction. Fully 40 percent of the men (20 out of 51) interviewed were back in the prison system. The estimated average sentence for our sample for their current conviction at the time of the interview was nearly three years. At an estimated cost of $23,000 per year in prison for each prisoner, it cost the states of Oregon and Utah about $1.4 million to keep those men with the disheartening result of their being rearrested within a few months. If one considers the 1,700 men per day released back into society across the U.S. and realizes that many (perhaps 70%) will be rearrested within a short time, one can readily calculate the enormous financial and general societal cost associated with rearrest and incarceration.

FOLLOW-UP INTERVIEWS

For several months our research team attempted to track and reconnect with the men originally contacted and to contact the identified partners/wives. We were able to connect and/or interview only seven of the women identified by men as a partner or spouse. We were able to connect with so few that any statistical comparisons were not possible. Also, as mentioned, the Clarke study in this collection reports on efforts to follow the women in the UK sample. That research team was much more successful. It should be noted that the men in our sample were aware that we were attempting to contact their spouses (they provided the contact information). We thought that perhaps even their knowing that we could talk to their partners might alter their responses to the questions asked. For now, that idea remains an untested methodological problem.

DISCUSSION

There are several important lessons to be learned from actually conducting research with men in this situation. First, access is possible albeit time consuming and difficult at times. While there were several methodological barriers encountered during this pilot project, there were surprising levels of cooperation with the prison staffs in all of the institutions. Our experience was that the prison staff were eager to have social scientists conduct meaningful and useful research. Given their limited research budgets, any assistance is welcomed. Second, we discovered that these incarcerated fathers would speak with us about delicate and personal topics. For the most part, they were courteous and respectful of our interviewers. We do not know how accurate the information they provided was, but the project demonstrated that information could be collected.

Third, we found that issues of literacy and comprehension ability were not a problem. We constructed our questionnaires so that they reflected about an eighth-grade reading level. We had no recorded incident of interviewers finding that the questions were too difficult or misunderstood. We also found that a key element of this process was the stories they were encouraged to tell. The ethnographic and con textual information we collected were essential to understanding the situation of these men.

We discovered that undergraduate and beginning-level graduate students could administer these interviews with success. With training and coaching they were able to conduct themselves with a high level of professionalism and efficacy. The mentoring and preparation aspect of this project was nearly as important as the data collection aspect. Early on, we had reservations about safety, reliability, and ability to produce consistent results in this unique environment. Those fears were allayed, and future research efforts in the prison system will be conducted using selected students.

We speculate that the physical arrangement during the interviews and who is present during the interviews may influence data collection and results. In future research, interviewers should attempt to conduct data-gathering efforts when there are no guards or other prisoners in the same room. This, of course, becomes a safety issue tradeoff. However, our guess is that the data become very suspect when a guard is even in the same room.

With regard to our findings, we noted that ambivalence is a key concept that repeatedly surfaced from both the quantitative and qualitative interviews. It became apparent that prison relationships are complicated and do not fit neatly on the simple continua or scales we attempted to measure. In one case, a prisoner spent several minutes explaining that he both loved and hated his partners. We frequently noted such ideas of ambivalence and contradictory feelings.

Additionally, we frequently heard expressions of what could be labeled "idealized expectations." This is a key concept that should be pursued in future research with these families. Their self-ratings about their fathering expertise were probably, to some degree, socially acceptable expressions. However, there was a consistent theme expressed by these men that they saw themselves as good fathers and as fairly involved with their children even though they were in prison and they had not seen their child for months. This disconnect needs to be explored in future research.

There were some unexpected responses we heard from these fathers about their plans for future involvement. For example, one prisoner to be released commented that he was a good father and really looked forward to reentering family life. He said, "I am so excited about seeing my kid that I will visit him as soon as I get out, after I get a fix." The point of this example is that when we entered this topical and sample domain, we were quite naive about what the range of responses might be.

Another key issue that became a hallmark of our experience was that we learned firsthand the importance of gaining trust. We found that, ideally, placing researchers in the prison for several weeks prior to the interview period would reassure the men about the credibility of the research team. As one can imagine, there is a fair amount of mistrust, suspicion, and anger that must be dealt with. We concluded that the research project would be easier to complete and results more reliable if we were to embed a researcher in the prison. That is, a quasi-staff person could be on-site for several weeks conducting the interviews and building prisoner and staff trust. This, we feel, would do much to facilitate meeting and recruiting participants. Additionally, this approach would improve sampling confidence as we gain the trust of prison staff members.

We also discovered that follow-up is one of the most difficult aspects of this process. We were unprepared for the incorrect or disconnected phone numbers, lack of permanence, and inability of even the parole officers to assist with finding men once out of prison. We had anticipated that this element of the research process would be difficult, but we were unprepared for the immense hurdles in finding and talking to these men once they were released.

We also discovered that "cold" contact with the mother is very difficult. Sending letters and then calling a mother whose child has an incarcerated father is problematic. She has little reason to trust anyone who may be associated with the prison system and might appear to be using some form of intrusive deception.

In sum, this methodological experiment found some success and many obstacles. In the end, however, it has become clear to our team that this line of research is important, even essential, and has high potential for policy and program application. Clearly, much of the scant prison programming that is being accomplished is done without much prior or follow-up foundational research. This research effort is necessary but difficult.

REFERENCES

Arditti, J., & McClintock C. (2001). Casualties of the war: Drug policy and families. Marriage and Family Review, 32, 11-32.

Beck, A.J., Karberg, J., & Harrison, P.M. (2002). Prison and jail inmates at midyear 2001. Washington, DC: Bureau of Justice Statistics.

Bowen, G., & Orthner, D. (1983). Sex-role congruency and marital quality. Journal of Marriage and the Family, 45,223-230.

Bureau of Justice Statistics (2002). Annual Parole Survey, 2000. Washington, DC: Bureau of Justice Statistics.

Fritsch, T.A., & Burkhead, J.D. (1981). Behavioral reactions to children to parental absence due to imprisonment. Family Relations, 30, 83-88.

Garland, D. (2001). The culture of control: Crime and social order in contemporary society. Chicago, IL: University of Chicago Press.

Hagan, J., & Dinovitzer, R. (1999). Collateral consequences of imprisonment for children, communities, and prisoners. In M. Tonry & J. Petersilia (Eds.), Prisons: Vol. 26. Chicago, IL: University of Chicago Press.

Hammett, T., Roberts, P., & Kennedy, L. (2001). The burden of infectious disease among inmates and releasees from correctional facilities. In The health status of soon-to-be released inmates: Vol. 2 (pp. 15-28). Washington, DC: National Institute of Justice.

Hairston, C. (2001). Fathers in prison: Responsible fatherhood and responsible public policies. Marriage and Family Review, 32, 111-135.

Kampfer, C. J. (1995). Post-traumatic stress reactions to children of imprisoned mothers. In K. Gabel & D. Johnson (Eds.), Children of incarcerated parents. New York: Lexington Books.

Lynch, J., & Sabol, W. (2004). Assessing the effects of mass incarceration on informal social control in communities. Criminology and Public Policy, 3, 267-294.

National Institute of Corrections (2002). Retrieved May 23, 2005 from http://aspe.hhs.gov/hsp/prison2home02.

Nelson, M., Deess, P., & Allen, C. (1999). The first month out: Post-incarceration experiences in New York City. New York: Vera Institute of Justice.

Nurse, A. (2002). Fatherhood arrested: Parenting from within the juvenile justice system. Nashville, TN: Vanderbilt Press.

Office of National Statistics (2005) Web tables on marital status, http://www.statistics.gov.uk/StatBase/Expodata/Spreadsheets/D8751 .xls.

Petersilia, J. (2003). When prisoners come home: Parole and prisoner reentry. New York: Oxford University Press.

Pattillo, M., Weiman, D., & Western, B. (2004). Imprisoning America: The social effects of mass incarceration. New York: Russell Sage Foundation.

Pope, C., Lovell, R., & Brandi, S. (2001). Introduction: Foundations of criminal justice research. In C. Pope, R. Lovell, & S. Brandl (Eds.), Voices from the field: Readings in criminal justice research (pp. 1-16). Belmont, CA: Wadsworth.

Sack, W.H. (1977). Children of imprisoned fathers. Psychiatry, 40, 163-174.

Spanier, G. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 15-28.

Tonry, M., & Petersilia, J. (Eds.) (1999). Prisons: Vol. 26. Chicago, IL: University of Chicago Press.

Travis, J., & Wahl, M. (2005). Prisoners once removed: The Impact of incarceration and reentry on children, families, and communities. Washington, DC: Urban Institute Press.

Western, B., & Pettit, B. (2002). Beyond crime and punishment: Prisons and inequality. Contexts, 1, 37-43.

The interviewing team included the following: Ben Carter, Ben Gibbs, Anita Armstrong, Kristen Sturgill, Christy Smith, Cassie Hanks, Jessica Smith, Robb Clawson, Matt Thorpe, Lahela Lindsay, Amy Tuttle, Ashley Kitchen, and Justin Dyer. Generous support for this project came from Brigham Young University: School of Family Life and Family Studies Center, the BYU Mentored Learning Grant Program, and the College of Family Home and Social Sciences.

Correspondence concerning this article should be addressed to Randal D. Day, Brigham Young University, School of Family Life, 2092B JFSB, Provo, UT 84602. Electronic mail: [e-mail].

Randal D. Day

Brigham Young University

Stephen J. Bahr

Brigham Young University

Allan C. Acock 

Oregon State University, Corvallis

Joyce A. Arditti

Virginia Polytechnic Institute

and State University