Abstract
Children experiencing parental incarceration face numerous additional disadvantages, but researchers have often relied on these other co-occurring factors primarily as controls. In this article, we focus on the intimate links between crime and incarceration, as well as on the broader family context within which parental incarceration often unfolds. Thus, parents’ drug use and criminal behavior that precedes and may follow incarceration periods may be ongoing stressors that directly affect child well-being. We also use our analyses to foreground mechanisms associated with social learning theories, including observations and communications that increase the child’s risk for criminal involvement and other problem outcomes. These related family experiences often channel the child’s own developing network ties (peers, romantic partners) that then serve as proximal influences. We explore these processes by drawing on qualitative and quantitative data from a study of the lives of a sample of respondents followed from adolescence to young adulthood, as well as on records searches of parents’ incarceration histories. Through our analyses, we find evidence that 1) some effects attributed to parental incarceration likely connect to unmeasured features of the broader family context, and b) together parental incarceration and the broader climate often constitute a tightly coupled package of family-related risks linked to intergenerational continuities in criminal behavior and other forms of social disadvantage.
Keywords: family dynamics, life course, parental incarceration, social learning, young adulthood
Research is accumulating rapidly on the negative consequences of recent incarceration trends for neighborhoods, families, and individuals (see Travis, Western, & Redburn, 2014, for a review). The findings from studies of effects on children who experience a parent’s incarceration (PI) have highlighted that the negative impact often reaches to the next generation. Researchers have documented effects across multiple domains ranging from delinquency to educational deficits and, eventually, to an intergenerational cycle of involvement with the criminal justice system (Foster & Hagan, 2015; Murray, Loeber, & Pardini, 2012). The results of studies with a focus on underlying mechanisms have highlighted the stressful, traumatic aspects of the experience (Arditti & Salva, 2015), negative effects on parent–child bonds, family stability, and parenting practices (Poehlmann-Tynan, 2015), as well as the deleterious role of secondary stigma/labeling (Comfort, 2007).
A potential limitation of this line of research is that, as Uggen (2013) has noted, the literature on consequences of incarceration has developed mainly along a separate track from the broader “causes of crime” and related criminological traditions, including the literature on intergenerational transmission. Most studies of PI effects have included controls for numerous sources of disadvantage, but a challenge is that data sets may lack detailed information about the parent’s criminal behavior (Kirk & Wakefield, 2018). In addition, as the objective of most studies has been to isolate the incarceration effect itself, this has required “zooming in” on the incarceration period rather than the kind of “zooming out” that would be more likely to direct attention to these other features of family climate. Wakefield and Apel (2018) recently offered an additional rationale for focusing attention primarily on the incarceration period. The authors suggested that in contrast to the sporadic and frequently less observable character of parents’ criminal involvement, parental absence resulting from incarceration is an experience that is almost by definition noticeable to the child. Thus, in focusing in on a discrete, timed event and its consequences, PI effects studies have fit well with the broader life-course tradition, which has also generally been centered on the impact of key events, transitions, and turning points (see, e.g., Elder, 1998).
Although researchers have documented considerable stresses and long-term corrosive effects of the experience of PI, a comprehensive treatment of mechanisms nevertheless requires additional scrutiny of the broader family context (BFC) within which incarceration often unfolds. It is important to focus theoretically and empirically on the nature of family life before and after the parent’s incarceration as, in most instances, these other periods last longer and may be more all encompassing relative to the incarceration period(s). Family systems theory and trauma-informed approaches are logical conceptual frameworks for locating PI within the context of a broader set of family experiences (Arditti, 2012). We wish, however, to focus attention as well on the family dynamics highlighted by social learning theorists. Such influence processes within the family are generally considered foundational, but they have been somewhat peripheral to research that has been aimed at examining PI effects.
Our view is that in families touched by incarceration, parents’ and other family members’ criminal behavior (especially substance use and violence) may be recurrent, observable features of children’s lives. Over the long haul, these aspects of family climate may not only expose the child to an array of stressful situations but also to a ready-made repertoire of attitudes, behaviors, and coping strategies that increase the likelihood of their eventual use. In addition, the realities linked to such family contexts may channel the child’s own developing social ties (peers, romantic partners) that then serve as proximal influences on behavior and reduced life chances. Consistent with our goal of further integrating these research traditions, however, we highlight that incarceration and other forms of system contact may themselves continue to shape parental attitudes and actions (including drug use and crime careers), the character of family networks, and children’s experiences as they form their own relationships. The PI literature has not been focused extensively on parental crime and associated network characteristics as “collateral consequences” with implications for child well-being; conversely, researchers who have focused on intergenerational transmission have not often grappled with the uniquely marginalizing effects of system contact.
Our goals here are primarily theoretical, and a general objective is to carve out a larger conceptual space in the life-course tradition for ongoing conditions, as well as for what we will call micro-events. The former are dimensions of family life that have a significant impact as recurrent, shared experiences. The latter are meaningful occurrences that punctuate life in general and the experience of one’s “home life” in particular, but that would not be considered a major transition or turning point. To locate further the experience of PI against the backdrop of this broader family landscape, we draw on analyses of qualitative and quantitative data collected in connection with a longitudinal study of a large, diverse sample of young people and their parents (the Toledo Adolescent Relationships Study—TARS).
1 |. BACKGROUND
1.1 |. Prior research linking crime and incarceration
Recent increased interest in the effects of incarceration on children has undoubtedly been influenced by studies indicating rapid growth in the use of incarceration within the United States, the large size of the prison population, and disproportionate effects on poor and minority individuals and families (Wildeman, 2009). The PI literature thus fits well in many respects with other basic criminological studies that have produced results revealing biases in sentencing decisions (Kutateladze, Andiloro, Johnson, & Spohn, 2014), inequitable practices associated with bail and other pretrial decisions (Arnold, Dobbie, & Yang, 2018), and deleterious consequences of system involvement, particularly for minority individuals (Pager, 2003). Yet even the findings from studies that have focused primarily on justice system disparities (e.g., by race/ethnicity and gender) indicate that prior offenses and current offense seriousness are robust predictors of the odds of incarceration and of sentence length (Steffensmeier, Ulmer, & Kramer, 1998).
Researchers have also documented that a high percentage of prisoners report a current history or background of substanceuse (Mumola&Karberg,2006). Consistent with such portraits, findings based on longitudinal follow-ups of adolescents have shown that early drug use increases odds of later incarceration (Slade et al., 2008). Yet in a recent analysis, Wakefield and Apel (2018) questioned whether most parents experiencing incarceration have exposed their children to crime directly, noting Hirschi’s (1969, p. 94) early view that even if a parent is committing criminal acts, he “does not publicize this fact to his children.” Nevertheless, the authors also presented National Longitudinal Survey of Youth (NLSY) data showing that adolescent respondents who went on to experience incarceration as adults reported higher levels of early substance use; greater frequency and variety of involvement in delinquent acts, gang, and other antisocial peer affiliations; and increased likelihood of carrying a handgun.
The parent’s criminal behavior is thus somewhat distinctive as a control as unlike other disadvantages, criminal involvement is generally a requirement for incarceration to occur.1 When viewed from a longer life-course lens, then, it is reasonable to expect that parents with incarceration experience will generally have longer, more involved criminal histories, relative to their counterparts who have not been incarcerated. Considering the “post”‘ side of the incarceration experience, criminological researchers have also documented high rates of recidivism and have found that relapses are common challenges for individuals with drug problems (Cullen, Jonson, & Nagin, 2011; Mowen & Visher, 2015). The findings of research on postrelease experiences are consistent with the focus in the PI literature on negative consequences of incarceration, but these findings reveal an additional rationale for considering parents’ behavior profiles as a potentially important component of what may be captured via a “PI experience” variable.
1.2 |. Disentangling effects of parental incarceration and the broader family climate
Some data sets that have been used frequently in studies of consequences of PI do not include detailed measures of parents’ criminal behavior and related aspects of BFC. For example, Roettger and Swisher (2011) found, relying on a nationally representative sample (The National Longitudinal Study of Adolescent Health—Add Health), that experiencing paternal incarceration was associated with increased odds of delinquency and arrest that persisted into young adulthood. Yet the authors noted the limitation that no questions were included in Add Health about the parent’s antisocial behavior, and thus, they could not rule out unmeasured effects of parental criminality. Unlike Add Health, analyses based on the Fragile Families and Child Well-being Study (FFCW) data often include controls for intimate partner violence (IPV) and parents’ substance use, and researchers have documented varying effects depending on whether these other factors are present. For example, they have shown that father’s incarceration is generally deleterious, but under conditions of IPV, the child may in effect be “better off” when the father has been incarcerated (Wildeman, 2010).
Although FFCW studies have thus included attention to other features of the family climate, a potential issue is the low reported rates of these parental behaviors within the FFCW study. For example, Geller, Cooper, Garfinkel, Schwartz-Soicher, and Mincy (2012) found that only 5 percent of respondents in families with an incarceration history had reported domestic violence at baseline, and less than 1 percent parents’ hard drug use in the past year. Similarly, Wakefield and Powell (2016) observed that the presence of parental drug use conditioned the effect of incarceration on children’s aggression, but they noted that based on available survey information, most fathers had evidenced “no red flags” (behaviorally) prior to their incarceration. When we juxtapose these low reported levels against findings based on the other lines of criminological research outlined earlier, however, it is not clear that questions included in such surveys have fully captured the nature and extent of these dimensions of family climate. At a minimum, for example, a focal incarceration event itself conveys that a level of involvement in criminal activity has unfolded prior to the period of confinement. Thus, to the degree that parental incarceration remains a marker for a higher level of offending or other problem behaviors within the family as well as system contact, some dynamics that connect to these unmeasured behaviors may figure into evaluations of incarceration effects. Consistent with this notion, in several surveys that included indices of parental behavior such as substance use, the findings revealed no link between PI and child outcomes once the parental behavior controls had been introduced (Phillips, Burns, Wagner, Kramer, & Robbins, 2002; Phillips, Erklani, Keeler, Costello, & Angold, 2006; see also Kinner, Alati, Najman, & Williams, 2007). In contrast, in other studies that included controls, their findings have revealed significant PI effects (see the meta-analysis conducted by Murray, Farrington, et al., 2012).
In recognizing these inconsistent results and the complexity of attending to “selection” problems (see, e.g., Copp, Giordano, Manning, & Longmore, 2018), researchers have recently drawn on a range of creative strategies to “zero in” more effectively on the specific effects of incarceration. For example, Porter and King (2015) compared Add Health respondents whose parents had experienced incarceration during childhood/adolescence with those whose parents would go on to experience a future incarceration when respondents were adults. Their results showed a significant effect of early relative to later PI on expressive but not instrumental delinquency. As the authors noted, however, a potential limitation is that although some aspects of propensity are accounted for, the early starting parents may yet differ in the seriousness/chronicity of their antisocial behavior profiles and related dimensions of BFC when compared with those whose incarceration only occurred later in life.2 Other scholars have relied on variability in criminal justice policies and procedures to account for these behavioral characteristics. For example, Wildeman and Andersen (2017) used a Danish policy change that expanded the use of noncustodial sentences (i.e., probation with community service) to compare the outcomes of children whose fathers were convicted of similar crimes. Their findings documented the effects of incarceration on the odds of children’s future criminal justice contact (see also Harding, Morenoff, Nguyen, & Bushway, 2018).
Scholars who have relied on such approaches have lent additional support to results shown in prior work using more traditional methods. A life-course lens is nevertheless potentially useful as we consider such incarceration events and their sequelae as components of a wider ranging set of family experiences that influence many aspects of children’s development, social network affiliations, and life chances. In this study, the longer life-course vantage point and more explicit focus on BFC will allow us to 1) evaluate critically depictions of BFC derived mainly from studies that have been focused more on the conceptual and measurement attention on PI, and 2) develop further our understanding of mechanisms underlying previously observed intergenerational trends. Family stress and trauma perspectives that have been developed as frameworks for studying PI effects offer a useful conceptual basis for exploring the impact of experiences associated with BFC. Our view, however, is that a comprehensive understanding of underlying mechanisms requires further integration of these traditions with the basic tenets of social learning theory (Sutherland, 1947).
1.3 |. Mechanisms underlying intergenerational effects
1.3.1 |. Family stress and trauma frameworks
The findings reported in the PI literature have highlighted multiple changes associated with PI and the stress this places on family functioning (see Arditti, 2012). Trauma-informed approaches are particularly useful as this perspective not only suggests multiple destabilizing consequences of PI for family life but also conveys that children’s strong emotional and even physiological reactions are important underpinnings of observed declines in children’s well-being. The results of quantitative and qualitative investigations have highlighted multiple types of such changes (e.g., those related to finances, including child support, housing circumstances, caregiver arrangements, parenting practices, and relationship quality/stability; Haney, 2018; Lopoo & Western, 2005; Schwartz-Soicher, Geller, & Garfinkel, 2011; Turney, 2014). Most researchers have focused particular attention on the experience of separation and feelings of loss as these changes have the potential to disrupt the parent–child bonding that is at the heart of healthy child development (Poehlmann-Tynan, Burnson, Runion, & Weymouth, 2017). Researchers have also focused on the effects of labeling and stigma, identifying spillover effects that may result in a pattern of children and other family members’ social withdrawal and isolation (Comfort, 2007; Foster & Hagan, 2015; Murray, 2007). In another important body of research, scholars have explored the impact of variability and the context of visitation and other forms of contact with parents during the period in which the parent is incarcerated (see, e.g., Poehlmann-Tynan, 2015).
Although issues such as visiting patterns are unique to the PI experience, basic processes identified, including family stresses, feelings of loss, and labeling may also be connected to the parent’s involvement in antisocial behavior and related dimensions of BFC. For example, a parent’s drug use in the period prior to an incarceration event may be perceived as stressful, associated with declining income, parental absences from the home, and marital conflict (Smith & Wilson, 2016). In addition, informal labels such as “crackhead” may be uniquely stigmatizing, adding to the marginalizing effects of labels formally applied (prisoner, ex-con; Copes, Hochstetler, & Williams, 2008; Pager, 2003). Furthermore, the child’s exposure to the parent’s behavior and BFC will often extend over a longer period of the child’s development, and to the extent that drug use or violence occurs within the home, may well be intimately experienced. Yet whether we consider the child’s response to the broader family climate, painful periods when a parent is incarcerated, or the whole package, dynamics emphasized by social learning theorists add an important dimension. Basic tenets of the theory address a core question, namely, why it is that young people often cope with these circumstances in the particular ways that they do.
1.3.2 |. Dynamics highlighted by social learning theory
Some researchers have found that parental incarceration is linked with well-being outcomes such as depression, but in several critical reviews, scholars have suggested that the most consistently observed effects are for behavior problems such as delinquency and aggression (Foster & Hagan, 2015; Murray et al., 2012; Travis et al., 2014). Effects have been documented in studies of externalizing problems in young children (Geller et al., 2012; Wakefield & Wildeman, 2014), as well as in analyses based on samples of youths followed into adulthood (Mears & Siennick, 2016; Murray, Farrington, et al., 2012; Swisher & Roettger, 2012). Thus, even when the conceptual focus is on the effects of the parent’s incarceration experience, the child often seems to evidence difficulties that look similar in contour to those of their parents. Such findings are generally evocative of social learning processes, but these dynamics have received little attention within the context of the PI effects research tradition.
In developing his theory of differential association, Sutherland (1947) emphasized that mechanisms involved in learning to commit crime are similar to learning other kinds of behaviors, and that intimate primary groups are likely key sources of influence. The family would seem to be an ideal learning lab as Sutherland emphasized that early, recurring, and significant associations are likely to be more critical to the process of acquiring definitions favorable or unfavorable to the violation of law. Despite the generally good fit—and foundational role of the family—more studies in the social learning tradition have been focused on the delinquency of peers (Akers, 2011; Haynie, 2002; Weerman, 2011). The frequent emphasis on the latter as a direct pathway to crime likely stems from several factors, including recognition that 1) peers become increasingly important around the time that delinquent behaviors begin to escalate, 2) youths often engage in these behaviors with friends, and 3) parents generally do not want their children to become delinquents (West & Farrington, 1977, p. 116). The idea that children lack full awareness of their parents’ actions is another theme that can be found not only in treatments of PI effects but also in connection with more general investigations of family influences on delinquency and research on intergenerational transmission (e.g., Farrington, Coid, & Murray, 2009).
It is thus traditional to concentrate on family factors such as attachment and supervision—things that may be missing from family life—that subsequently set up the conditions for the more obviously criminogenic influence of delinquent peers. Yet based on the results of a previous qualitative study of the lives of a sample of serious offenders and their children (∼70 percent of the parent sample had garnered adult incarceration experience), we theorized that tapping these family dynamics does not result in a complete portrait of the child’s lived experience nor of the family processes that foster intergenerational risk (Giordano, 2010). Attachment and supervision are important dimensions of parenting, but they do not fully capture what life is like within a given family, or do they reveal all of the sources of variability that may be systematically related to heightened risk for intergenerational continuity in criminal behavior.
2 |. CURRENT STUDY
In the current study, we follow up on the previous investigation by relying on a larger and more heterogeneous sample (TARS) that includes a broader range of respondents who vary in their exposure to PI and in the characteristics of their family climates. The study also benefits from a longitudinal design and longer period of assessment as these children were followed from adolescence into their young adult years. Consistent with the findings from prior research, we expect that PI will be negatively related to various indices of child well-being. In line with a social learning perspective, however, we draw attention to the broader family climate within which these experiences often unfold. We posit that a comprehensive treatment of mechanisms underlying negative consequences for child well-being, particularly behavior problems the child evidences, will often encompass 1) direct observation of parents’ problem behaviors; 2) ongoing interaction and communication that give depth and nuance to children’s understandings about behaviors that are desirable, tolerated under specific circumstances, or subject to negative assessments (Akers, 2011); 3) communications and observations related to other behavior patterns, attitudes, and world views that transcend the strictly criminal but add to the child’s risk portfolio nevertheless; 4) involvement with a “wider circle” of family members and associates that potentially increases the level of exposure to criminogenic influences; and 5) channeling effects on youths’ own developing networks that increase the odds of affiliation with “risky” peers and romantic partners.
Importantly, the parent’s actions and related dimensions of family climate may be directly affected by the incarceration experience. This finding is consistent with the general thrust of the incarceration effects literature, but in our view, these behavioral consequences—and their impact—have not been sufficiently highlighted. Because crime and incarceration are often closely linked and reciprocally related, then children may experience these family circumstances as a kind of gestalt or Home Life writ large, rather than viewing the parent’s incarceration as a clear turning point. Consistent with the emphases of trauma-informed perspectives, the child’s negative emotional responses to specific incidents or the totality of BFC (including PI as a component) are integral to understanding mechanisms linking family life to negative life outcomes. This idea of reactivity to all of these forms of family exposure adds an important dimension to the neutral tenor and traditional “passive vessel” depictions of the criminal learning process (Giordano, 2010). Furthermore, a life-course perspective on social learning highlights that as individuals mature, extrafamilial influences take on added significance as sources of definitions and influence. Yet our expectation is that although these ties may foster new life-course directions, experiences associated with PI and BFC often constitute a formidable package of family-related disadvantages that limit access to the conventionalizing potential of more prosocial peers and romantic partners.
In-depth interviews conducted with young adult TARS respondents who had experienced PI, as well as with other disadvantaged youth whose backgrounds do not include this experience, provided a basis for identifying key dimensions of the broader family context within which PI often occurs. These qualitative data access the young adult child’s perspective about where parental incarceration “fits” as an influence on their lives, and offer a unique perspective for evaluating assumptions contained within the existing PI literature. In particular, examination of these data provides a window on the relative salience of PI and the BFC from the child point of view and it helps us address the issue of children’s levels of awareness of parents’ criminal behaviors. We focus primarily on the experience of PI and the BFC with reference to the traditional formative years, defined here as prior to age 18. As these interviews were conducted in conjunction with a larger data collection effort, some observations about family life and effects that emerged during the process of analyzing the qualitative data were subsequently evaluated more systematically by relying on the full sample of respondents and the structured data.
The PI measure is derived from parents’ self-report and records searches of criminal backgrounds of mothers and fathers. The BFC is measured by a summary index that includes questions about parents’ IPV, use of coercive discipline, early problem behavior, and current drug/alcohol use, as well as children’s perceptions of a conflictual family environment while “growing up.” In addition to developing descriptive profiles of the PI and no PI subgroups, regression models were used to examine the associations between PI and BFC and two young adult outcomes (self-reported criminal behavior and intimate partner violence), net of neighborhood poverty, other sociodemographic characteristics, and traditional family factors (parental support and control).3 We then drew on latent class analysis to explore further our notion that PI and BFC and other forms of disadvantage often constitute a “package” of linked and reciprocally related risks. Finally, by relying on the structured and in-depth interview data, we explore the role of PI and BFC as influences on the character and impact of respondents’ own young adult network ties, as indexed by the criminal behavior of peers and romantic partners.
3 |. DATA
TARS was based on a stratified random sample of 1,321 adolescents and their parents/guardians. At wave 1, parents/caregivers (G1) completed a questionnaire, whereas responses of the core G2 respondents (adolescents at wave 1) were generally collected at respondents’ homes (using laptops). Data were collected in 2001, 2002, 2004, 2006, and 2011. The initial sampling frame of the TARS study encompassed 62 schools across seven school districts and was drawn from enrollment records of the 7th, 9th, and 11th grades. School attendance, however, was not a requirement for inclusion, and unlike in some community studies, respondents were thoroughly pursued for follow-up interviews, even if incarcerated. The stratified, random sample was devised by the National Opinion Research Center and includes oversamples of Black and Hispanic adolescents. Of the initial sample of 1,321 respondents, 1,021 valid respondents, or 77 percent of wave 1, were retained at wave 5. The results of attrition analyses are available in appendix B in the online supporting information.4
For both qualitative and quantitative analyses, parental incarceration experience was based on information provided in the parent questionnaire at wave 1, supplemented with court records searches of the backgrounds of 2,300 biological mothers and fathers of the G2 respondents. The question included in the parent questionnaire did not distinguish the gender of the parent incarcerated, and thus, for our analyses of the effects of parent gender, although not a central focus here, we had to rely exclusively on the records data (see appendix B in the online supporting information for more detail about the analyses related to the gender of the incarcerated parent). Based on the information from parent reports combined with the additional records searches, we identified 218 respondents who had experienced parental incarceration prior to age 18.
In-depth qualitative interviews were conducted with a subsample of approximately 100 TARS respondents at most waves of the study (for a more complete discussion of the criteria for selection, interview procedures, and analytic strategies, see appendix A in the online supporting information). Given our interest in risk behaviors and disadvantaged circumstances throughout the TARS study, we expected that some of these individuals would have PI experience, and we subsequently determined that 43 of the individuals who participated in the wave 5 qualitative component (the most recent round of interviews) had experienced parental incarceration. The result is a purposive sample, but we note that in comparisons with the total sample of PI-experienced respondents who completed the structured interviews, this qualitative subgroup did not differ on any of the focal variables or other sociodemographic indicators. In addition, although we concentrate our discussion here mainly on the most recent set of interviews (wave 5), where these or other respondents had participated in one or more of the earlier rounds of in-depth interviews, we also analyzed the narratives elicited at the earlier time periods (n = 39). Finally, the narratives of respondents whose parent(s) were generally disadvantaged but had not experienced incarceration were a further basis of comparison; for this analysis, we randomly selected 20 such respondents (from waves 4 and 5) for further study and contrasted their narratives with the PI-experience narratives.
Although these analytic steps and additional checks on emerging ideas increased our confidence in the results described, it is important to highlight the difference in our approach and studies specifically designed with a focus on the incarceration period. Several important studies have “zoomed in” on this period, and questions asked of respondents generally revolve around ways in which incarceration has affected relationships, family life, or child well-being (Braman, 2007; Wakefield & Wildeman, 2014). The interviews conducted in connection with this study are wider ranging and in effect “zoom out,” with the objective of understanding how respondents are doing as young adults, as well as of eliciting their views on factors that may have influenced their current life circumstances. Thus, to the degree that references to PI or the broader family climate were included within the narratives, it is because respondents considered these details salient as formative influences or explanatory details. The strategy of asking primarily about incarceration effects has the limitation of failing to consider many aspects of family climate we illuminate here; conversely, our approach does not permit a detailed exploration of PI effects close to the time in which the child experienced it, and based on specific questions designed to elicit this information.
In the quantitative analyses, we explore similar terrain but provide a more systematic assessment of linkages between PI, BFC, and two young adult outcomes (crime, IPV). We draw on responses to structured questions completed by the full sample and introduce traditional controls. For the quantitative analyses, we focus on parent reports about their behavior at wave 1 and respondents’ reports of criminal behavior and IPV as young adults. Some dimensions of the broader family climate (family conflict, witnessing parents’ IPV) are drawn from G2’s retrospective reports. Measures of young adult relationship characteristics (peer and partner criminal behavior) are drawn from wave 4. The analytic sample for the quantitative analyses includes all respondents with valid responses on the outcome variables and on PI status who participated in the structured interviews. These restrictions result in a final analytic sample of 656 respondents. Additional information about the quantitative analyses, including scale items and properties, is provided in appendix B in the online supporting information.
4 |. RESULTS
By drawing on the structured data and full sample, basic descriptive comparisons indicate that young adult respondents who had experienced incarceration of a parent prior to age 18 scored higher than other TARS respondents on self-reported criminal involvement as young adults, and they were more likely to report intimate partner violence (see table 1). Consistent with prior work, those in the PI subsample were more likely to be Black or Hispanic, live in a high-poverty neighborhood as children, and score lower on other socioeconomic indicators (mother’s education and employment). The two subgroups did not differ on the family support/attachment to parents and supervision scales, but they are consistent with our conceptual focus and expectations, PI-experienced respondents scored significantly higher (i.e., in the direction of greater risk/problem circumstances) on the BFC index. We note also that on average, the former scored higher on each individual component of the scale (i.e., more likely to report parents’ IPV, coercive parenting, early and current parent substance use, as well as evidencing higher scores on the items tapping overall family conflict).
TABLE 1.
Means/percentages for indicators of family context and sociodemographic characteristics according to history of parental incarceration for the full sample (N = 656)
| Variables | Full Sample (N = 656) |
History of Parental Incarceration (n = 218) | No Parental Incarceration (n = 438) | ||
|---|---|---|---|---|---|
| Mean/Percentage | SD | Range | |||
| Dependent Variables | |||||
| Crime | .72 | 1.15 | 0–6 | .98*** | .59 |
| Intimate partner violence | 21.80% | 33.03%*** | 16.21% | ||
| Peer crime | 1.82 | 1.71 | 0–6 | 2.31*** | 1.58 |
| Partner crime | .79 | 1.20 | 0–6 | 1.09*** | .65 |
| Independent Variables | |||||
| Parental incarceration | 33.23% | — | — | ||
| Broader family climatea | .62 | 8.34 | −8.83—40.52 | 5.03*** | −1.58 |
| Characteristics of young adult network | |||||
| Criminal behavior of peers (wave 4) | 1.38 | 1.59 | 0–6 | 1.73*** | 1.21 |
| Criminal behavior of romantic partner (wave 4) | .57 | 1.06 | 0–6 | .76** | .48 |
| Traditional Parenting Factors | |||||
| Parental support | 4.24 | .56 | 2.25–5 | 4.21 | 4.25 |
| Parental control | 3.40 | .38 | 1.43–4 | 3.42 | 3.39 |
| Sociodemographic Characteristics | |||||
| Gender (male) | |||||
| Female | 53.81% | 53.67% | 53.88% | ||
| Race/ethnicity (White) | |||||
| Black | 17.84% | 36.70%*** | 8.45% | ||
| Hispanic | 10.82% | 20.18%*** | 6.16% | ||
| Age (wave 5) | 25.41 | 1.79 | 22–29 | 25.36 | 25.43 |
| Mother less than high school | 10.52% | 20.64%*** | 5.48% | ||
| Mother employment | 78.81% | 71.56%** | 82.42% | ||
| Mother co-resident with child’s father | 57.47% | 28.44%*** | 71.92% | ||
| Neighborhood poverty | 13.32 | 13.98 | 0–70.13 | 21.28*** | 9.36 |
Broader family climate includes items tapping family conflict, parent IPV, coercive parenting, parent’s early problem behavior, and parent’s adult alcohol/substance use.
p < .01;
p < .001.
Use of the structured scales and larger sample offers an efficient means to assess basic differences in behavior and backgrounds of the two subgroups. Yet the in-depth interviews conducted with a subset of these respondents highlight the serious nature of the problem behaviors and relationship difficulties that characterized many of the PI-experienced young adults’ personal histories. For example, the no-PI subgroup includes 23-year-old Stephanie, whose narrative summary was labeled “a nurse who hates sick people,” and Robert, a chef trying to replace his seasonal job with full-time employment. This group also includes Jake, a 23-year-old college student who had experienced criminal justice contact, but told the interviewer this consisted of two nights in what he called “big boy jail” after hosting a party for his girlfriend’s 21st birthday (underage drinking). Thus, although those in the no-PI subgroup did not describe problem-free lives, most seemed to be making progress in navigating the transition to adulthood.
The PI group, in contrast, includes 24-year-old Alicia, who was interviewed in a federal prison where she was serving 17 years on a burglary charge, and 29-year-old Justin, whose interview began with a query from the interviewer about why the police were “out front” when she pulled up to his house (driving under a suspended license related to an earlier conviction for selling drugs). Another respondent, 30-year-old Corey, described the stresses associated with living in what he jokingly referred to as a “no-bedroom house” with his girlfriend’s three children and her parents, and a continuing battle with his ex-girlfriend Leah (“She’s on my text message saying you deadbeat fucker. I hate you. I wish you would fucking die, but it’s ok because child support just put a warrant out on you so enjoy your fucking freedom while you can.”). Corey’s background includes prior IPV, and the narrative raises the specter of future altercations (“if I could do something and get away with it then me and my son could just live happily ever after … I think that is one domestic violence case I would have to take”). The in-depth interview data thus tend to convey more effectively than self-report scores the seriousness of the behavior profiles and other problem circumstances of many PI-experienced respondents.5
5 |. HOME LIFE: PARENTAL INCARCERATION AND THE BROADER FAMILY CLIMATE AS FORMATIVE EXPERIENCES
The qualitative interview guide was focused heavily on respondents’ extrafamilial relationships and on their assessments of the success or perceived difficulties they had encountered as they had navigated the transition to adulthood. Nevertheless, respondents frequently brought up aspects of their earlier family lives in connection with their discussions of how they got to the “here and now” they were attempting to describe. On initial review of the lengthy life history narratives, it seemed that a small percentage of PI-experienced respondents included references to parental incarceration as an influence on the direction their lives had taken. In contrast, references to the broader family environment (e.g., parents’ drug use or IPV or other family violence or father belonged to a gang) seemed to be more commonplace. To examine this provisional observation more systematically, two blind coders tallied all such references in the set of 43 narratives of the young adult respondents who had experienced parental incarceration prior to the child’s 18th birthday. Their results revealed that across more than 2,068 pages of transcribed materials, only 5 percent (2 of the 43) included references to PI experiences in connection with their discussions of their formative years, and an additional 9 percent (n = 4) referenced a more recent period when the parent had been in jail or prison. In contrast, approximately 60 percent (26 respondents) included discussions of the aspects of the broader family climate, and all of those who mentioned parental incarceration also referenced parental deviance/family climate issues. Furthermore, the descriptions of BFC were often detailed as respondents referenced ongoing conditions as well as specific incidents that illustrated a more general point about their home lives (our notion of micro-events).
The results of the analyses of these family-related sections of the narratives indicate the following: First, many respondents seemed to have intimate knowledge about problematic aspects of the parents’ behaviors, an observation that addresses the issue of the observability of actions that may have been associated with parents’ criminal justice contacts. Second, descriptions of parental drug use, IPV, and other criminal behaviors were frequent themes across the narratives of many respondents, suggesting that these aspects of family background are common, rather than rare occurrences, as has been suggested in some prior work in the PI tradition. Third, respondents often specifically connected these experiences to their own current circumstances, suggesting a greater perceived impact of BFC relative to consequences that they linked directly to the PI they had also experienced. Finally, respondents’ discussions of family life frequently referenced the problem behaviors of family members other than the focal parent with incarceration experience. This suggests that whether viewed from the lens of the family stress/trauma perspective or basic tenets of social learning theory, a comprehensive treatment of the BFC and its impact will necessarily include attention to the behavioral repertoires along with criminal justice exposure of these other members of the family.
Clearly, the results of prior studies, particularly those in which researchers have relied on designs that have allowed them to zero in on incarceration effects, have highlighted the negative consequences of incarceration on a range of childhood and young adult life circumstances. Nevertheless, it is of interest that within this sample group, parental incarceration was not more often narrated as a key turning point, formative influence, or feature of their early family lives. It is likely that interviews with the same individuals as younger children would have produced more references to parental incarceration and, importantly, some PI-related factors that may have deeply affected their lives (e.g., a parent’s inability to gain employment as a result of felony conviction) might not be as apparent from the child’s perspective. A further consideration is that, particularly within the context of a community sample, incarceration experiences may consist of one or more short jail stays rather than of a prison sentence of greater duration. With these caveats in mind, a tentative conclusion is that from the child-as-young-adult vantage point, BFC often seemed to loom larger as a set of core family-related concerns and influences.6
5.1 |. References to PI focused on the “formative” years
We begin our discussion by focusing on the narratives of Ryan, age 26, and Whitney, 27, the two respondents who brought up PI in connection with recollections of their childhood and adolescent experiences. Ryan’s account was coded as forging a “direct link” between incarceration and behavior/well-being, but in this instance, Ryan told the interviewer that his best memories of childhood were the years when his grandparents had custody, after his mother was sent to prison:
She was hanging around with the wrong crowd … booze, smoking crack … that’s why when my grandma took custody of us I was happy. Cause I remember the hard times my mom and them had, and we would call our grandpa all the time, and he’d come pick us up and take us away from all the problems, and when they told us they were taking custody of us, it was real nice.
Ryan referenced problematic aspects of his family life as ongoing conditions that preceded his mother’s incarceration. In addition, this respondent vividly recalled specific incidents (micro-events) associated with this phase of his childhood (i.e., those times he had to call his grandparents) that deepens the portrait of significant stress and instability during this period. The sequence Ryan described accords well with court records, which revealed that prior to the incarceration period, Ryan’s mother had been arrested for theft and passing bad checks charges when Ryan was 3 and 4, multiple soliciting charges around age 7, forgery when this respondent was 10, and felony child endangerment at around age 15. Ryan’s narrative also revealed that he had witnessed IPV (“my mom and dad always fought. When we were younger, that’s where we witnessed it all … my dad would hit my mom”).
Ryan’s recollection of the positive times when he lived with his grandparents would thus seem in some respects similar to previous work in the “conditional effects” tradition. As noted earlier, researchers have found that under some conditions (e.g., the presence of IPV in the home), the child might well be “better off” when a parent is incarcerated. Yet a component of such arguments is that these problematic features of family life are uncommon. The frequent references to parental substance use, IPV, and other difficulties discussed across many of the in-depth interviews (along with the group-level differences in the structured BFC index shown in table 1), are not entirely consistent with this assessment. For example, when we relied on a limited single-item indicator included in the BFC index, we found that 49 percent of those with PI backgrounds reported that they had witnessed parents’ IPV, a figure more than double the reports provided by the non-PI respondents, and considerably higher than the 5 percent figure reported in connection with an FFCW analysis (Geller et al., 2012).7 Furthermore, based on information gleaned from discussions of the longer time period covered by the narrative accounts, this figure likely represents an underestimate of total exposure. For example, Ryan not only focused attention on his parents’ IPV (consistent with the wording of the structured item), but also he brought up his mother’s subsequent victimization by a later partner (“and uh my mom’s boyfriend would hit her”).
The life-course view afforded by these narrative accounts indicates additional complications to the “better off” implications of conditional effects analyses. Although the parent’s time in prison may for some youths provide a respite from the instabilities and stressful situations related to BFC, for most children, the parent’s confinement will not span the whole of childhood and adolescence. Thus, the apparently “better off” outcome will necessarily reflect a provisional determination. Yet as researchers have reported in the broader literatures on reentry problems and drug relapse odds, parents’ lives after jail or prison often come with disappointments and reduced life chances, and for some, with a return to previous patterns, including drug use and other criminal behavior (see, e.g., Mowen & Visher, 2015). These later stages are also integral to an understanding of the child’s family history and subsequent development. Ryan described his experiences later on in high school:
[Then] my mom got custody of me, at the age of 16 and I moved to her house. And I started just, I wouldn’t have to go to school. If she would’ve [made me go], I woulda graduated high school, but she didn’t make me. I woke up, 11 o’clock just laying there. Smoke some weed and that’s about it. Now I wish I wouldn’t have done that but you can’t go on in the past. She always told me I’d rather you smoke it with me than go out here and get something laced and end up dead. And I thought it was the coolest thing to be a mom to do that, but on the other hand, I don’t think it’s so cool [now].
Ryan indicated that his family life during this time included routines (waking up late, smoking weed with his mom) that negatively affected his academic progress. We also see in this example some dynamics emphasized by social learning theorists, including direct observation of his mother’s behavior, as well as parental communications that indicate a general acceptance of Ryan’s use of marijuana, including a rationale for smoking at home. We can also use this simple example to highlight that attitudes and actions related to noncriminal arenas (here, school) contribute to the portrait and potential impact of BFC, as Ryan’s mother apparently did not press about specific homework assignments, or emphasize the importance of school attendance.
The narratives were a primary source for developing our ideas about the general importance of BFC for a comprehensive view of family factors that, along with PI experiences, are likely to influence child behavior and well-being. We also drew, however, on the larger sample and multiple regression analyses to explore the issue of the relative salience of PI and BFC with respect to two specific outcomes—young adult criminal involvement and intimate partner violence. As shown in table 2, bivariate results indicate that PI is significantly associated with self-reports of young adult crime and IPV. In full models that include the BFC index and other disadvantage controls, however, the association between PI and the two outcomes is no longer significant. In supplemental models that were focused on paternal and maternal incarceration, the results reveal a similar pattern of associations (see table B1 in appendix B in the online supporting information). We note that in each model examined, BFC alone does not fully mediate the PI association, which indicates the importance of considering disadvantage along with (behavioral) indicators of family context. Furthermore, consistent with the long window of assessment and our theoretical discussion, some aspects of BFC may have been influenced by the parent’s incarceration experience.
TABLE 2.
Coefficients and odds ratios for the negative binomial and logistic regression of crime and IPV on parental incarceration, family context, and sociodemographic characteristics (N = 656)
| Crime |
IPV |
|||||||
|---|---|---|---|---|---|---|---|---|
| Zero Order |
Model 1 |
Model 2 |
Model 3 |
Zero Order |
Model 1 |
Model 2 |
Model 3 |
|
| b | b | b | b | OR | OR | OR | OR | |
| Parental Incarceration | .503*** | .313* | .238 | .036 | 2.549*** | 1.626* | 1.173 | 1.071 |
| Broader Family Climate | .036*** | .030*** | .029*** | .012* | 1.082*** | 1.071*** | 1.064*** | 1.058*** |
| Characteristics of Young | ||||||||
| Adult Network | ||||||||
| Criminal behavior of peers (wave 4) | .408*** | .275*** | 1.269*** | |||||
| Criminal behavior of romantic partner (wave 4) | .430*** | .215*** | 1.340*** | |||||
| Traditional Parenting Factors | ||||||||
| Parental support | −.143 | −.125 | −.065 | 1.002 | 1.151 | 1.206 | ||
| Parental control | .241 | .344* | .291* | .980 | 1.102 | 1.052 | ||
| Sociodemographic | ||||||||
| Characteristics | ||||||||
| Gender (male) | ||||||||
| Female | −.734*** | −.774*** | −.550*** | .932 | .869 | .943 | ||
| Race/ethnicity (White) | ||||||||
| Black | .514*** | .203 | .166 | 2.298*** | 1.123 | 1.134 | ||
| Hispanic | .431* | .203 | .012 | 2.440** | 1.416 | 1.362 | ||
| Age | −.065 | −.059 | −.024 | 1.011 | 1.018 | 1.036 | ||
| Mother less than high school | .209 | −.125 | −.174 | 2.097** | 1.003 | .993 | ||
| Mother employment | −.276 | −.213 | −.221 | .677 | .872 | .866 | ||
| Mother co-resident with child’s father | −.360** | −.056 | −.014 | .385*** | .636* | .638 | ||
| Neighborhood poverty | .011** | −.002 | .007 | 1.031*** | 1.009 | 1.012 | ||
| Overdispersion parameter | .997 | .702 | .145 | |||||
| AIC | 1,510.85 | 1,478.81 | 1,324.54 | |||||
| BIC | 1,528.79 | 1,541.62 | 1,396.32 | |||||
| X2 | 59.59*** | 70.05*** | 81.44*** | |||||
p < .05;
p < .01;
p < .001.
5.2 |. Broader family climate and parental incarceration: Often a package deal
A limitation of regression analyses is that they are not ideal for conveying the frequently linked, reciprocally related nature of these experiences. To illustrate, Whitney, the second respondent who included a reference to her father’s early incarceration history, developed a narrative that accords with our notion of a cluster or package of related risks. Whitney specifically brought up her father’s incarceration, but she did so in a way that connected the experience to other aspects of her family background: “As soon as I turned 18, I rolled out. I couldn’t take it no more. My mom was, they was together for years, but then they broke up but she was on drugs … he used to beat her … he was in and out of jail.” Whitney’s conclusion that she “couldn’t take it no more” referred specifically to life at home with her father (the parent with incarceration experience); in turn, this living arrangement in her mid-to-late high school years was necessary as a result of her mother’s death from a drug overdose. Whitney also stressed that her father’s use of violence extended to her and to her sister (“my daddy used to beat everybody—me, mom, my sister”) and offered a straightforward explanation that resonates with a social learning emphasis (“he used to get a whooping by his mom a lot”).
The results from a latent class analysis (see table 3 and figure 1; for further details about procedures, see appendix B in the online supporting information) reveal additional support for the idea that these family experiences tend to be interrelated. The findings indicate that a two-class solution fit these data well; that is, within the context of this community sample, these features of family life tend to be linked. Respondents in class 1 were characterized by the experience of parental incarceration and higher than average scores on the family climate index, as well as by being more likely to be minority youth, reside in a family structure type other than “two-parent biological” during adolescence, and live in a high-poverty neighborhood. Conversely, it was unlikely that the backgrounds of those in class 2 included these features. Of interest here is the apparent absence of other theoretically relevant classes (i.e., high scores on BFC but no PI experience or low BFC but experienced PI). And, as would be expected based on the regression results, membership in class 1was a strong predictor of the two young adult outcomes. This notion of a tightly coupled package of risks is not incompatible with findings of research that are focused on the negative effects of PI. Yet the qualitative and quantitative results indicate the need to give additional weight to the broader patterns of behavior and other features of family life that are, along with various other forms of disadvantage, frequently associated with incarceration experience.
TABLE 3.
Model fit information for competing latent class models (N = 656)
| Number of classes | G2 | df | AIC | BIC | Entropy |
|---|---|---|---|---|---|
| 1 | 598.45 | 26 | 608.45 | 630.88 | 1.00 |
| 2a | 63.49 | 20 | 85.49 | 134.83 | .77 |
| 3 | 28.56 | 14 | 62.56 | 138.83 | .81 |
| 4 | 9.53 | 8 | 55.53 | 158.71 | .70 |
| 5 | 3.87 | 2 | 61.87 | 191.96 | .70 |
Selected as final model.
FIGURE 1.
Probabilities of parental incarceration, exposure to the broader family climate, nonintact family structure, minority status, and neighborhood poverty by class membership
5.3 |. Key dimensions of the broader family climate
5.3.1 |. Close to home and beyond
Ryan and Whitney’s narratives are useful in illustrating the degree to which these respondents saw PI and BFC as a fundamentally intertwined set of family-related experiences. Yet it is important to reiterate that these narratives are not typical, in that most (95 percent) PI-experienced respondents did not include any references to their parents’ earlier incarceration histories. Thus, the results of our analyses of the larger set of narratives without explicit mentions necessarily contributed to our view of the important role of the BFC for understanding life-course paths and associated behaviors of these respondents.
The narratives of this larger subgroup were generally similar to the sections of Whitney and Ryan’s accounts that were focused on BFC, particularly in the frequent references to parents’ substance use, as well as to various forms of IPV. Consistent with our focus on ongoing conditions, respondents’ references make clear that these actions often occurred within the home, and tended to be recurrent features of the child’s experience. In addition, their comments, alongside information gleaned from court records, at times offer a nuanced appreciation of the sequencing of these linked family experiences. For example, Ryan, quoted earlier, had pointed to the instability in his family life prior to his mother’s incarceration. His comments are similar to those of Marcus, a respondent in the no-mentions subgroup:
I feel like I kind of grew up fast. Based on my lifestyle and where I came from, just all the things that were thrown at me.… Well my parents separated when I was young. But they lived together, so I seen my dad fighting and I grew up with drug abuse with my dad for years. Waking up and seeing him passed out in the living room with a beer and pills, and whatnot [later on, crack].
Marcus’s discussion of his early family life illustrates our more general conceptual argument about some limitations of the life-course perspective’s heavy emphasis on transition events. In this example, the accumulation of closely observed everyday experiences and small-scale events (seeing his dad “passed out in the living room”) may be stressful, formative, and consequential. And in his case, the ongoing conditions that characterized his home life seemed to precipitate a more obvious turning point—a decision to move out of the house when he was still in his early teens (“I kind of wanted to get out of the house so I left the house and lived on the south end. Moved in with my friends and was taking care of myself and I was like twelve or thirteen years old, out running the streets all night”). Marcus also made it clear that it was during this period that he began to commit robberies and burglaries with his friends. Of note, his move and the beginning of a “street life” occurred prior to his father’s incarceration: Records indicate that his father spent 2.5 years in prison for forgery on an arrest that occurred when Marcus was ∼14. Arrest record data also support his general portrait of life at home prior to that time, as in the years before his committing offense, his father George had accumulated a series of domestic violence charges when Marcus was 7, 9, and 12.
It is also important to note that although Marcus emphasized his difficult childhood, his comments throughout nevertheless reflected caring and concern for his father, and continuing contact with him as an adult. For example, at the most recent interview, Marcus expressed worry that his father had developed a serious heart condition (“he’s got stents and stuff”), but he noted that George continued to use crack and drink heavily on a daily basis. Thus, it is perhaps not surprising that the family-centered sections of his life history narrative are focused mostly on the role of his father’s decades-spanning problem with substance use rather than on his incarceration experience.
Although many of the descriptions of “home life” were focused on what life was like “at home,” some respondents conveyed knowledge, albeit imperfect, about parents’ actions that took place farther afield. For example, the incarceration literature has been focused primarily on the negative effects of father absence as a result of incarceration, but 27-year-old Craig’s discussion of parental absence during the formative years was focused on his father Jacob’s involvement in an organized drug gang:
My father was never around. He was involved in the drug game a lot, and he was wanted by some drug guys, you know.… He didn’t want us to know nothing about it, but we knew.… He owed them a lot of money—and he couldn’t come around us. If they found him around us, they would have killed him right then and there, and us.
Although father absence was a significant theme within the narrative, Craig, in his early-years discussion, also referenced periods of direct contact. This aspect of his discussion echoed the more common themes of drug use and violence (“yea, I watched my father pistol whip my mother … when we was living in Deshler”). Craig also described a vivid memory (another example of a micro-event) that seemed in context designed to illustrate the seriousness of his father’s involvement in illegal activities. Craig recounted the story of the first time he had held a gun in anticipation of a potentially dangerous drug handoff: “My dad comes running in the house and he threw me and my brother guns … and said, here if they, if any, if it goes bad, aim for the heads.” In this example, he conveys an awareness of his father’s involvement in illegal ventures, yet the vividness of his recollection also indicates that he may have considered this a stressful if not traumatic early experience. And although this respondent does not explicitly “connect the dots” to his own later involvement in illegal behavior, we note that Craig is currently serving time for an offense that includes a gun specification (along with aggravated burglary and attempted murder).
Other respondents forge more explicit connections between the parent’s actions and the respondents’ subsequent patterns of behavior. For example, Whitney, quoted at the outset, indicated that after her mother died of a drug overdose, she was “kinda in a funk that was hard to get out of,” and during this period, she got “deeper into drugs” as a way to cope. Chad’s narrative is similar, in that he recognized that he was headed in the direction of problems that had derailed his mother’s life:
When she passed, I started using more to block out the pain, you know, not think about it. It really tore me up … that’s when I started injecting heroin.… But that was the best way of blocking out the pain for me. Once I lost her, I felt like I lost everything. It’s not that I thought it would be different—I just didn’t care.
5.3.2 |. The wider circle
Although we have focused primarily on the behavior of the parent who had been incarcerated, the content of the narratives (and by inference the reach of BFC) is not as clearly delimited. Thus, during the process of analyzing these materials, it was sometimes difficult, without referring to the official records or parent questionnaire, to figure out which of the parents had accumulated the incarceration experience. The related phenomenon of children experiencing the incarceration of additional family members has been addressed within the context of the PI literature but primarily as this adds to feelings of personal loss, rather than as a window on behavioral influences across the wider family circle (Wildeman & Wakefield, 2014). Yet the content of these narratives aligns with the results of prior criminological research in which a trend toward assortative mating, sibling influences on crime, and the concentration of offending in families is documented (Farrington, Jolliffe, Loeber, Stouthamer-Loeber, & Kalb, 2001; Krueger, Moffitt, Caspi, Bleske, & Silva, 1998; Slomkowski, Rende, Conger, Simons, & Conger, 2001). Thus the other parent, subsequent partners, siblings, and other individuals in the parent’s orbit contribute significantly to family climate and by extension to social influence. As an example, Whitney, quoted earlier, experienced the death of her mother from a drug overdose, even though her father was coded as the parent with the incarceration experience.
Researchers interested in the experiences of incarcerated women and their children have included attention to partner characteristics (see, e.g., Wildeman & Turney, 2014). In other studies of PI effects, however, discussions about the partner or other family members’ actions have generally concentrated on reactions to the incarceration event (e.g., effects on financial circumstances and social withdrawal), rather than on more general assessments of their behavioral histories. Thus, for example, although it is intuitive to consider some family members—notably grandparents—as a generally more prosocial influence, this cannot be assumed a priori (Lance: “My grandpa had my grandma, he had her face in the bathtub full of water…, and I started crying and like ‘grandpa leave grandma alone!’ ‘Leave grandma alone!’”). These individuals are important as additional reference others, and because they may be called on as caregivers—in general, and when a parent has been sent to prison (Hanlon, Carswell, & Rose, 2007; Johnson & Waldfogel, 2002):
My uncles and their girlfriends… it’s, it [intimate partner violence] was pretty common in my family. I remember cause my grandparents raised me and my sisters. I remember them always arguing and stuff, and my grandpa, he would get real aggressive with my grandma. And we would have to get in the middle of it. Like as far back as I could remember. I was little, little. (Krista, age 25)
It, just my family’s just really aggressive and outspoken and just mean and they’re always fighting. Literally at every function that we have, somebody’s fighting … and then with my stepdad … I kinda blame him because maybe if she woulda picked somebody better I woulda turned out better. I think watching them fight all the time made me violent. (Rebekah, age 26)
My cousin he was doing robberies before … when he broke down his plan to me it just seemed like you know this is a definite go and you know there was really nothing to worry about [Alicia is in prison for this offense]. (Alicia, age 24)
Although a comprehensive treatment of sibling effects is beyond the scope of this analysis, clearly attitudes and behavior profiles of these similarly aged family members contribute to the child’s experience of a particular family context, and potentially to the child’s risk portfolio. Principles of identification are strong, and siblings are generally in the process of accomplishing similar developmental tasks. Jeff forges a direct link:
I grew up with, you know, my sister and her ex a lot and they were abusive, you know, they talk to each other in certain ways, yea like beating each other or whatever, and I guess I was around them a lot so I must’ve … picked it up in the back of my mind and stored it there and one day just used it I guess.… And maybe consciously, picking up that’s how I need to treat a girl, cause he obviously had a girl, and he was keeping her, and that’s what I wanted.
In this example, Jeff’s reference to “picking it up and storing it” and “one day just used it” seems valid as a depiction of family-related dynamics emphasized by social learning theorists. Yet it is also important to consider in the current context the robust finding that criminal involvement and substance use are often linked to friendships with others who have similar behavioral profiles (Haynie & Osgood, 2005; Matsueda & Anderson, 1998). Thus, the parent’s affiliations outside the family also have potential implications for the child’s level of exposure to various forms of risk. To illustrate, Tasha’s father could be characterized as a low-level offender. His incarceration history includes a night in jail on a DUI/drug abuse charge when Tasha was 4, followed the next year by a 20-day period in the Toledo workhouse after he was convicted of this offense. Yet her father’s court records include subsequent criminal justice contacts (e.g., DUI, concealed weapon, domestic violence, disorderly conduct, and trespass) that were not associated with additional jail time but revealed patterns of behavior that coordinated well with Tasha’s account. This 27-year-old respondent told the interviewer that her father and mother fought a lot, although she never directly observed them hitting each other:
My dad, however, was an alcoholic … and that did take a toll on us when we were younger, um, cuz my dad did have a lot of parties. He had a pool table in the basement and I was um, raped by one of his friends when I was 14. And my brother had heard me screaming one night and finally walked into the room ‘cause he thought I was dreaming or something.
5.3.3 |. Beyond crime
As we noted earlier, attitudes and behaviors that are not strictly speaking “criminal definitions” are a large part of what parents and other family members communicate. Dimensions of family climate that relate to these other domains are nevertheless foundational as they shape the child’s conduct with respect to a host of life experiences with implications for outcomes such as crime and IPV. Examples of this wider arena of definitions include Ryan’s comments about never having graduated from high school, as well as his regret that his mother had not insisted that he attend school each day. Similarly, Jeff not only observed IPV in his sister’s relationship but in the “way they talked to each other” as well. These broader patterns of communication Jeff references are important aspects of socialization within the family; in numerous studies, scholars have documented that negative styles of communication are a robust predictor of IPV (Giordano, Copp, Longmore, & Manning, 2015). Mackenzie’s discussion provides another example of such relationship “lessons” in reflections about her father’s romantic relationship experiences: “My dad played women my whole life. That’s why I don’t trust nobody. They always cheat” (see, e.g., Miller, 2008). A comprehensive treatment of such attitudes and behaviors is clearly beyond the scope of this investigation, but in our view, these noncrime understandings constitute critical, under-researched dimensions of BFC.
5.3.4 |. Linked lives: PI, BFC, and the next generation of associations
From the narrative excerpts quoted earlier, we can point to specific dimensions of family life that may precede and follow incarceration, and that are likely to influence overall well-being and outcomes such as criminal behavior and IPV. Yet as children begin to navigate the transition to adulthood, extrafamilial relationships gain in importance as sources of reference, support, and companionship. These changes provide new contexts that are integral to the “continuous development” assumptions of the life-course perspective. At the same time, social proximity, shared understandings, and labeling and exclusion dynamics may favor elements of continuity in the character of network ties. Mackenzie provides perhaps the most compact example of the idea of network channeling as this respondent told the interviewer that she had dated and eventually moved in with her father’s drug dealer Devon. Suggesting at least a tacit acceptance of her involvement with Devon, Mackenzie noted that her father allowed the couple use of his car in exchange for access to Devon’s supply of drugs. Mackenzie’s involvement with Devon thus continued the general trajectory, rather than providing entrée to a new set of social relationships that could support a more prosocial life direction.
To examine these linkages more systematically across the full sample, the results of analyses indicate that net of traditional socioeconomic disadvantage and other controls, both PI and BFC are associated with higher levels of young adult peer and romantic partner criminal behavior (see table B3 in the online supporting information).8 In turn, in full models focused on the behavioral outcomes (see table 2), peer and partner criminal behavior are significantly related to young adult criminal involvement. Furthermore, as shown in table 2, romantic partner criminal behavior is significantly linked to the respondent’s odds of reporting IPV (see also Herrera, Wiersma, & Cleveland, 2008).
The narrative data add to the portrait as we can use them to point to specific pathways through which “home life” may influence the character of young adult social ties. For example, Marcus (the young man who moved in with his friends) described a sequence that has long been emphasized in traditional criminological theorizing (see, e.g., Shaw, 1930). Marcus indicated that ongoing family problems were related to his premature “exit” from the home and increased susceptibility to delinquent companions (“[We were] doing the usual stuff that kids do over there, you know causing trouble … breaking into stores, robberies of people who were walking on the street … I was around it.”). In contrast, Mackenzie’s romantic partnership seemed to result from direct exposure to her father’s contacts. Other respondents pointed to the effects of labeling and social exclusion as a result of aspects of BFC, as referenced in Brandi’s and Tasha’s narratives.
Brandi focused primarily on difficulties associated with life with her mother, even though her father was absent as a result of incarceration a significant portion of her childhood and adolescence. In her longer narrative, however, she also forged an explicit connection between life at home and her friendships as a young adult. This respondent’s description of her family life was included in response to a basic question about whether she still “hung out” with the young girl who lived in the house immediately behind hers:
No. No, because it’s, we have different beliefs and everything that I choose to do is what they choose to stay away from. You know, everybody has grown up, I mean … I—I had a shitty ass mom, you know what I mean, instead of toilet paper and paying rent she, we’d have to wipe our ass with coffee filters, you know what I mean. We smoked a lot of weed, but really we got kicked out every month—always flip floppin’ back and forth.
This respondent’s discussion of growing apart from her friend started out with an agentic view of their different orientations. Yet Brandi subsequently pivoted to a more general description of her earlier family life, illustrated with references to specific micro-events (the toilet paper incident, getting kicked out of apartments) to suggest why her childhood friend may have distanced herself from Brandi and her family. In this instance, focusing primarily on the stigma of parental incarceration, which is also an important part of her background, would not provide a comprehensive picture of all of the family circumstances that she believes have set the two young women on different paths.
Twenty-seven-year-old Tasha, whose family experience was described briefly earlier (raped by father’s friend at age 14), developed a long-term relationship with Brad, whose own family life seemed to have been even more difficult. Tasha felt that Brad understood all that she had gone through, however, particularly lingering depression related to the earlier victimization. Thus, their backgrounds of difficult family lives may have been part of their connection, even as this linked her to a partner and family network characterized by an even more limited supply of valued social capital:
He’s got his sensitivities. He’s grown up with a mom that was a crackhead. I mean, he’s turned out pretty good for considering how his family is. He just prayed that he would find somebody to accept him for him, and that wouldn’t judge him for his family. [later] The first two years together we were both doing drugs. On my birthday he um broke my arm.… It was more because he was hanging out with a couple of his buddies, sitting there doing drugs and he didn’t want to give me none, basically. And I was asking for ‘em, trying to hold onto him and my arm got twisted in the wrong way, and it broke.
Similar to Brandi, Tasha references negative social judgments based on family background, providing another illustration that the parent’s incarceration status is but one element of a larger set of social “sorting” and labeling processes that may influence the character of young adult social ties (Bryan, 2017). In this instance, concordance on family background may have conferred a level of shared understandings, but it did not minimize and indeed likely amplified the level of risk (i.e., Tasha’s references to drug use, IPV, and affiliation with drug-using friends).
6 |. DISCUSSION
There is widespread agreement that the heavy reliance on incarceration in the United States has negative effects on individuals as well as for the health of communities and societies as a whole (Western & Pettit, 2010). The findings from studies of the effects on children reveal significant consequences of PI for well-being and behavior that often presage longer term patterns of intergenerational inequality (Wildeman, 2009). In recognizing the other kinds of disadvantage such children often face, researchers have increasingly drawn on statistical techniques and on a variety of other creative methodological strategies to distinguish the effect of incarceration from other co-occurring sources of disadvantage. Theoretical treatments of mechanisms associated with negative effects on children have been focused on stress/trauma associated with the parent’s incarceration, as well as on the realities of tangible (e.g., housing and finances) and more subjective (e.g., feelings of stigma and decreased attachment to parents) changes that connect to the PI experience.
In contrast to studies in which researchers have zeroed in on the incarceration effects, here we relied on a longer life-course lens and accessed the perspectives of young adults who had experienced PI during childhood or adolescence. With this vantage point, we could examine the experience of PI within a BFC. We argued that the parent’s criminal behavior is a specific kind of disadvantage that has an intimate connection to incarceration but that has not been sufficiently foregrounded in studies that fall within this tradition. The findings from research on incarceration and its effects indicate strong support for key aspects of labeling theory, given inequalities in the application of formal sanctions, and consequences of system involvement for a broad range of life chances. Nevertheless, criminal behavior remains a significant predictor of the odds of experiencing incarceration.
The need to nudge the parent’s behavior patterns into a more prominent position conceptually and empirically is generally supported by results from other criminological traditions that have not been focused specifically on PI and child well-being. Research findings have shown that prior offending and early onset of delinquency are related to later incarceration odds, and scholars have documented that a high percentage of inmates’ backgrounds include previous or current substance use. Furthermore, in studies of postrelease adjustment, researchers have pointed to high rates of recidivism. Within the context of a traditional survey sample, then, parents with a PI background will also tend to have more significant behavioral histories relative to their counterparts who have not had system contact. The results of analyses of the TARS interviews reveal support for this assessment. We drew further on qualitative and quantitative data from this study to illustrate specific ways in which these realities may be experienced at the family level, over the long haul, and from the child’s point of view.
Most PI-experienced young adults interviewed focused the family-related sections of their narratives on issues such as parents’ drug use, violence within the family, and related aspects of BFC. The respondents often included detailed descriptions of their home lives, sometimes incorporating discussions of what we labeled “micro-events” to illuminate a more general point about the character of the BFC. Conversely, few respondents incorporated specific mentions of the PI experience. These accounts are clearly retrospective, but they can be used to provide a unique perspective on the issue of the perceived impact and relative salience of what are often linked family experiences. The quantitative results based on structured indices and the full sample indicate support for the need for additional research on ongoing family conditions, as well as on the effects tied directly to the PI experience. Consistent with prior work, respondents who had experienced PI prior to age 18 reported higher levels of young adult crime and IPV when compared with those who had not experienced the incarceration of a parent. Yet these associations were no longer significant in models that included an index of BFC and other disadvantage controls.
We do not conclude from the quantitative and qualitative results that the effects of PI are spurious. Given the long window of assessment, aspects of parental behavior and family climate are likely to have been affected by the marginalizing effects of incarceration experience(s). Yet “ongoing conditions” of family life, including parents’ antisocial behaviors, are critically important as these occur “up close,” face-valid as formative, and generally last longer than the incarceration periods. And, given the limited measurement of parents’ behavior and related aspects of climate in many studies of PI effects, it is likely that PI remains a marker for a range of BFC dynamics as well as for tapping incarceration-specific effects. We also provide a specific example of a more general critique of the traditional emphasis of the life-course perspective on timed events and turning points.
An efficient theoretical approach would be simply to extend the family stress/trauma perspective to cover these other aspects of the child’s experience. Research findings on stress processes have highlighted the negative effects of chronic stressors as well as those tied to major life events (Avison & Turner, 1988). Yet we argued here that dynamics identified by social learning theorists add an important dimension to our understanding of underlying family processes associated with intergenerational trends. This study’s findings and those of other research in the PI tradition reveal similarities in the parent’s behaviors and specific difficulties later evidenced by their children. Such findings highlight the play of more basic social influence processes, as sometimes outlined in “linking statements” included in the narrative accounts (“my family’s just really aggressive … they’re always fighting … I think watching them fight all the time made me violent”). Thus, as stressful situations unfold, both related and unrelated to PI, exposure to parents’ behavioral repertoires increases the likelihood that the child will draw on similar coping strategies.
An overall objective of these analyses has been to identify key dimensions of BFC and to suggest links to children’s behavior and well-being. Some results of treatments of PI effects have indicated either that most parents do not evidence serious problem behaviors prior to their incarceration or that such actions generally remain outside the child’s awareness. Examination of the in-depth interviews resulted in some contrary evidence, particularly about the perceived salience and sequencing of these experiences, and the observability of parental behaviors. Yet more detailed assessments of levels and seriousness of parental actions are needed include both parents’ behavior histories, and ideally those of other members of the family.9 The BFC summary index we relied on in the quantitative analyses was limited as many of the survey items referenced the behavior of the parent who completed the survey, most often the child’s mother. This should have resulted, however, in a conservative estimate of BFC effects. Other traditions within criminology should be leveraged (e.g., trajectory analyses, long-term patterns of persistence/desistance, and network methods—here applied to family configurations rather than to delinquent peers), that, along with timing of incarceration, would provide a more complete picture of the impact of BFC in relation to PI experience(s).
In social learning theory, in addition to direct observation, social influence occurs via ongoing patterns of interaction and communication. Studies are needed that are aimed at moving beyond global measures of “definitions favorable to the violation of law” to identify specific parental attitudes that may be linked to patterns of intergenerational continuity. Developing more nuanced measures is particularly important where the referent is parent–child communication as clearly most parents do not want their children to become delinquent or to engage in aggressive actions against significant others. We also noted that much of the time BFC revolves around noncriminal concerns, yet attitudes and behaviors relating to other domains (we used the examples of school and romantic life) remain largely unexplored terrain for criminologists—even though these arenas are known to be crime “inhibitors.” The larger point is that family climates vary considerably and are not captured fully either with poverty indicators or traditional parenting variables such as attachment and supervision.
Our use of the life-course lens revealed a more basic research challenge, whether the focus is on BFC, PI effects, or linkages between them—namely, the number of changes in living circumstances PI-experienced children often accumulate during their formative years. The in-depth interview data contained examples in which instabilities seemed tied to BFC conditions rather than to PI, but PI itself has clearly been linked to family instability either directly (e.g., as may result from maternal incarceration) or indirectly (e.g., loss of income or divorce; e.g., Lopoo & Western, 2005; Siegel, 2011). Zooming in on a specific incarceration period is a useful strategy for gauging effects of PI on child well-being, but it may provide a limited view of children’s long-term prospects if we do not take into account the full array of living circumstances that complete the picture of “home life” for particular children.
The longer life-course lens we advanced here also allowed us to make linkages between family of origin experiences (both BFC and PI) and the character of later ties with friends and romantic partners. The results of our focus on the family’s channeling role indicate the long reach of these early experiences. Yet these results also indicate that the characteristics of young adult relationships explains the additional variance in the two outcomes we examined (crime, IPV). Therefore, we have provided evidence that extrafamilial relationships may serve as (social) anchors for moving in a more positive life direction. Additional research on factors associated with such discontinuities (in effect “bucking the trend”) should be a high priority.
A limitation of the study is that as we focused on a community sample, many instances of parental incarceration consist of short jail stays rather than of lengthy incarceration periods. Thus, it may be unrealistic to expect that these young adult respondents would focus heavily on a parent’s earlier, and at times short, stay in a county facility. This undoubtedly influenced our assessment that, for example, when juxtaposed against an ongoing problem such as a parent’s long-term struggle with substance use, the experience of PI may not always represent a key family turning point. Yet in other studies in the PI tradition, scholars have also drawn on community-based samples and have emphasized underlying mechanisms that may fit more readily the situations children face whose parents have experienced lengthier periods of incarceration. As other researchers have concluded, studies are needed that distinguish type and length of incarceration (e.g., Kirk & Wakefield, 2018; Travis et al., 2014). Our expectation, however, is that children whose parents were incarcerated for longer periods will not only experience more severe feelings of loss/stigma along the lines of previous theorizing, but they will often have coped with family realities such as those described in this investigation (i.e., these parents often have more extensive offense histories and drug-use careers). And consistent with the child’s vantage point and several histories we described, negative effects on child well-being may connect to an extended pattern of lower level system contacts and the behavior related to those contacts, rather than just to serious crimes or longer incarceration periods.
Conceptualizing incarceration as part of a larger risk set that encompasses family members’ behavior profiles has policy implications. Although we emphasized some limitations of conceptualizing the parent’s incarceration as the primary source of all of these difficulties and disadvantages, parental incarceration experience is nevertheless an efficient means to identify an extremely, if not uniquely, high-risk group of children. In addition, although it is generally accepted that incarceration is an ineffectual response to complex societal and individual-level problems, the “package” idea and these data indicate that, at least in the short term, reducing the use of incarceration in itself is not likely to be sufficient to address the equally complex needs of children growing up in these families. It is thus imperative to direct significant resources to parents in more fundamental areas (e.g., increased access to high-quality drug treatment and programs focused on healthy relationship dynamics), whether in connection with alternatives to prison or while a parent is under confinement.
Furthermore, programming to address the needs of children experiencing PI has been focused mainly on mitigating the negative effects of the PI experience itself (e.g., providing opportunities to maintain contact). It is also critical, however, to provide assistance so that children develop positive strategies for coping with these underlying family issues such as parents’ drug use and family violence, as well as with difficulties linked specifically to the PI experience. For example, several interviews included rather harsh assessments about the parents’ actions and life choices (“I had a shitty-ass mom”), which reveal the need for additional counseling and education about addiction and about other life problems the parent may have faced. And, based on the portrait of the broader family climate that emerged from these interviews, when one parent does go to prison, additional oversight is needed of the child’s subsequent living arrangements. These arrangements most often involve the other parent, grandparent, or other relatives, and there is often little-to-no follow-up to ensure the safety and stability of these family circumstances (Johnston, 2006). Finally, our results indicated that respondents who had gone on to develop relationships with more prosocial peers and partners were less likely to report serious young adult problems. From these findings, then, we can underscore the need to provide young people with more options that allow them to develop prosocial connections—during the incarceration period, and throughout their formative years.
Supplementary Material
Acknowledgments
Funding information
National Institute of Justice, Grant/Award Numbers: 2009-IJ-CX-0503, 2010-MU-MU-0031; Eunice Kennedy Shriver National Institute of Child Health and Human Development, Grant/Award Numbers: HD036223, HD044206, HD66087, R24HD050959
AUTHOR BIOGRAPHIES
Peggy C. Giordano is a distinguished research professor of sociology at Bowling Green State University. Her research is centered on basic social network processes, and the ways in which close relationships influence delinquency and criminal behavior over the life course. She is especially interested in family dynamics associated with the intergenerational transmission of crime and other problem outcomes, as well as in the impact of extrafamilial influences such as peers and romantic partners.
Jennifer E. Copp is an assistant professor in the College of Criminology and Criminal Justice at Florida State University. Her research is aimed at examining the influence of intimate and familial relationships on developmental outcomes across a variety of domains, including crime and other problem behaviors. Recently, she has considered how incarceration, and other forms of system contact, influence future behavior, family life, and child well-being.
Wendy D. Manning is a Dr. Howard E. Aldrich and Penny Daum Aldrich Distinguished Research Professor of Sociology at Bowling Green State University, director of the Center for Family and Demographic Research, and co-director of the National Center for Family and Marriage Research. Her research is focused on relationships that exist outside the boundaries of marriage, including cohabitation, and the effects of varied family forms on relationship dynamics, stability, and parenting practices.
Monica A. Longmore is a professor of sociology at Bowling Green State University. Her interests include parenting and family dynamics, social psychological processes, and the ways in which these family circumstances and subjective understandings affect adolescent and young adult behavior and well-being. Her recent research has been focused on precursors and consequences of intimate partner violence.
Footnotes
Additional supporting information can be found in the listing for this article in the Wiley Online Library at http://onlinelibrary.wiley.com/doi/10.1111/crim.2019.57.issue-3/issuetoc.
SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Wrongful convictions are an important exception.
For example, early onset has been associated with the seriousness of delinquency involvement and the length of criminal careers (Piquero, Farrington, & Blumstein, 2007).
To limit the scope of our discussion, we focus on an outcome of interest to criminologists (crime) and on a second (IPV) that indexes the quality of young adult relationships, which is also an actionable offense.
Additional supporting information can be found in the listing for this article in theWiley Online Library at http://onlinelibrary.wiley.com/doi/10.1111/crim.2019.57.issue-3/issuetoc.
The narratives of those in the two subgroups (PI, no-PI) contain negative cases, including respondents in the no-PI group who reported criminal involvement or IPV, as well as those in the PI group whose lives were relatively on-track. Yet whether relying on the qualitative or quantitative data, results indicate that the PI subgroup on average evidenced more problem behaviors (and family backgrounds) relative to those in the no-PI sample.
It is possible that some respondents did not mention parental incarceration due to the perceived stigma associated with this event, but given their candor about other difficult aspects of their early home lives and current circumstances, in our view, this does not offer a complete explanation for the relative absence of these mentions in the lengthy narratives.
A total of 49 percent of PI-experienced respondents, compared with 19 percent of non-PI respondents (t = –8.35, p < .0001).
This finding is similar to results of a recent AddHealth analysis, which showed that adolescents who experience parental incarceration associate with peer groups that are, on average, more delinquent (Bryan, 2017).We add to the portrait, in documenting that BFC explains additional variance, and that both PI and BFC are related to the character of the individual’s romantic ties.
Although we have argued that parents’ behaviors, networks, and lifestyles are distinguishable from socioeconomic disadvantage, the latter constitute an ever-present, consequential backdrop. Thus, for example, affluent drug-involved parents rarely face the prospect of system contact.
REFERENCES
- Akers RL (2011). Social learning and social structure: A general theory of crime and deviance. New Brunswick, NJ: Transaction. [Google Scholar]
- Arditti JA (2012). Parental incarceration and the family: Psychological and social effects of imprisonment on children, parents, and caregivers. New York: New York University Press. [Google Scholar]
- Arditti JA, & Salva J. (2015). Parental incarceration and child trauma symptoms in single caregiver homes. Journal of Child and Family Studies, 24, 551–561. [Google Scholar]
- Arnold D, Dobbie W, & Yang CS (2018). Racial bias in bail decisions. The Quarterly Journal of Economics, 133, 1885–1932. [Google Scholar]
- Avison WR, & Turner RJ (1988). Stressful life events and depressive symptoms: Disaggregating the effects of acute stressors and chronic strains. Journal of Health and Social Behavior, 29, 253–264. [PubMed] [Google Scholar]
- Braman D. (2007). Doing time on the outside: Incarceration and family life in urban America. Ann Arbor: The University of Michigan Press. [Google Scholar]
- Bryan B. (2017). Paternal incarceration and adolescent social network disadvantage. Demography, 54, 1477–1501. [DOI] [PubMed] [Google Scholar]
- Comfort M. (2007). Punishment beyond the legal offender. Annual Review of Law and Social Science, 3, 271–296. [Google Scholar]
- Copes H, Hochstetler A, & Williams JP (2008). “We weren’t like no regular dope fiends”: Negotiating hustler and crackhead identities. Social Problems, 55, 254–270. [Google Scholar]
- Copp JE, Giordano PC, Manning WD, & Longmore MA (2018). Parental incarceration and child well-being: Conceptual and practical concerns regarding the use of propensity scores. Socius: Sociological Research for a Dynamic World, 4, 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cullen FT, Jonson CL, & Nagin DS (2011). Prisons do not reduce recidivism: The high cost of ignoring science. The Prison Journal, 91, 48S–65S. [Google Scholar]
- Elder GH Jr. (1998). The life course as developmental theory. Child Development, 69, 1–12. [PubMed] [Google Scholar]
- Farrington DP, Coid JW, & Murray J. (2009). Family factors in the intergenerational transmission of offending. Criminal Behaviour and Mental Health, 19, 109–124. [DOI] [PubMed] [Google Scholar]
- Farrington DP, Jolliffe D, Loeber R, Stouthamer-Loeber M, & Kalb LM (2001). The concentration of offenders in families, and family criminality in the prediction of boys’ delinquency. Journal of Adolescence, 24, 579–596. [DOI] [PubMed] [Google Scholar]
- Foster H, & Hagan J. (2015). Punishment regimes and the multilevel effects of parental incarceration: Intergenerational, intersectional, and interinstitutional models of social inequality and systematic exclusion. Annual Review of Sociology, 41, 135–158. [Google Scholar]
- Geller A, Cooper CE, Garfinkel I, Schwartz-Soicher O, & Mincy RB (2012). Beyond absenteeism: Father incarceration and child development. Demography, 49, 49–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Giordano PC (2010). Legacies of crime: A follow-up of the children of highly delinquent girls and boys. New York: Cambridge University Press. [Google Scholar]
- Giordano PC, Copp JE, Longmore MA, & Manning WD (2015). Contested domains, verbal “amplifiers,” and intimate partner violence in young adulthood. Social Forces, 94, 923–951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haney L. (2018). Incarcerated fatherhood: The entanglements of child support debt and mass imprisonment. American Journal of Sociology, 124, 1–48. [Google Scholar]
- Hanlon TE, Carswell SB, & Rose M. (2007). Research on the care taking of children of incarcerated parents: Findings and their service delivery implications. Children and Youth Services Review, 29, 348–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harding DJ, Morenoff JD, Nguyen AP, & Bushway SD (2018). Imprisonment and labor market outcomes: Evidence from a natural experiment. American Journal of Sociology, 124, 49–110. [Google Scholar]
- Haynie DL (2002). Friendship networks and delinquency: The relative nature of peer delinquency. Journal of Quantitative Criminology, 18, 99–134. [Google Scholar]
- Haynie DL, & Osgood DW (2005). Reconsidering peers and delinquency: How do peers matter? Social Forces, 84, 1109–1130. [Google Scholar]
- Herrera VM, Wiersma JD, & Cleveland HH (2008). The influence of individual and partner characteristics on the perpetration of intimate partner violence in young adult relationships. Journal of Youth and Adolescence, 37, 284–296. [Google Scholar]
- Hirschi T. (1969). Causes of delinquency. Berkeley: University of California Press. [Google Scholar]
- Johnson EI, & Waldfogel J. (2002). Parental incarceration: Recent trends and implications for child welfare. Social Service Review, 76, 460–479. [Google Scholar]
- Johnston D. (2006). The wrong road: Efforts to understand the effects of parental crime and incarceration. Criminology & Public Policy, 5, 703–720. [Google Scholar]
- Kinner SA, Alati R, Najman JM, & Williams GM (2007). Do paternal arrest and imprisonment lead to child behaviour problems and substance use? A longitudinal analysis. Journal of Child Psychology and Psychiatry, 48, 1148–1156. [DOI] [PubMed] [Google Scholar]
- Kirk DS, & Wakefield S. (2018). Collateral consequences of punishment: A critical review and path forward. Annual Review of Criminology, 1, 171–194. [Google Scholar]
- Krueger RF, Moffitt TE, Caspi A, Bleske A, & Silva PA (1998). Assortative mating for antisocial behavior: Developmental and methodological implications. Behavior Genetics, 28, 173–186. [DOI] [PubMed] [Google Scholar]
- Kutateladze BL, Andiloro NR, Johnson BD, & Spohn CC (2014). Cumulative disadvantage: Examining racial and ethnic disparity in prosecution and sentencing. Criminology, 52, 514–551. [Google Scholar]
- Lopoo LM, & Western B. (2005). Incarceration and the formation and stability of marital unions. Journal of Marriage and Family, 67, 721–734. [Google Scholar]
- Matsueda RL, & Anderson K. (1998). The dynamics of delinquent peers and delinquent behavior. Criminology, 36, 269–308. [Google Scholar]
- Mears DP, & Siennick SE (2016). Young adult outcomes and the life-course penalties of parental incarceration. Journal of Research in Crime and Delinquency, 53, 3–35. [Google Scholar]
- Miller J. (2008). Getting played: African American girls, urban inequality, and gendered violence. New York: New York University Press. [Google Scholar]
- Mowen TJ, & Visher CA (2015). Drug use and crime after incarceration: The role of family support and family conflict. Justice Quarterly, 32, 337–359. [Google Scholar]
- Mumola CJ, & Karberg JC (2006). Drug use and dependence, state and federal prisoners, 2004 (Special Report NCJ213530). Washington, DC: Bureau of Justice Statistics. [Google Scholar]
- Murray J. (2007). The cycle of punishment: Social exclusion of prisoners and their children. Criminology and Criminal Justice, 7, 55–81. [Google Scholar]
- Murray J, Farrington DP, & Sekol I. (2012). Children’s antisocial behavior, mental health, drug use, and educational performance after parental incarceration: A systematic review and meta-analysis. Psychological Bulletin, 138, 175–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Murray J, Loeber R, & Pardini D. (2012). Parental involvement in the criminal justice system and the development of youth theft, marijuana use, depression, and poor academic performance. Criminology, 50, 255–302. [Google Scholar]
- Pager D. (2003). The mark of a criminal record. American Journal of Sociology, 108, 937–975. [Google Scholar]
- Phillips SD, Burns BJ, Wagner HR, Kramer TL, & Robbins JM (2002). Parental incarceration among adolescents receiving mental health services. Journal of Child and Family Studies, 11, 385–399. [Google Scholar]
- Phillips SD, Erklani A, Keeler GP, Costello AJ, & Angold A. (2006). Disentangling the risks: Parent criminal justice involvement and children’s exposure to family risks. Criminology & Public Policy, 5, 677–702. [Google Scholar]
- Piquero AR, Farrington DP, & Blumstein A. (2007). Key issues in criminal career research: New analyses of the Cambridge Study in Delinquent Development. New York: Cambridge University Press. [Google Scholar]
- Poehlmann-Tynan J. (Ed.). (2015). Children’s contact with incarcerated parents: Implications for policy and intervention. New York: Springer International. [Google Scholar]
- Poehlmann-Tynan J, Burnson C, Runion H, & Weymouth LA (2017). Attachment in young children with incarcerated fathers. Development and Psychopathology, 29, 389–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Porter LC, & King RD (2015). Absent fathers or absent variables? A new look at paternal incarceration and delinquency. Journal of Research in Crime and Delinquency, 55, 414–443. [Google Scholar]
- Roettger M, & Swisher RR (2011). Racial and ethnic differences in associations between father’s incarceration and son’s delinquency and arrest. Criminology, 49, 1109–1147. [Google Scholar]
- Schwartz-Soicher O, Geller A, & Garfinkel I. (2011). The effect of parental incarceration on material hardship. Social Service Review, 85, 447–473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shaw CR (1930). The jack-roller. Chicago: University of Chicago Press. [Google Scholar]
- Siegel JA (2011). Disrupted childhoods: Children of women in prison. Piscataway, NJ: Rutgers University Press. [Google Scholar]
- Slade EP, Stuart EA, Salkever DS, Karakus M, Green KM, & Ialongo N. (2008). Impacts of age of onset of substance use disorders on risk of adult incarceration among disadvantaged urban youth: A propensity score matching approach. Drug and Alcohol Dependence, 95, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Slomkowski C, Rende R, Conger KJ, Simons RL, & Conger RD (2001). Sisters, brothers, and delinquency: Evaluating social influence during early and middle adolescence. Child Development, 72, 271–283. [DOI] [PubMed] [Google Scholar]
- Smith VC, & Wilson CR (2016). Families affected by parental substance use. Pediatrics, 138, e20167575. [DOI] [PubMed] [Google Scholar]
- Steffensmeier DJ, Ulmer J, & Kramer J. (1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, Black, and male. Criminology, 36, 763–797. [Google Scholar]
- Sutherland EH (1947). Principles of criminology (4th ed.). Chicago: Lippincott. [Google Scholar]
- Swisher RR, & Roettger ME (2012). Father’s incarceration and youth delinquency and depression: Examining differences by race and ethnicity. Journal of Research on Adolescence, 22, 597–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Travis J, Western B, & Redburn FS (Eds.). (2014). The growth of incarceration in the United States: Exploring causes and consequences. Washington, DC: National Academies Press. [Google Scholar]
- Turney K. (2014). The consequences of paternal incarceration for maternal neglect and harsh parenting. Social Forces, 92, 1607–1636. [Google Scholar]
- Uggen C. (2013, August). From crime to punishment (and back again). Paper presented at the annual meeting of the American Sociological Association, New York, NY. [Google Scholar]
- Wakefield S, & Apel RJ (2018). Criminological perspectives on parental incarceration. In Wildeman C, Haskins AR, & Poehlmann-Tynan J. (Eds.), When parents are incarcerated: Interdisciplinary research and interventions to support children. Washington, DC: American Psychological Association. [Google Scholar]
- Wakefield S, & Powell K. (2016). Distinguishing petty offenders from serious criminals in the estimation of family life effects. The ANNALS of the American Academy of Political and Social Science, 665, 195–212. [Google Scholar]
- Wakefield S, & Wildeman C. (2014). Children of the prison boom: Mass incarceration and the future of American inequality. New York: Oxford University Press. [Google Scholar]
- Weerman FM (2011). Delinquent peers in context: A longitudinal network analysis of selection and influence effects. Criminology, 49, 253–286. [Google Scholar]
- West DJ, & Farrington DP (1977). The delinquency way of life. London, England: Heinemann. [Google Scholar]
- Western B, & Pettit B. (2010). Incarceration and social inequality. Daedalus, 139, 8–19. [DOI] [PubMed] [Google Scholar]
- Wildeman C. (2009). Parental imprisonment, the prison boom, and the concentration of childhood disadvantage. Demography, 46, 265–280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wildeman C. (2010). Paternal incarceration and children’s physically aggressive behaviors: Evidence from the Fragile Families and Child Wellbeing Study. Social Forces, 89, 285–309. [Google Scholar]
- Wildeman C, & Andersen SH (2017). Paternal incarceration and children’s risk of being charged by early adulthood: Evidence from a Danish policy shock. Criminology, 55, 32–58. [Google Scholar]
- Wildeman C, & Turney K. (2014). Positive, negative, or null? The effects of maternal incarceration on children’s behavioral problems. Demography, 51, 1041–1068. [DOI] [PubMed] [Google Scholar]
- Wildeman C, & Wakefield S. (2014). The long arm of the law: The concentration of incarceration in families in the era of mass incarceration. Journal of Gender, Race and Justice, 17, 367–389. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.