Abstract
This article extends research on the association between paternal incarceration and family functioning by differentiating between families with fathers who have been incarcerated in local jails, state prisons, federal prisons, and unknown types of facilities. Data from the Fragile Families and Child Wellbeing Study (FFCW) enable this finer grained analysis. We show that there are few observable differences between families with fathers incarcerated in a local, state, or federal facility and the 53 percent of families with fathers incarcerated in an unknown facility type. We test the association between facility type and family functioning using a series of fixed effects models, showing strong associations between facility type and only two of seven family outcomes. The evidence presented here suggests that family functioning does not markedly vary by type of facility, but this finding needs to be substantiated in future research.
By the close of the twentieth century, incarceration had become a common experience not only for poor and working-class African American men (e.g., Bonczar 2003; Pettit and Western 2004) but also for their families (e.g., Lee et al. 2015; Wildeman 2009). Because incarceration is so common and heavily unequally distributed in the United States, there has been great interest in how incarceration affects the subsequent life chances of both men (Kling 2006; Loeffler 2013; Massoglia 2008; Pager 2003; Schnittker and John 2007; Western 2002) and the parents, partners, and progeny that they leave behind (Arditti 2012; Braman 2004; Comfort 2007, 2008; Foster and Hagan 2007; Geller et al. 2012; Lee, Porter, and Comfort 2014; Nurse 2002; Turney 2014b; Turney and Wildeman 2013; Wakefield and Wildeman 2013; Wildeman, Schnittker, and Turney 2012; Wildeman 2010). With a few noteworthy exceptions (e.g., Kling 2006; Loeffler 2013; Comfort 2007, 2008; Turney and Wildeman 2013; Wildeman 2010), research in this area suggests that incarceration not only harms the life chances of the adult men who cycle through the system but also those tied to them (for reviews, see especially Foster and Hagan 2015; Wakefield and Uggen 2010; Wildeman and Muller 2012).
Although research to date makes important contributions to our understanding of family life and social stratification in the contemporary United States, it still suffers from a number of limitations. Among the most important of these limitations is that virtually none of this research successfully differentiates between—or attempts to differentiate between—the consequences of jail incarceration, state prison incarceration, and federal prison incarceration for family outcomes (see especially Massoglia and Warner 2011, 852–54), making it impossible to know whether the effect of incarceration is relatively uniform, even if these effects likely work through different mechanisms across different facility types, or if it varies across different facility types. This limitation applies to relevant quantitative and qualitative research alike. Qualitative work in this area tends to focus either on a small number of specific prisons (e.g., Arditti 2012; Comfort 2008; Turanovic, Rodriguez, and Pratt 2012) or lump incarceration in a host of different types of facilities into one category (e.g., Braman 2004). Quantitative research has similar limitations, as the most oft-used datasets for considering the effects of incarceration—the National Longitudinal Survey of Youth 1979, the Project on Human Development in Chicago Neighborhoods, and the Fragile Families and Child Wellbeing Study (FFCW)—either include measures of incarceration that provide weak proxies for local jail incarceration (although strong proxies for prison incarceration; Western 2002, 530) or have never before released data on correctional facility types (Wakefield and Wildeman 2011, 797). 1
This inability to differentiate between local jail incarceration, state prison incarceration, and federal prison incarceration limits our understanding of how criminal justice contact shapes family life in three ways. First, distance to family and visitation procedures differ based on facility type and also shape the frequency of contact between inmates and their family members. For instance, although jail inmates are often incarcerated very close to their homes, state and federal prisoners are often imprisoned quite far from their families, potentially making visitation far more difficult for prison inmates than for jail inmates (e.g., Christian 2005; Comfort 2008). But state and federal prisons also generally have more transparent visiting guidelines that are applied with greater consistency than those of local jails (but see Comfort 2008), meaning that jail inmates may not have as strong an advantage in terms of the number of visits they experience relative to prison inmates as one might expect. Inmate visitation and the maintenance and nurturing of preincarceration social ties are associated with smoother adjustments to incarceration (Clark 2001; Wooldredge 1999) and better postrelease outcomes (Bales and Mears 2008; Ryan and Yang 2005; Visher and Travis 2003; Wolff and Draine 2004). Because most research on distance and visitation has focused on how contact with family and friends while incarcerated is related to in-facility behavior and recidivism, little is known about family functioning outcomes (Visher and Travis 2003; but see Nurse 2002).
Second, conditions of confinement differ across institution types. For instance, jails tend to offer fewer physical and mental health services than do prisons, which may shape the effects of incarceration on family functioning. Different types of correctional facilities, moreover, house different types of inmates, and the distinctive mix of inmates in local jails, state prisons, and federal prisons could also shape how incarceration affects family functioning through a variety of channels. Prior research on confinement indicates that an individual’s own physical and mental health can have broader social effects, influencing or covarying with the health of other family or household members (e.g., Gallagher and Mechanic 1996; Tammentie et al. 2004; Townsend, Miller, and Guo 2001). These findings, combined with the knowledge that the link between incarceration and mental health is explained in no small part by the link between incarceration and family processes (Turney, Wildeman, and Schnittker 2012), suggest that heterogeneity by facility type may be associated with varying levels of postincarceration well-being and relationships between incarcerated fathers and their current or former partners.
Third, because many jail inmates are awaiting trial, there could be greater uncertainty surrounding the time of release for jail inmates relative to inmates of state or federal correctional facilities. As criminal justice practices and policies across the country shifted away from a rehabilitative model to one with more determinate sentencing practices, a wealth of literature has emerged supporting the conclusion that increased certainty around term length is associated with less psychological distress among inmates (e.g., Goodstein 1984; Goodstein and Hudack 1982; Parisi 1982). Differing levels of ambiguity around term length by facility type may therefore be associated with varying experiences of incarceration for inmates, which may indirectly influence family processes or inmates’ families. Despite the various ways in which incarceration in a local jail, a state prison, and a federal prison differ, no existing research has simultaneously considered how these types of institutional contacts differentially affect family functioning.
In this study, we use never-before-available data from the FFCW on the type of facility in which fathers are incarcerated to show how the association between current paternal incarceration and family functioning varies by whether the father is incarcerated in a local jail, a state prison, or a federal prison. These family functioning indicators include whether (1) the mother and father’s relationship has dissolved and (2) the mother has a new partner, as well as maternal reports of (3) relationship quality, (4) material hardship, (5) engagement with the focal child, (6) depression, and (7) parenting stress.
The analysis proceeds in three stages. In the first, we show that there are few observable differences between fathers who are currently incarcerated in a local jail, state prison, or federal prison and fathers with an unknown facility type, as we do not have information on facility type for about 53 percent of currently incarcerated fathers. In the second stage, we show descriptive differences in family functioning between families of currently incarcerated fathers and other FFCW families by facility type. In the third stage, we use fixed effects models to consider how the association between paternal incarceration and family functioning varies across facility types. Before doing so, however, we include two between-wave transition matrices to demonstrate that there is enough between-wave change in incarceration status for fixed effects models to be appropriate.
Although the results are somewhat underpowered—because only 47 percent of mothers reported on the type of correctional facility in which the father was currently incarcerated 2 and because a small share of fathers are currently incarcerated at any given wave in the FFCW data—the results nonetheless provide support for a number of important conclusions. First, descriptive differences in family functioning between families by current paternal incarceration are substantial, but differences in family functioning by facility type are muted, as there are few statistically significant or substantively large associations. This indicates that either the paternal incarceration–family functioning association is insensitive to correctional facility type or that between-facility type differences in the characteristics of the incarcerated fathers and the families in which they are embedded are suppressing differential associations across facility types.
Second, families experiencing paternal incarceration were more likely to have had the parental union dissolve and to have the mother move on to a new partner, even after adjusting for fixed but unobserved traits of families. This is unsurprising given the well-documented association between incarceration and union dissolution (e.g., Lopoo and Western 2005; Turney 2015b). But with these two exceptions, the fixed effects models provide little evidence that the descriptive differences in family functioning are driven by incarceration or that there are any large or statistically significant differences across facility types in effects on family life.
Despite data limitations, these analyses lend important insights into the field and suggest directions for future research. First, although the large amount of missing data might be seen as suggesting otherwise, mothers who know that their child’s father is incarcerated are willing and able to comment on facility type, meaning that subsequent research on the differential consequences of incarceration across facility types can expect a similar battery of questions to accurately indicate when and where a loved one is incarcerated. Second, because the results provide little support for the hypothesis that paternal incarceration in any one type of correctional facility is any more detrimental—or beneficial—for family functioning than another, researchers interested in the connections between incarceration and family life should be attentive to a range of criminal justice contacts rather than privileging one or another. Finally, these analyses suggest the need to think about how facility type might moderate other well-known associations such as the association between paternal incarceration and child development and well-being (e.g., Geller et al. 2012; Wildeman 2010) and incarceration and adult mental health outcomes (e.g., Turney Wildeman, and Schnittker 2012; Wildeman, Turney, and Schnittker 2014).
Data and Methods
Data
Following much research on the consequences of paternal 3 incarceration for family life (e.g., Turney 2014b, 2015b), family functioning (e.g., Turney 2014a, 2015a; Turney and Wildeman 2013), and child well-being (e.g., Geller et al. 2012; Haskins 2014; Turney and Haskins 2014; Wildeman 2010, 2014), we use data from the FFCW study. Mothers and fathers were interviewed in person at baseline, shortly after their children were born, and again when their children were approximately one, three, five, and nine years old. More information on the data can be found elsewhere (Reichman et al. 2001), but the primary benefits of the data for studying the consequences of incarceration for family life are that they are longitudinal, include a large sample of disadvantaged children who have experienced paternal incarceration, and include a range of outcomes related to the well-being of fathers, mothers, children, and families. Therefore, they facilitate the use of a strong research design to consider the consequences of incarceration for many indicators of family functioning.
The FFCW data have been used extensively to consider the consequences of incarceration for family life, but they nonetheless have a number of noteworthy limitations. First, attrition is high in the FFCW data, as is generally the case in longitudinal studies including marginalized populations. Second, information on offense type is only available from mothers at two waves and has a great deal of missing data, making it difficult to consider variations in the consequences of incarceration by offense type (but see Turney 2015a; Wildeman 2010). Third, the data include little information on the duration and frequency of incarceration (but see Turney 2015a). Fourth, the public release data include no information on the type of correctional facility in which the individual is currently incarcerated.
Our analytic sample includes mothers with nonmissing data on all outcome variables at the five-year survey (N = 4,069), each of which we describe below in detail. Since we use the same outcome measures at the one-year interview and the three-year interview, we impute observations missing this information at those earlier surveys to preserve observations.
Explanatory variables
Our explanatory variables, which form the core of our theoretical and empirical contributions, are drawn from unique never-before-released data on the type of correctional facility in which the father is incarcerated at the one-, three-, and five-year interviews. At each of these waves, mothers who reported that the father was incarcerated were asked the following: “Is this a local, state, or federal facility?” 4 Incarcerated fathers were not asked what type of correctional facility they were incarcerated in, and, hence we rely on maternal reports of correctional facility type.
As Table 1 shows, over the course of the three surveys, many mothers reported that the father was currently incarcerated in a local jail, a state prison, or a federal prison. Over the three waves of data, there were 139 periods of current paternal jail incarceration reported, 173 periods of current paternal state prison incarceration reported, and 67 periods of current paternal federal prison incarceration reported. Thus, there were 379 total instances when the mother reported the father to be currently incarcerated in a specific correctional facility type. There were also 432 periods of current paternal incarceration when the father was incarcerated in an unknown correctional facility type, meaning that facility type is missing for 53 percent of observations.
Mothers’ Responses to Facility Type Questions, by Year
NOTE: Tabulations are unweighted. Facility type information is taken from the one-, three-, and five-year surveys of mothers. The “not asked” category includes those who were not asked the question as part of the survey protocol or were not part of the sample in that survey year.
As Table 1 indicates, much of this missing information is not due to mothers’ lack of knowledge about the type of facility the father is in but rather is due to not being asked about the facility type. There are three reasons mothers were not asked about incarcerated fathers’ type of correctional facility. Mothers were not asked if they did not participate in the survey that year, if they lived in one of the two cities in which the question was not asked at the one-year survey, or if they reported the father was not incarcerated (but fathers are considered currently incarcerated if either the mother or father reports incarceration).5,6
Table 1 shows that, in the vast majority of cases in which the mother reported the father was currently incarcerated and was asked about his facility type, mothers know the facility type. Of the 65.7 percent of mothers who were asked about facility type at the one-year interview, 83.1 percent (54.6 percent/65.7 percent) reported that he was currently incarcerated in a local jail, a state prison, or a federal prison, and only 16.9 percent (11.1 percent/65.7 percent) reported that they did not know the current facility type. Thus, mothers who know that the father is currently incarcerated often also know the type of correctional facility.
Dependent variables
We included seven measures of family functioning as our dependent variables, all of which have been described elsewhere (Turney 2015a, 2015b; Turney and Wildeman 2013). We used measures of (1) whether the mother and father had separated, (2) whether the mother reported a new romantic partner, (3) mother-reported relationship quality with the father, (4) mother-reported material hardship, (5) mother-reported parental engagement, (6) mother-reported depression, and (7) mother-reported parenting stress. These measures provide insight into family structure, the quality of family life, and maternal material and psychological well-being.
Other variables
We provide descriptive information for the following measures, weighting this information so this analytic stage is representative of the full FFCW birth cohort sample: paternal race and ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, non-Hispanic other race), father’s foreign-born status, paternal age, paternal education (less than high school, high school diploma or GED, more than high school), father’s current marital status with the child’s mother, father’s employment status immediately prior to incarceration, and father’s previous incarceration. We provide descriptive information only for prior incarceration for the three- and five-year follow-up surveys because there is limited information on prior incarceration at the one-year survey.
Methods
The analysis proceeds in three stages. In the first stage, we considered how the demographic characteristics of currently incarcerated fathers for whom facility type is unknown (N = 432) compare to those for whom facility type is known (N = 379). We see this straightforward stage as important because it provides insight into whether there is anything obviously unusual, at least in terms of key demographic characteristics, about the unknown facility type group. In the second stage, we presented descriptive differences between families with and without currently incarcerated fathers by facility type for all seven of our family functioning measures. In the third stage, we used fixed effects models to show how movements in and out of various types of correctional facilities are associated with changes in family functioning. For these analyses, we have standardized all coefficients (with a mean of 0 and a standard deviation of 1) to compare the magnitude of associations across a range of different types of outcome measures.
Although this is an exceptionally rigorous research design, since it adjusts for all time-invariant differences between families and focuses only on changes in family functioning that are associated with changes in facility type, the method also comes with a couple of noteworthy limitations. Perhaps most importantly, because of the small number of fathers incarcerated in specific types of facilities, the analysis is somewhat underpowered, although not severely so, as Table 2 indicates. Second, these models key on both movements into a correctional facility and movements out of a correctional facility to generate estimates of paternal incarceration (and, specifically, facility type). This means that the estimates are partially a function of becoming incarcerated and partially a function of becoming an ex-inmate. Although keying on the shift from not incarcerated to currently incarcerated is reasonable and estimates a process of interest, the move from currently incarcerated to formerly incarcerated, while an important social process itself (Western et al. 2015), may bias our estimates toward zero, as research on the consequences of incarceration implies that current incarceration and a history of incarceration differentially affect family life (see Wildeman and Muller 2012).
Incarceration Status and Facility Type Transitions
NOTE: Tables report the distribution of end facility type state for fathers within each starting facility type state. Tabulations are unweighted.
Table 2, which shows the starting and ending state of fathers across each of the three survey waves, indicates that there is more movement both between facility types and in and out of any facility than might be expected. Of the fifty-three fathers incarcerated in a local jail at the three-year interview, 7 only four were incarcerated in a local jail at the five-year interview, with more not incarcerated (thirty-two), incarcerated in a state prison (six), or incarcerated in an unknown facility type (eleven). A fixed effects model, which adjusts for unobserved differences between families, appears to be an appropriate choice, even if the number of observations in each type of facility is somewhat smaller than would be ideal.
Results
How unusual are fathers currently incarcerated in an unknown facility type?
In the first analytic stage, shown in Table 3, we compared the background characteristics of fathers incarcerated in an unknown facility type to those of fathers incarcerated in a local, state, or federal facility. In terms of race, nativity status, age, marital status, employment status, and prior incarceration, the differences between fathers in an unknown facility type and other fathers are muted. There are some differences in terms of educational attainment, as fathers in an unknown type of facility are less likely than fathers in other groups to have dropped out of high school. But there are no other indications, at least on the basis of background characteristics, that fathers with unknown facility type are any different than fathers with known facility type.
Descriptives of Incarcerated Fathers, by Facility Type
NOTE: Estimates are weighted using the 1-year national weights. Descriptives come from the pooled one-, three-, and five-year surveys.
Descriptive differences in family functioning
In the second analytic stage, shown in Table 4, we descriptively considered how facility type may differentially affect family functioning. Consistent with research in this area (e.g., Turney 2014a, 2015b), the first two columns of Table 2 demonstrate that families in which the father is incarcerated have less favorable family functioning outcomes than do families in which the father is not incarcerated. Essentially, this shows that we can replicate prior research.
Descriptives of Family Functioning Measures, by Facility Type
NOTE: Estimates are unweighted. Facility type and outcome measures are taken from the pooled, one-, three-, and five-year surveys.
We next turn to differences in family functioning across facility types, our key theoretical and analytical contribution. No longer being involved in a romantic relationship with the focal child’s other parent was the norm for families in which the father was currently incarcerated in a local jail (72.7 percent), a state prison (76.9 percent), a federal prison (77.6 percent), and an unknown type of facility (76.2 percent). Additionally, many mothers reported relationships with a new romantic partner, and these descriptive statistics were relatively similar across mothers connected to fathers incarcerated in local jails (36.0 percent), state prisons (35.8 percent), federal prisons (43.3 percent), and unknown type of correctional facility (33.6 percent). Therefore, for both of these outcomes, differences between families by facility type were quite muted, although if the fathers were in federal prison, mothers were more likely to have had their unions dissolve and to have moved in with a new partner.
Mothers who share children with fathers incarcerated in federal prisons, compared with mothers who share children with fathers in other facility types, had other outcomes that are slightly more favorable. They reported less material hardship, less depression, and less parenting stress than other mothers, although they also reported the lowest levels of engagement with their children (although few of these differences are statistically significant given the small number of observations). Differences between mothers who share children with fathers incarcerated in local jails, state prisons, and some unknown type of correctional facility were more muted and had no clear pattern across these five outcomes. Thus, at least on the basis of descriptive differences in family functioning, there are few indications that differences across facility type are substantial.
Results from fixed effects models estimating family functioning
The results shown in Table 4 only provide descriptive information on family functioning across facility type, and a more rigorous test of these associations is needed. In the third and final analytic stage, which can be seen in Figure 1, 8 we considered the association between current paternal local jail, state prison, federal prison, and unknown facility type incarceration and seven indicators of family functioning, adjusting for all stable characteristics of families, both those that are observed and those that are not, to provide the strongest test possible with these data.
Current Paternal Incarceration and Family Outcomes by Facility Type
Consistent with the descriptive data, jail incarceration, state prison incarceration, federal prison incarceration, and unknown facility type incarceration are associated with a statistically significant increase in the probability of the mother and father separating and of the mother having a new romantic partner. These effects, moreover, are relatively small—between .02 and .04 standard deviations—but visual inspection suggests that they are quite uniform. 9
Beyond these two outcomes, however, there are few statistically significant associations for any of the facility types. Both current federal prison incarceration and unknown facility type incarceration are associated with statistically significant declines in mothers’ reports of relationship quality, and paternal incarceration in an unknown facility is also associated with a statistically significant decrease in mothers’ engagement. But the only other statistically significant association is between state prison incarceration and increases in maternal parenting stress. The coefficients throughout these analyses and the magnitude of the differences between the coefficients across various facility types, moreover, are quite small.
Discussion and Conclusions
We know far more about the consequences of having a family member—especially a father—incarcerated for family life now than we did even a few years ago, as a comparison of more and less recent reviews on incarceration and family life makes abundantly clear (e.g., Comfort 2007; Foster and Hagan 2015; Hagan and Dinovitzer 1999; Murray and Farrington 2008; Wildeman, Wakefield, and Turney 2013). We have stronger causal tests of the relationship between paternal incarceration and family well-being (e.g., Andersen and Wildeman 2014; Porter and King 2015), a more nuanced understanding of how incarceration could affect different types of families in different ways (e.g., Comfort 2008; Turney and Wildeman 2015), a clearer understanding of the mechanisms through which incarceration harms family members (e.g., Turney, Wildeman, and Schnittker 2012; Wildeman, Schnittker, and Turney 2012), and a better sense for how many families currently have a family member imprisoned (Lee et al. 2015) or have had one imprisoned (Wildeman 2009).
Despite great progress in understanding the spillover effects of incarceration on family life, significant gaps in research remain. In our opinion, one of the most important gaps has to do with our lack of knowledge regarding how the correctional facility type—local jail, state prison, or federal prison—might shape the consequences of incarceration for families. In this article, we used unique data from the FFCW—the dataset most often used to consider the familial consequences of incarceration, and a never-before-released indicator of the type of correctional facility in which a father is incarcerated—to address this pressing gap in the literature on the familial consequences of incarceration.
The results from our analyses provide support for a number of conclusions. First, even though we do not know the type of correctional facility in which 53 percent of currently incarcerated fathers reside, the differences—at least in terms of demographics—between these fathers and other currently incarcerated fathers are small, suggesting that there is nothing obviously unusual about fathers who are in an unknown facility type. Second, and most importantly, although the families of currently incarcerated fathers experienced disadvantages across all of the seven outcomes we considered, findings suggest such disadvantage results from father’s incarceration for only two of those outcomes. These findings could be driven by the small number of incarcerated fathers or because we consider the effects of current incarceration rather than any postrelease effects. Furthermore, even for the two outcomes where the association between paternal incarceration and family functioning was strong—whether the mother and father separated and whether the mother has a new romantic partner—there was little evidence of variation across facility types in the effects of paternal incarceration on family life, which is a key conclusion.
Although it is typical to immediately follow a results summary with some discussion of implications, we think doing so might be a bit bold because this is the first analysis of this type and the number of between-wave switches in and out of incarceration, while not miniscule, also leads to fairly large confidence intervals for our key outcomes. We, thus, instead discuss limitations and future research.
First, and most importantly, because of the large amount of missing data on the specific fathers’ type of correctional facility and the lack of information on the type of correctional facility of their prior incarceration, our analysis was somewhat underpowered, and delivered results that were underwhelming, provided the small differences across institution type in the paternal incarceration–family functioning association were driven not by truly small differences but instead by small sample sizes and large standard errors. This provides suggestions for future research: researchers working in this area should endeavor to gain better information on the types of correctional facilities in which both currently and formerly incarcerated men and women have been confined. This could be done by linking administrative data with survey data or by testing expanded versions of survey questions that provide more information on specific types of correctional facilities (including providing respondents with the names of correctional facilities to choose from).
Second, although the differences we found were not substantial, and, hence, there was no great need to discuss the differential effects of confinement in specific types of correctional facilities on family outcomes, the analysis was nonetheless limited by the fact that we had (somewhat) reliable information on correctional facility type, but even more limited on the duration or frequency of incarceration or the type of crime that the father had been convicted of across waves (or charged with, as would be relevant for fathers awaiting trial in local jails). The fact that we were able to include information on only one of these puzzle pieces in this article is a limitation because it makes it impossible for us to know whether the differential consequences of paternal incarceration for family outcomes by correctional facility type are driven by the duration of incarceration, frequency of incarceration, conditions of confinement, distance from family, or some combination thereof. Future research in this area should therefore endeavor to provide information on all of these pieces of information simultaneously. The National Longitudinal Survey of Adolescent Health data, which have also been broadly analyzed when considering the consequences of parental incarceration for children and families (e.g., Foster and Hagan 2007, 2009; Hagan and Foster 2012a, 2012b; Roettger and Boardman 2012; Roettger and Swisher 2011), do have more complete information on these pieces of information than do other datasets. However, since all reports in these data are based on child’s reports of parent’s incarceration, 10 it might be the case that, especially for incarcerations when the child was very young, the information gleaned from these data may be less reliable than data gleaned from parental reports of incarceration.
Third, the fixed effects modeling strategy is somewhat limited because estimates produced by these models are based on a combination of (1) the effects of fathers moving from not being incarcerated to being incarcerated and (2) the effects of fathers moving from being incarcerated to not being incarcerated, which might have very different effects. Future research on how family consequences of incarceration vary by the type of correctional facility might do well to differentiate the effects of entry and reentry, a task that was beyond the scope of this article because of the relatively limited amount of data and the models employed.
Although the limitations of this article are considerable and primarily a function of data limitations endemic to this type of research, the contributions of this article are also substantial. Perhaps most importantly, by showing no consistent differences across facility types in the consequences of current paternal incarceration, this article suggests that jail incarceration may be just as important for family functioning as the long stints upstate or out of state that garner so much more scholarly and public attention. As such, future research should not only try to better understand variation in the consequences of criminal justice for families across a host of nuanced dimensions but also focus far more on jails than research has historically.
Footnotes
NOTE:
We are especially grateful to Kate Jaeger for furnishing the data and enduring endless inquiries from the corresponding author about what other data on facility type might be available. We are also grateful to Hedy Lee, Sara Wakefield, Steve Raphael, the anonymous ANNALS reviewer, and the conference participants for comments. All remaining errors are ours.
Notes
Christopher Wildeman is an associate professor of policy analysis and management at Cornell University and a senior researcher at the Rockwool Foundation Research Unit. His work considers the prevalence, causes, and consequences of contact with the criminal justice system and the child welfare system for families. His first book, Children of the Prison Boom: Mass Incarceration and the Future of American Inequality (with Sara Wakefield) was published by Oxford University Press in 2013.
Kristin Turney is an associate professor of sociology at the University of California, Irvine. Her research investigates the complex, dynamic role of families in creating and exacerbating social inequalities. She is currently working on a qualitative interview project that involves interviewing jail inmates and their family members—including romantic partners, children, and mothers—both during their incarceration and after release.
Youngmin Yi is a PhD student in sociology and policy analysis and management and an affiliate of the Cornell Population Center and Center for the Study of Inequality at Cornell University. Her research interests include family demography; race, ethnicity, and immigration; and social networks and social stratification.