Abstract
This article examines heterogeneity in adverse events and conditions and how low-income African American young adults respond. Although nearly all individuals in the sample report at least one instance of adversity, the nature and frequency of adversity varies, as do the responses. Some individuals see their lives and plans derailed; others engage in more protective strategies. For still others, adversity presents a difficult trade-off between surviving and thriving. We formalize this trade-off as an extension of a basic model of costly human capital investments. The model shows that a rational, fully informed individual facing this brutal trade-off, in an effort to survive the fallout of adversity, may optimally choose not to make high-return investments that promote thriving in the future. Improved policy design would recognize this type of trade-off.
Volumes of social science research show that different forms of adversity predict diminished social, educational, and economic outcomes among children and young adults. Adversity, moreover, includes a wide range of circumstances, patterns, and events, from long-standing disadvantageous conditions, such as living in a violent neighborhood or growing up with an addicted family member, to shorter disruptive episodes or events that could last merely seconds, such as witnessing or being a victim of a shooting. The impacts of exposure to adversity can occur immediately and also reverberate over the life course (see Felitti et al. 1998; Schafer, Ferraro, and Mustillo 2011). According to the U.S. Department of Justice (2020), 60 percent of American children have been exposed to violence, crime, or abuse, which can occur at home, at school, or in their neighborhood, with higher rates among Black children. For example, Dean Kilpatrick, Benjamin Saunders, and Daniel Smith (2003) find that 57.2 percent of Black children, versus 34.3 percent of White children, have witnessed acts of violence in their lifetime.
The consequences of exposure to adversity and disruptive events are dire, including low performance at school, poor physical and mental health, violent behavior, and criminality, among many others. Moreover, these impacts can vary; a long-standing literature on resilience (Troy et al. 2022) finds that not all adverse events lead to significant disruption in functioning, and a growing body of research examining heterogeneous impacts of adversity suggests that individuals who face similar adverse events may exhibit different responses in ways that affect the long-run consequences of adversity (Aquino, Brand, and Torche 2022). For instance, evidence suggests that females face worse mental health consequences after exposure to violence than males do (Fitzpatrick and Boldizar 1993).
Despite an increasing focus on heterogeneity, few studies examine variation in decision-making in response to adversity and disruption. Yet the strategies people develop and the choices they make in the face of adversity, including efforts to manage and mitigate the short-run consequences of adversity, can likely help explain heterogeneous consequences. This includes variation in whether and how the consequences of adversity extend to long-run outcomes over the lifecycle and thus contribute to or perpetuate inequality.
In this article, we examine heterogeneity in adverse experiences as well as subsequent behavior and decision-making in response to these disruptions. We use data from a sample of 150 low-income African American youth born in high-poverty neighborhoods in Baltimore. Data were collected via in-depth semi-structured interviews, and cover significant ground, from life histories to future hopes, and thus offer rich narratives on adversity and responses to it. Although our focus on low-income Black youth suggests homogeneity along important dimensions of disadvantage—indeed, we find that nearly all respondents experienced some form of serious adversity—our sample exhibits important variation in how adverse events manifest and how young people respond to them. Research documents that young adults growing up in neighborhoods of concentrated disadvantage face serious adversity and economic barriers to educational and occupational attainment, but less commonly examines the range of reactive behaviors and decision-making in the wake of these difficult events (for exceptions, see Small 2004; Edin and Kefalas 2005; Hannerz 1969). Moreover, as we explain, reported adversity shows temporal heterogeneity, from short one-time instances (disruptive events) to long-standing circumstances (adverse conditions). Thus, when we use the term adversity, we refer to both, although shorter disruptive events are often more remarked upon in youths’ accounts.
Our analysis begins with a categorization of young adults’ descriptions of adversity. We create a typology that allows us to observe not only the frequency but also the nature of disruption and adversity along several dimensions of heterogeneity, including the time frame. For example, some youth live in violent and poor neighborhoods or homes, where victimization, addiction, or family instability are long-standing and woven into the fabric of their lives (for other longer-duration adverse circumstances, see in this issue Turney et al. 2024; Bailey et al. 2024; Rauscher and Cao 2024). Others also experience more acute shocks, such as the death or incarceration of a family member or bouts of homelessness (on more acute events, see in this issue Baranowska-Rataj, Högberg, and Voßemer 2024; Alcaíno and Argote 2024; Khalid et al. 2024). In general, most respondents who report a destabilizing disruptive event do so against a backdrop of ongoing adverse conditions; for example, being a victim of a violent assault may be reported by someone who also describes living in a neighborhood where violent crime occurs regularly.
We also examine respondents’ perceptions of adversity. The interviews were not designed to ask about adverse events per se but to understand how social context shapes the transition to adulthood. Thus that we learn so much about adversity is telling in its own right. Although it is not always possible to draw conclusions given that interviewers did not explicitly ask about adverse events or perceptions of adversity, the perceptions we do observe provide prima facie evidence that what we as researchers characterize as adversity is indeed perceived as such by respondents. In their narratives, many young adults also describe what happened and what they did in the wake of these adverse events and shocks, which in turn supports the idea that we can describe some of their reactive and intentional responses and decision-making processes. Thus, this article not only describes adverse events in great detail, but also examines how youth explain their own rationales for how they responded to these events—prioritizing their agency over our analysis.
This leads to a second and central part of our study, which is to describe how young adults respond to adversity in general and disruptive events in particular. We find significant heterogeneity in these responses, including both reactive coping strategies and intentional protective decisions, which have implications for longer-term trajectories and outcomes, and may be potential mechanisms that help generate or perpetuate social and economic inequality. In particular, we identify three basic categories of responses. Responses in the first category amount to different forms of derailment, including running away, living on the streets, turning to crime to make money, or coping with trauma by using addictive substances. This response is best understood less as an active or deliberate way to manage adversity and more as a reaction to the perception—perhaps correct in many cases—that there is little to be done to improve circumstances given the depth of adversity faced. The second category of responses includes protective actions and behaviors that may counteract or help respondents avoid the worst aspects of adversity and, moreover, are likely to be beneficial in the long term. For example, a focus on school or extracurricular activities can provide a safe haven from violence and is also an investment in human capital with potential long-term payoffs.
A third and intriguing category—one that requires additional focus—embodies a set of behaviors and choices respondents make that may be helpful in the short term but could be harmful in the long term, such as dropping out of school to care for relatives and social avoidance or withdrawal strategies. Avoidance strategies, for example, may protect youth from violence in the near term, but might also backfire in the long term by limiting youths’ exploration of social networks and activities, which could then limit their eventual schooling and career options. In other words, some strategies that can help individuals survive (that is, avoid some of the worst outcomes, such as death, incarceration, persistent criminality, homelessness, or neglect of an ailing or vulnerable family member), may diminish their ability to thrive (that is, reach their potential or make sustained progress toward their stated goals, or find activities at school, home, or work that they find meaningful or engaging).1
In the final part of this analysis, we posit a simple model, motivated by our empirical findings, especially heterogeneous responses and the potential trade-off between surviving and thriving that some individuals seem to face. The model formally characterizes this type of trade-off, which can lead individuals to adopt strategies that make it difficult to thrive and thereby captures one way that adversity can generate and perpetuate inequality. The model also incorporates the idea that the trade-off between surviving and thriving may be irrelevant to individuals for whom survival is nearly assured, but for many people who face violence, instability, and scarcity, it may be a recurring factor affecting their choices.
In general, the model captures the idea that individuals living in poverty can face a brutal trade-off: surviving, even if doing so blunts their ability to thrive. An implication is that policies to break cycles of poverty could focus on interventions designed around recognition of this trade-off. As a simple example, afterschool programs to help students prepare for postsecondary education may not be useful for students who must take care of an ailing relative or who fear for their personal safety or literal survival when they travel to or from school (Burdick-Will 2013); similarly, postsecondary educational pathways and the financial aid policies to support them may require too stringent a timetable for completion than some youth can manage given the need to stop out and work to support a vulnerable household (see Hart 2019).
Our approach to investigating adversity is unique and multidisciplinary. Approaches to studying adversity vary both within and across disciplines and include the collection and analysis of qualitative data (such as semi-structured interviews and ethnographic observations), experimental or causal inference methods applied to “big data,” and nationally representative data sets, to estimate treatment effects of adverse and disruptive experiences. Still other methods focus on the development of theoretical models of dynamic decision-making that can be matched to data and employed to simulate responses to counterfactual policies that reduce adversity or mitigate some of its consequences. Each approach can help us understand how adversity weaves its way through individuals’ decisions and experiences over the life course and thus how adversity generates and perpetuates inequality. However, the conceptual frameworks and methods are nearly always used in isolation when, presumably, bringing more tools to bear could lead to stronger evidence of the nature and consequences of adversity and, thus, better policy to address it (for another example, see also Bergman et al. 2023).2 Given the set of methods we use, this study represents an attempt to help bridge disciplinary divides.
LITERATURE REVIEW
Hundreds if not thousands of studies in the empirical social sciences examine adversity and its consequences (see Torche, Fletcher, and Brand 2024, this issue). We do not try to cover this entire body of literature, but instead focus on studies that illustrate approaches to which we see this article as a direct contribution. We consider examples of research from economics and sociology (among other fields) on adversity, especially in high-poverty neighborhoods; the interaction between poverty and low human capital investments, which helps perpetuate poverty; and interactions between poverty and decision-making more generally.
Adversity in High-Poverty Contexts
A well-developed literature spanning thirty years looks at how urban poverty—especially neighborhood disadvantage—presents adversities that shape child and youth outcomes (Mayer and Jencks 1989; Wilson 1987; Chetty, Hendren, and Katz 2016). In particular, scholars find that young people growing up in high poverty, violent communities are more likely to drop out of high school, engage in delinquency, and become teenage parents, and less likely to go to college (Leventhal and Brooks-Gunn 2000; Sampson, Morenoff, and Gannon-Rowley 2002; Sampson, Sharkey, and Raudenbush 2008; Sharkey 2010; Wodtke et al. 2011; Harding 2003; Papachristos, Hureau, and Braga 2013). More recently, many economists have also turned their attention to neighborhood-level factors that capture different forms of adversity for low-income children and families and also explain disadvantaged outcomes, such as health disparities (Currie 2011), violence (Grogger 1997), incarceration of parents and siblings (Norris, Pecenco, and Weaver 2021; Bhuller et al. 2018), eviction or foreclosures (Diamond, Guren, and Tan 2020; Collinson et al. 2022), exposure to pollution (Currie et al. 2009), discrimination (Lang and Manove 2011), and juvenile incarceration (Aizer and Doyle 2015).
Yet the specific mechanisms through which these adverse events experienced by young adults like those in our sample (that is, those facing the same types of family, school, or neighborhood disadvantage) affect their development—though well theorized—remain relatively understudied (Sharkey and Faber 2014; Galster and Sharkey 2017). Variation in neighborhood and structural factors such as schools (Schwartz 2010; Johnson 2019), exposure to violence (Sharkey 2010; Sharkey and Torrats-Espinosa 2017), and policing regimes (Neil and Sampson 2021) predict unequal educational and economic outcomes, and additional work has suggested that violent, high-poverty neighborhoods diminish parents’ mental health and their efficacy (Ludwig et al. 2013; Diez Roux and Mair 2010). Yet research focusing on how such factors—violence, policing, parental efficacy under duress, and inadequate school conditions—directly shape the decisions of young people that are most consequential for their future attainment is still scant. We know youth face these conditions in their daily lives in low-income neighborhoods or households, but exactly how do these adversities and constraints translate into diminished outcomes, and how do we explain heterogeneity in these outcomes for youth from similar backgrounds?
Beyond providing a detailed taxonomy of adversity and disruptive events from a sample of low-income Black youth, our contribution is to leverage interdisciplinary frameworks and qualitative data to formalize a specific mechanism through which adversity can have long-term consequences consequences. We argue that adversity can generate a trade-off between mitigating the immediate consequences of adversity and improving longer-term outcomes. We formalize this idea as a simple model of human capital in which long-term investments imply large—and largely unappreciated or invisible—costs that make them suboptimal.
Poverty and Human Capital Accumulation
The literature in economics on human capital accumulation, defined as the set of skills, traits, experiences, and other factors that predict lifecycle outcomes, is massive. Traditionally, models of human capital incorporate education and work experience (Mincer 1958; Becker 1962) or different kinds of productive abilities, such as manual versus academic skill (Willis and Rosen 1979). Economists have since incorporated new dimensions into our understanding of what constitutes human capital, such as health (Grossman 1972) and socioemotional skills (Heckman and Rubinstein 2001), among many others. A large literature continues to extend our understanding of human capital, focusing on how individuals make costly investments that increase their human capital, a canonical example being investments in education. An important question that continues to generate scholarship is why disadvantaged groups often invest less in their human capital; for example, why do people from poor families complete less education even after conditioning on measures of academic performance, such as test scores? Earlier work has offered many plausible explanations, including credit constraints (Lochner and Monge-Naranjo 2012; Hai and Heckman 2017), impatience (Levitt et al. 2016), and incorrect information about the returns to human capital investments (Cunha, Elo, and Culhane 2022), among others.
Sociological research, especially scholarship in urban and cultural sociology, comes closer to our article in recognizing that contending with poverty can itself create additional hardships, which thus reinforce disadvantage and inequality in human capital outcomes; this literature also argues that heterogeneity is significant in these processes (for reviews, see Newman and Massengill 2006; Small, Harding, and Lamont 2010). Low-income African American youth, whose neighborhoods have long been more socially isolated and economically disadvantaged than the neighborhoods of White youth (Sampson 2012; Sharkey 2008), must make difficult trade-offs as they cope with adverse events and navigate their social networks, families, and communities. For example, experiences of school or neighborhood adversity or instability lead some youth to isolate or strategically cut their social ties (Chan Tack and Small 2017; Clampet-Lundquist et al. 2011; Koogler 2019; Rosenblatt, Edin, and Zhu 2015; Trinidad 2021; Warner, Warner, and Kuhl 2017) and take on or exit certain challenging family roles (Burton 2007; Haynie et al. 2009; Turney, Liu, and Marin 2024, this issue). Nikki Jones (2010) finds that some young African American women in Philadelphia survived violent and risky neighborhoods by fighting, and others by practicing “situational avoidance.”
Scholars also find that youth in low-income neighborhoods pursue low-wage jobs and sub-baccalaureate education after high school—instead of four-year degrees—because they perceive these to be viable options to avoid the street, meaningful ways to provide for one’s family in the near term, and tethers to long-term aspirations that in the short term, seem impossible to realize (Cox 2017; Newman 1999; Deterding 2015; Holland and DeLuca 2016). Such research suggests that rather than a lack of interest in optimally investing in their futures, youth can be strategic and adaptive in the face of limited resources and environments that make their goals much harder to achieve (see Harding 2010).Indeed, significant work in urban sociology argues that we must account for the importance of poverty, neighborhoods, and adverse experiences when trying to understand the outcomes of low-income young adults, as these contexts shape how they make investments in their education and perceive trade-offs between different options (Anderson 1999; Elliott et al. 1996; Wilson 1987; Young 1999). Such insights also help explain why their efforts do not necessarily accumulate into conventional indicators of social mobility and success (Anderson 1999; Elliott et al. 1996; Wilson 1987; Young 1999). This perspective is consistent with and guides our work in this article.
Less explicitly developed in earlier research is the idea that there may be circumstances or contexts in which surviving and thriving are in direct conflict with one another in the sense that investments that make the former more probable make the latter less probable. Some precedent for this idea is found in models of long-run investments focused on individuals living at or near subsistence in developing economies (see, for example, Banerjee and Duflo 2007). Individuals living near subsistence may not save even despite high returns if doing so means they cannot meet their most basic needs, such as consuming sufficient calories. Similarly, if an investment in long-term human capital lowers the likelihood of avoiding disastrous outcomes, rational individuals may not make them.
Poverty and Decision-Making
More generally, we relate to earlier scholarship on decision-making among disadvantaged groups that goes beyond decisions about human capital investments. However, some distinctions are significant. Research in economics has often argued that people living through scarcity, disruption, or adversity may exhibit suboptimal decision-making, for example, due to lack of information or a failure to properly incorporate long-term ramifications of their behavior (Mullainathan and Shafir 2013; Bertrand, Mullainathan, and Shafir 2004). Often categorized under the umbrella term behavioral economics, multiple modeling approaches predict suboptimal investments in long-term outcomes, such as savings and human capital (education, for example). These include incorrect information, such as biased beliefs about the returns to long-term investments (see Cunha, Elo, and Culhane 2022) or a preference for short-run payoffs that thwart plans that have payoffs far into the future (see Levitt et al. 2016).3
Other lines of research, common in sociology, have focused less formally on decision-making and rationality per se (see discussion in Bruch and Feinberg 2017; DeLuca and Jang-Trettien 2020) but have long acknowledged some of the inherent trade-offs that low-income individuals may face when deciding how to achieve mainstream goals such as marriage and parenthood without well-resourced avenues to do so (Edin and Lein 1997; Bell et al. 2018). Similarly, some sociologists have explored the adaptative behaviors some low-income men enact to achieve alternative goals (such as drug dealing and pursuing working-class jobs) when opportunity for higher aspirations is blocked (MacLeod 1987; Willis 1977; Anderson 1999); many of these decisions essentially raise the costs of long-term investments in education or other forms of human capital. As Mario Small, David Harding, and Michèle Lamont (2010, 9) point out, acting on “the ‘right’ set of values or beliefs may actually undermine one’s mobility when exercised in a difficult context.”
Our article, especially the model we develop, suggests that a rational, fully informed, and dynamically optimizing individual seeking to survive may engage in behaviors that undermine their ability to thrive. For example, social isolation may be the optimal choice to survive adversity or danger even if doing so reduces labor-market opportunities because it makes success in school more elusive. Although suboptimal decision-making may certainly play a role in the lives of the youth in our study, as in many contexts, the model we propose does not require these features to explain seemingly counterproductive behavior. Instead, the model is based on the idea that choices may be fully rational. In our case, long-term investments in human capital may entail sizable costs for disadvantaged individuals (costs that are underappreciated or invisible to some researchers and policymakers even if they are well known to others) such that it appears that decisions affecting investments in human capital are suboptimal even though they are not.
DATA AND METHODS
This article uses qualitative data gathered from in-depth semi-structured narrative interviews to examine heterogeneity in the experience of adverse events, and how youth respond to adverse events, with attention to how such responses and decision-making might relate to outcomes in the transition to adulthood. We draw on a sample of 150 low-income African American youth and young adults living in high-poverty, racially isolated neighborhoods in the city of Baltimore, Maryland (table 1). The data originated from a mixed-methods follow-up study of families in Baltimore who were part of the Moving to Opportunity (MTO) program, an experimental housing and neighborhood intervention implemented in the 1990s.4 A total of 636 families in Baltimore participated in this program, and a qualitative component was later added to examine the transition to adulthood among youth in the study (directed by Susan Clampet-Lundquist, Stefanie DeLuca, and Kathryn Edin).
Descriptive Characteristics of Youth Sample
A stratified random sample of two hundred youth from the MTO study (ages fifteen to twenty-four) were chosen (stratified by gender, age and treatment arm), and 75 percent agreed to participate in the qualitative portion of the study (N = 150). Most interviews were conducted in respondents’ homes, and 96 percent of respondents still lived in the Baltimore area at the time of the interview. All names have been changed, many to pseudonyms chosen by the respondents themselves. Forty-nine percent of the sample was female, and the mean age of respondents was 19.6 years old.5 The sample at baseline was significantly disadvantaged—youths and their families were public housing tenants, who were among the poorest residents in Baltimore at the time. Mean household income (in 2022 dollars) was $10,580.60, roughly 42.9 percent of the poverty line in 1994. Some 63.2 percent of the sample lived in households receiving Aid to Families with Dependent Children. Most of the youth (74.2 percent) lived in households where the head of household had never married, and where the head of household was not working (76.3 percent). Only 32 percent of the sample had parents who completed a high school degree or the equivalent, and only 13.3 percent had parents who ever attended any kind of postsecondary institution (almost none were four-year). Among youth respondents, 25.3 percent were currently in high school at the time of the study, 9.3 percent had no high school diploma or GED (General Educational Development) certificate, and 24 percent had a GED or high school diploma as their highest level of education. Although 40 percent of the sample (69 percent of the high school graduates) ever attended any postsecondary educational institution, only 10.7 percent ever attended a four-year institution; 16.7 percent had ever attended trade school; and 12.7 percent ever attended a two-year institution.6
The interviews covered a wide scope of topics and experiences around respondents’ transition into adulthood (for more on sample and interview design, see DeLuca, Clampet-Lundquist, and Edin 2016). The semi-structured, in-depth interviews included open-ended questions about employment, education, neighborhoods, friends and family, risky behavior, and mental health. Youth were asked directly about their college and career preparation, postsecondary decision-making process, and, for youth who were interviewed after high school, their experiences in the labor market and postsecondary institutions. However, discussions of adverse experiences emerged inductively in these interviews—they were not asked about directly. The sample allows us to understand heterogeneity among a group of youth often studied as if their trajectories were the same. Their shared socioeconomic origins explain how they ended up in the sample, but we observe significant differences in their pathways into adulthood and responses to adversity.
Methods
These data were not collected with the intention to analyze heterogeneity in adversity and disruption. Despite the difficulty in simplifying the complex and layered landscapes of adversity in respondents’ lives, we developed our coding scheme by first identifying several important aspects of adversity where heterogeneity could be observed, such as geographic and social proximity, and recency. Duration is a key dimension where we observe differences as respondents describe both long-standing adverse conditions along with individual events that could last a few seconds or something in between, such as relatively short periods of scarcity or crisis. Indeed, they often report several; for example, it is typical for disruptive events to occur during a longer period of adverse conditions.
Our typology thus extends beyond the well-known Adverse Childhood Experiences Study (ACES) (Felitti et al. 1998). Whereas the ACES includes self-reports of adverse experiences such as living in a household with domestic violence, substance abuse, or child abuse, our conception of adversity extends to include hardships such as neighborhood- and school-level violence, untimely death of family and friends, and absent parents. Further, rather than capturing only the presence of adversity, we also account for temporality, the proximity of the adversity to the respondent, and whether an adversity occurred multiple times. Even though our analysis tends to focus on responses to short disruptive events, our taxonomy includes longer periods of detrimental conditions. This means we can offer a fuller picture of what respondents report and also avoid imposing arbitrary temporal cutoffs in our decision of what to include as adversity. For example, the incarceration of a family member suggests a disruptive event (the arrest or conviction) but also leads to a potentially long period of disadvantageous conditions.
To generate a typology of adverse experiences, the research team documented each report of adversity and characterized each along the following dimensions: the social settings in which they took place (school, family, neighborhood, friend group, and other); recency (coded dichotomously for whether within the past two years); and duration (coded as continuous for uninterrupted periods, intermittent for multiple discrete events or periods, and one-time for discrete events). More detailed descriptions of our analysis process and these categories are presented in the online appendix (https://www.rsfjournal.org/content/10/1/103/tab-supplemental). Rather than simply trying to track each form of adversity as an isolated phenomenon, we also noted instances in which one source of adversity or disruption seemed to be rooted in a prior form of adversity to identify potential downstream effects. As we explain, some instances of diversity can be seen as negative shocks or events to which respondents actively respond. Others seem to reflect reverberations of some initial instance of diversity (such as eviction leading to homelessness, drug use, and school absences). We found a range of such experiences, including but not limited to assault, domestic violence, unplanned pregnancy, family members’ struggles with addiction, financial instability, housing instability, school instability, and inadequate resources.
Rather than solely drawing on our characterizations as researchers or basing our understanding only on existing studies of poverty, we also include descriptions of how youth themselves characterize adverse events. Our research team reviewed cases seeking to understand how respondents framed adversity, coding for negative, destabilizing, and neutral perceptions. We also found evidence of youth perceiving disruptive events as clarifying points, positive turning points, or as experiences that were not uncommon among their familial and social networks. The original qualitative interviews did not explicitly ask about perceptions. Consequently, although we may be able to identify certain perceptions when the respondent discusses them, we are careful not to consider the absence of an expressed perception as a lack thereof. We account for this by coding only explicit mentions of perceptions with an indicator variable. We code the lack of any specific mention as not applicable. The research team also provided justifying narrative data from the interviews for each perception coded. This was done to ground our coding decisions in each respondent’s language as expressed and to justify analysis and coding decisions.
Next, the analysis considers the responses to disruptive events and adversity that youth describe within the contexts of their daily lives and their life course trajectories. The research team coded for responses and strategies by thoroughly reading field notes and transcripts and noting instances where the respondent reports some consequence or strategy in response to adversity or as a preventive measure to guard against potential adversity. Examples include isolating from friends, co-workers, family, or neighborhood spaces as strategies for physical protection; becoming a caretaker for a family member or partner, due to illness or loss of income; or youth changing their college and career plans to assist family members. Detailed descriptions of these response types are available in the appendix. A systematic review of the qualitative data yielded a broader taxonomy of responses: reactive coping mechanisms and derailed pathways, proactive and protective strategies, and tough trade-offs. We discuss these later and in the appendix.
Although we separate these aspects of adversity for analytic clarity and illustration, the dimensions, perceptions, and responses to adversity that youth shared are often interconnected or conditionally related in the interview narratives. As such, evidence of each of the three dimensions will appear across each of the subsequent findings sections. This is especially evident when we present within- and between-case examples from the qualitative data to illustrate the range of responses to adverse events. The cases presented here were chosen to not only include diversity by gender, age, and response, but also because these cases are typical of the accounts that fit into these three emergent categories of responses. Different types of responses to adversity can also be present within-case. In other words, in our main analysis, we describe responses to instances of disruptive events rather than characterize individuals’ cumulative trajectories, the caveat being that we note when an instance of disruption appears to be a direct consequence of a previous instance.
FINDINGS
We present three main sets of findings: the range and prevalence of adversity in our sample, heterogeneous responses to adversity, and a mathematical formalization of the trade-off between surviving and thriving using an economic model of human capital investments.
Range and Prevalence of Adversity
We examined each case to identify instances of adverse events reported by respondents. Table 2 shows that 919 reports of adversity were reported by 150 individuals. Figure 1 provides a histogram for person-level counts to illustrate how the frequency of adversity is distributed across the sample. The average is 6.2 and the median is six. One individual reported no adversity and several individuals report fifteen or more distinct adverse events. More than half of the youth, 52 percent, reported between two and eight adversities.
Categories of Adversity in Youth Sample
Distribution of Adversities per Respondent
Source: Authors’ tabulation.
Note: A histogram of the number of adversities for each respondent including all 919 adversities described in table 2.
To better demonstrate which kinds of adversity different individuals experience, we report the prevalence of various adverse circumstances and events in table 2. The first column describes the type of event—for example, neighborhood violence. The second and third columns show the frequency of the adversity as individuals in the sample report it, as a count and a percentage, respectively; the fourth column shows the total instances reported across all sample members. For example, in the first row, ninety-one (of 150) respondents (60.7 percent of the sample) reported experiencing neighborhood violence. A total of 107 instances of neighborhood violence were reported in the sample (11.6 percent of all instances of adversity).
Table 2 shows the wide range of adversity in the lives of the youth in our study. For example, violence can occur at school, in the neighborhood, or at home. Other indicators of economic disadvantage include job loss, addiction problems of a family member, and housing instability. Some youth report encounters with police, some of which they characterize as harassment or misconduct. Other youth reported instances of family instability, such as estrangement or incarceration of a family member. Co-occurring with (or perhaps caused by) some of these adversities are physical and mental health problems, some diagnosed by clinical providers, others self-described. The vast temporal range includes disruptive events, long-standing disadvantageous conditions, and many others, such as bouts of homelessness or parents’ divorce or separation, a process that can last many months. School can provide a safe haven in the face of adversity, but many youth also report problems there, including violence and inadequate resources.7 Only one respondent in 150 does not report any adverse events during the interviews.
Table 3 provides additional characteristics of adversity reported by respondents. The table is organized similarly to table 2, that is, first at the individual and then at the adversity-instance level. We first coded instances of adversity according to their social locus (school, family, friend, neighborhood, or other). We also consider temporal heterogeneity, such as a single recent disruptive event, an intermittent source of adversity, or a longer-standing disadvantageous condition. What emerges is considerable range in the proximity, duration, and perceptions of adversity that youth face. For most categories, both social proximity and temporality, a majority of individuals report at least one corresponding adverse event or circumstance. For example, 140 individuals (93.3 percent of the sample) report an adverse event within the past two years; 125 (83 percent) report a neighborhood-level adversity.
Dimensions of Adversity in Youth Sample
We also examined how individuals framed adversity. Specifically, 105 individuals (70 percent) expressed a negative perception for one or more adversities, and fifty (33.3 percent) viewed one or more adversities as destabilizing. Notably, sixty-three respondents (42 percent) viewed at least one adversity as clarifying, and twenty (13 percent) perceived at least one adversity as positive (such as a helpful turning point). Approximately thirty (20 percent) viewed one or more adversities as ordinary, or not uncommon in their familial or social networks (see Aquino, Brand, and Torche 2022). A caveat to reports of perceptions is that in many cases where we have coded an adverse event or circumstance, no perception was reported. This occurred for 430 instances of adversity (roughly 46.8 percent of all reported adversities); for another twenty-seven (3 percent), the coder was unsure whether any text corresponded to a perception. However, respondents were never asked how they framed adversity, leaving the absence of a reported perception as difficult to interpret. In general, the counts and prevalences of perceptions must be interpreted with care. Most important, they provide some evidence that what we code as adversity is indeed perceived as such by youth, at least in some cases.
Heterogeneous Responses to Adversity
In the wake of these adverse events, youth described a range of consequences, decisions, reactions, coping mechanisms, and strategies.8 We observed at least three types of responses in the interviews, all varying in the extent to which they appeared to be reactions or derailments in the face of shocks versus deliberate or strategic decisions.9 The first set may be described as reactive coping mechanisms and derailed pathways, which include more impulsive and emotional responses that might have provided immediate relief or respite, but also ended up being quite costly in terms of school completion and employment. These included responses such as dropping out of school, running away to live on the streets, or substance use. A second set of responses are described as proactive and protective strategies that youth enacted to shield themselves from exposure to risks in their neighborhoods, schools, and families, and might also have eventually promoted long-term educational and professional goals, including cultivating deeper relationships with parents and mentors; investing in hobbies, personal interests, and extracurricular activities; and selective friendship and strategic avoidance of people and places they believed would jeopardize their plans. A third set of responses provided insight into some of the tough trade-offs youth were forced to make—these were decisions that seemed to provide short-term stability and survival, but could also backfire or have potentially negative long-term consequences, limiting educational and career pathways. These included leaving school to be a caregiver for sick relatives, leaving school to make money to support family, more severe social isolation, and avoiding school to stay safe.
Reactive Coping Mechanisms and Derailed Pathways (Neither Survive nor Thrive)
In the aftermath of adverse events described earlier, it is perhaps not surprising that a number of youth described feeling frightened, angry, or desperate, and made decisions they later came to regret. These consequences and reactions are those that might typically come to mind when thinking about how adverse events predict disadvantage. Some describe being thrown into a spiral, sharing accounts with less evidence of deliberation and choice in the wake of adverse events. When Daphne, twenty-one, reflected on how she dealt with the difficult events she experienced in her life, including her brother’s murder, her father’s incarceration, repeated episodes of school instability and fighting, an unexpected pregnancy with an abusive partner, and abuse from an uncle, she said with flat affect, “Can’t feel happy because it’s nothing to be happy about. Can’t feel sad because you can’t do nothing about it. It’s already done happened. You just got to go with it, just for real, deal with it.” Viewing their current situations as so difficult, some youth resorted to actions that if anything made things worse.
Terry, twenty-three, grew up in a family of nine. He told the interviewer that his drug-addicted mother was so emotionally abusive that he repeatedly ran away from his home in suburban Columbia, Maryland, sleeping on the street by the time he was eight years old. Eventually, he became a ward of the state and cycled through group homes. He rejoined the family when he was fifteen, but found once more the emotional abuse he had tried to escape. He fell into a depression after this reunion and began binge drinking. To escape his mother’s house again, he moved in with a brother, who then asked him to move out when the brother married; after this, he returned to his coping mechanism of sleeping in Leakin Park, hiding his belongings in the woods while he attended school. Reflecting on this period, he said, “I prayed and smoked a lot of weed,” and started hanging out with a group of other homeless youth. As he described their connection to each other, he explained: “People—out of a desperation—they cling to each other, complete strangers. . . . But they were never really friends. Kids, they drink together, they smoke together, because they don’t have anybody. . . It’s like this depression thing in the air, man, pain in people’s eyes and everything. . . . Nobody knows what to do, so they just party and [do] drugs.” Out of sheer determination, Terry was able to graduate from high school and eventually worked at one of the homeless shelters where he lived, but shared that he is still “learning to be comfortable inside a house,” something he still associates with his childhood abuse.
Anna, twenty-two, experienced repeated spells of housing instability growing up alongside repeated spells of paternal incarceration. When she was interviewed, Anna explained that she had had an unwanted pregnancy at age fourteen and that her parents helped push her to give the baby up for adoption. Looking back on what she called a “rough time” in her life, she deeply regretted giving her child away. She said that she felt unloved by her family and the turmoil of the unplanned pregnancy and subsequent adoption led her to run away from home and drop out of high school after ninth grade. She survived by engaging in sex work and living on the streets for two years before deciding to move back with her family. Anna explained that at the time, dropping out made sense to her: “I just didn’t complete it [high school] because my life wasn’t goin’ right at that time. . . . I felt as though quitting was the right way to go. But now I feel as though it wasn’t, like, I didn’t get my high school diploma, I don’t have my GED.”
When Jamison, twenty-one, was growing up, both of his parents struggled with substance use—his father with drugs and his mother with alcohol. Near tears throughout the interview, Jamison shared that he suffered from depression and turned first to weed, and then to alcohol, drinking every day, first thing in the morning. He had dropped out of high school to make money, and then lost his first job at a restaurant because he was drinking at work. His grandparents took him in to live with them in a retirement community in Florida to get straight and try to attend college. He left after three weeks and returned to Baltimore because his mother said she missed him, and because he started to feel isolated. He also felt that if he went back home, he could get a job through one of his cousins or his uncle. On returning to Baltimore, unfortunately, he did not find the fresh start he had hoped for:
But, I thought I had a master plan. . . . So, I was like, “I can get a job when I get back in Baltimore. I can save up money, work toward getting a car, and everything.” So, I came back here and nothing went right. Nobody called, I was getting frustrated. I’m thinking, “I have no criminal record, I got my GED, I’m planning on going to school.” They like, “But you don’t have any college.” . . . So, nothing went right. I look out for my mother and my family a lot so, they like, “You want anything?” and I’m like, “Yeah, can I get a drink?” and so, I came back drinking, drinking real heavy and stuff like that.
At the time of his interview, Jamison had found work at a restaurant, but struggled still with alcohol every day. He was hoping to return to Florida again to attend community college while living with his grandparents.
Shortly before Marco (twenty-two) was interviewed, his maternal uncle, who Marco said “was more like my dad for me,” had recently died of heart failure, which he said was the result of years of drug abuse. Adverse events were not new to Marco: growing up, his mother was in and out of the house and on the street doing drugs, leaving him to care for his siblings. At the time of his interview, he had not seen his father in ten years—his father was in jail for stealing cars. In high school, Marco had trouble sleeping and missed school regularly. At sixteen, he was incarcerated for armed robbery, which meant that he missed the summer school session he had needed to attend to make up for failing grades, and then he subsequently dropped out. By the end of the study, he had moved in with his grandmother in a nearby suburb, was unemployed, and had not yet gotten his GED. Reflecting on the past he said, “And like if I could, I probably wouldn’t of never missed all them days in school like I have, or been late like that. I don’t know, just a lot of things that I think is different from when I was a kid.”
Proactive and Protective Strategies (Survive and Thrive Aligned)
Although some of the youth who were interviewed turned to the streets, alcohol, and drugs, dropping out of school and sometimes work without much of a forward-looking trajectory, most did not. Many youth instead enacted deliberate strategies to prevent such unwanted outcomes. Some were identity projects—specific activities that provided youth with a sense of self and purpose in their daily lives (for details on identity projects, see DeLuca, Clampet-Lundquist, and Edin 2016). In particular, some of the youth said that they engaged more deeply in hobbies or sports, which provided a channel for positive connection with peers. Others invested more deeply in relationships with particular friends and family members or teachers, actively staying away from places and people who they felt might derail them or get in the way of their goals. By keeping busy, they explained, they could also avoid being in the wrong place at the wrong time, where violence might find them.
Hobbies and Positive Channels
Despite struggling with a severe bipolar disorder and having been hospitalized, Tony, twenty-one, found his footing after becoming deeply involved in the church and eventually finding a passion to pursue a career in pharmacy. Both of Tony’s brothers sold drugs and ended up in prison rather than following through on plans to go to college. Between his brothers’ imprisonment and the violence he saw in his neighborhood, he tried to avoid getting involved with too many people. In high school and after, he stayed to himself most of the time: “I did what I had to do, then went home. . . . I don’t want anyone to throw me off, so I’m like if you talkin’ negativity about doin’ something, I don’t even wanna hang around you, like you can do that by yourself.” Instead Tony invested in two close friends—Britney, with whom he took a biology class at a local community college and shared career aspirations in science, and Joshua, whom he knew from high school and who was in the Navy. He remained close with members of his family as well, such as his older sister, who was a school teacher. He also cultivated ties at his church, including Mr. Carter, who paid for him to go on a men’s Bible study retreat.
Tony explained how church helped motivate him and give him optimism: “It was interesting. I had, it’s a lot of people that went through a lot of things in their lives, and I thought I was the only one that went through something. . . . I’m still learning now, but they educating me and keep my, like I said, my spirit just, you gotta keep the Lord on your side because all things are possible doin.” As was often the case for respondents who found a hobby or positive channel to ground themselves, such investments could also benefit their longer-term educational and career prospects. In addition to the church and his close friends, Tony sought out other mentors at a youth program, who gave him job interviewing skills, which also eventually led to the internship that sparked his dream to become a pharmacist. When he was interviewed, Tony was working through the community college courses he needed to complete to transfer to the University of Maryland and study pharmacy.
Ashley, eighteen, provides another perspective on how some youth found positive outlets to process adverse events. She was among the younger respondents in the study, but easily recounted many disruptive experiences. Across eight moves she made as a child, Ashley recalled navigating violent schools and neighborhoods and the death of a friend in gang violence. When her mother was diagnosed with leukemia, she switched to a homeschooling arrangement but did not graduate. Her father had recently died unexpectedly of heart disease, and she was still grieving when she was interviewed. To cope with these events, Ashley wrote in her journal: “If you go through something, they always say the best thing to get it out is to write it down or talk to somebody. . . . What I write in my journal is stuff that I go through, probably everyday life that you won’t really compare with somebody, [because] they going through something.” Ashley had come to see writing not just as a hobby and an outlet to process the loss of her father, but also as a possible vocation. Although she was struggling to earn her GED by the end of the study, she was still hopeful about going to college and saw journalism as the field of study she might want to pursue.
Selective Avoidance, Selective Investment
The most common response to adverse events in one’s family life (death, drug use, and incarceration) and more general adverse circumstances (neighborhood violence and drug dealing) was to be deliberate about who one spent time with or became friends with, echoing findings from Anjanette Chan Tack and Mario Small (2017) and Holly Koogler (2019). More than 63 percent of youth in the sample mentioned using socially selective or avoidance strategies to avert the trouble they saw around them. Marcus, age nineteen, for example, explained how he combined his passion for sports with selective friendships. He explained:
I like to play basketball, I really like to play, I’m just a sports guy. I stay out of trouble a lot. I will be outside all year, I will be around a lot of people. I’ve got two good friends and one girlfriend. That’s about it. . . . Stayin out of trouble, like don’t hang around people out there sellin’ drugs or nothing, or I know a lot of people out here sellin’ drugs, I don’t be around that. I’ll be out or I’ll be in the house all day, that’s about it. And [with] my mother—that’s about it. . . . we just play all day till the sun go down.
Jessica, twenty-three, described how a friend group focused around a common set of educational goals helped her stay on track. She spoke glowingly about friends she made in elementary school and kept through college, a group who referred to themselves as the Circle of Success. She spoke of these friends as “bright” and “intelligent,” and described how they made it easy to both have fun and stay in school and on the path to college. Jessica delineated between this friend group and her peers in the neighborhood whom she referred to as “knuckle heads.” She elaborated, “I don’t talk to them, they are crazy. These people don’t have no goals. I don’t have nothing to do with that.” Through these selective friendships, Jessica was able to develop an identity related to academic success that distinguished her and her friends from their other high school peers.
Much like Jessica and her Circle of Success, Megan, seventeen, avoided friends who did not have the same long-term plans as she did:
Certain people I do avoid, just because they don't have the same goals as me. I'm worried about getting money for school books and they're worried about getting money for a new outfit. And some of the girls are really promiscuous and I prefer not to hang out with them, just because like—not because they’re bad people, it’s just because they do things that I’m not comfortable with and I don’t wanna hang out with them and like pick up their bad habits.
Megan also bonded with a mentor—a middle school art teacher who took her to high school open houses and helped her fill out applications. They still stayed in touch through email. When she was interviewed, Megan was in her senior year of high school and focused on applying to out-of-state four-year colleges to study broadcasting or film.
Developing Relationships with Family and Other Adults
Some youth also invested more time in their relationships with immediate or extended family as they sought to avoid negative peers. Delmont, nineteen, had endured one friend's death and another’s being incarcerated, as well as housing instability that led to multiple school changes. These experiences left him very reluctant to meet and engage with new people. As Delmont explained, “I do not wanna know that many people” so that no one has “no reason to have an altercation.” Asked what he did to avoid trouble and stay safe, Delmont explained that he spent much of his time each week with his father and uncle, cultivating his passion for rapping and exercising to stay in shape for his football team. When it came to rap, he shared, “First, it was jokin’ around, but then we got serious with it and people actually like it. . . . I try to do it, I would do it every day if I could, ‘cause I got a passion for it.” By the time Delmont was in high school, he and his father and uncle regularly performed at paid shows in Baltimore. When the research team last spoke with him, Delmont had been admitted to a four-year college in Baltimore, where he planned to start school the following fall.
For Bella, nineteen, playing sports and getting involved with clubs at her high school brought her closer to her teachers and a small group of friends, while keeping her away from peers she saw as being on the wrong path. “I wasn’t a very friendly person in high school just because the people that were there were not people I would want to have involved in my life. [My high school] was not full of a lot of positive people.” She credits one of her teachers, who was also her basketball coach, for seeing something in her and guiding her plans. She explained how her teacher/coach didn’t want Bella “to fall down that path of others and get distracted from school, so in order to keep myself occupied that’s when I really became in touch with the groups and sports and all that stuff.” Bella ended up playing multiple sports and joined the school’s entrepreneurship program, in addition to participating in the volunteer club. She developed a small set of friends she remained close with. Staying this busy had a benefit: “I didn’t get home ’til late, but it kept me out of trouble.”
After high school, Bella was able to draw on the support of her coach, who was “a really big help,” to successfully navigate the process of applying for academic scholarships. When Bella was interviewed, she was heading into her second year at a four-year college in Maryland, pursuing business and finance, interests she developed while participating in her high school’s entrepreneurship program. She was already thinking about pursuing graduate education.
Tough Trade-offs (Survive But Maybe Not Thrive?)
Another set of responses to adverse events was a sort of triage approach, where the loss of family income, sick relatives, or fear of personal safety at school required decisions and strategies in the immediate term. Some youth dropped out of school or changed their educational plans to take care of relatives.10 Others withdrew from social life more severely, “hunkering down” and rarely leaving the house, avoiding school and avoiding friendships. Although the rare exception, a few youth in the sample turned to selling drugs to make money. Unfortunately, these decisions, understandable in the context of family and other needs, risked diminishing longer-term educational or professional trajectories.
Leaving or Delaying School to Care for Family
One tough decision that youth faced was whether to stay in high school or college or to take care of family in times of need. For example, Sherika, seventeen, described how she had generally earned mostly As in her ninth grade year but struggled to maintain those grades when she started skipping school to keep her sister safe from an abusive partner. Sherika explained, “I have a sister who was being abused and she lived right around the corner from me. And she would call me and she was scared to stay in the house. And I knew if I was there, he wouldn’t touch her. And she wanted me to stay with her, so I stayed with her.” She said that this situation made it hard to be at school on time or at all some days and that skipping was not something she enjoyed: “It wasn’t like I was just hooking to have fun. It was like, I felt like I was here to protect my sister.” Although she felt she kept her sister safe, she still feels as if it was a “mistake,” because she received very poor grades for many of her tenth and eleventh grade classes, something that weighed heavily on her as she looked to the future.
Isaac, twenty-three, made it to a two-year college despite significant difficulties growing up, including the death when he was young of his father and getting arrested shortly after graduating high school, serving a six-month sentence for agreeing to give his friends a ride after they had committed a robbery. When Isaac was in his second year of college and on track to graduate with an associate’s degree, his twin sister died unexpectedly of a rare disease, leaving behind a son. With no other siblings and a mother who has health issues related to a disability, Isaac felt as if he had no choice but to leave school and take care of his nephew. When his sister died, he told the interviewer, “I just dropped everything and I just came home.” Isaac expressed how he wanted to finish college but it simply was not possible in the near term given his new life situation: “It’s not that I don’t want to go back to school, it’s just like trying to find time with school and then my nephew and then work. . . . My concern right now is money.” Isaac had not returned to school by the end of our study.
Just after Sierra, twenty-one, graduated high school, her mother lost her job at a hospital, leaving them both worried about how they would make ends meet. Sierra described how this destabilizing “downfall” meant her own college dreams had to be put on hold: “We had like this little downfall in our family, whereas I had to wait a while [to go to school], I had to wait. . . . I was thinking about Baltimore City Community College . . . take some classes, get a little education. But I was like, ‘I’m gonna wait, I’m gonna try to help my mother. I don’t wanna be crying cause we’re homeless or we don’t have this or that, I didn’t want that. So I waited.” Sierra described the unfortunate trade-off: “[right now] I’d rather have a home than my education.” Looking ahead, she still wants to go to community college, but explains that she “wants to be sure” that her family is on solid ground before she feels that she can pursue her original goal.
Similarly, Martin, twenty-two, recounted the difficulties he endured in high school, shuffling between schools and hanging out with a girlfriend who got in fights and was eventually expelled. In the eleventh grade, Martin was put in the difficult position of having to either continue in high school or help out when his mother was hospitalized: “I actually quit school when I was like seventeen. I basically stopped because my mother, she was in the hospital. . . . So, I basically did what I had to do as far as going and getting a job and, you know, even though I wasn’t making much of nothing, but I buy food, buy toilet paper, buy milk, or whatever else I needed to do for the house.” Shortly after, a subsequent girlfriend got pregnant, and Martin also took on the caretaking responsibilities of a new baby with her. By the end of the study, Martin had made a few attempts but had yet to earn his GED.
Withdrawal from Social Ties and School
Whereas Delmont, Bella, and Jessica protected themselves by avoiding certain people and places, instead turning selectively toward some of their family members and friends, other youth turned away from social ties more severely, deliberately isolating themselves in an effort to focus on their goals, and stay safe, themes also discussed by Peter Rosenblatt, Kathryn Edin, and Queenie Zhu (2015) and Holly Koogler (2019). Previous adversity had led to a generalized sense of mistrust so pervasive that some youth thought it was better to go at it alone, echoing Sandra Smith’s (2007) account of defensive individualism. For example, when asked, “And so can you tell me about your two closest friends?” Nathan, nineteen, responded, “I don’t have friends.” He avoided even clubs and activities in high school, “so I don’t lose track.”
Ralph, twenty-one, experienced a range of adverse events growing up—including his mother’s drug abuse and a high school where violence and disruption were common. In the front of his mind, though, was his experience of having been recently attacked by a man who was part of a group that Ralph considered close friends. As he told one interviewer, “It was . . . very unexpected. I didn’t even suspect nothing like that. I just . . . man, I told myself: that will never happen again.” In light of the pain and shock of this event, Ralph felt he was better off not making friends and that he didn’t need them: “I don’t do too much hanging no more. This world, man. I don’t like hanging out no more. . . . I cut all these friends; I don’t have no friends. I don’t want no friends, I am finished with friends. I am a grown man, so I don’t need no friends. I am better off by myself.”
Tyler, twenty-one, said that he rarely left the house at all: “I feel like that’s the best way to protect yourself. Rather than shootin’ or stabbin’ or killin’ somebody I’d rather just stay in the house because that’s all, that’s the only other outcomes that’s outside.”
When Michael, eighteen, was interviewed, he said, “I really have no close friends, to tell you the truth. . . and really my friends down here I don’t even hang around cause all them is just doin’ all the wrong things, which I’m not tryin’ to do, so I just stay away.” He described how his decision to socially isolate was motivated by a desire not to end up in prison again, and by a lack of trust in people after one of his friends was killed by another friend. Michael did time for selling weed, although he reflects that things could have been worse. “I could of got locked up for serious stuff that I didn’t get locked up for. And that must be, I just thought like that must be my wakeup call.” He says that another reason for turning himself around was to be a good role model for his little brother.
Another tough trade-off was the decision to not attend school in order to stay safe (Burdick-Will 2013). A recurring theme in an interview with Kelly, twenty-two, was that her high school could not keep students safe, let alone provide an environment conducive to learning. On one occasion, she related, “There was a riot after school. Everybody was throwing food. It was people that didn’t go to the school up there sitting at the lunch table and people didn’t even [know] that they was in there at first until the fight had started. And I remember a couple people getting stabbed, people getting maced, it was terrible.” As Kelly reflected, “It just be crazy. Like, we were not here for this. We was supposed to be learning.” Kelly explained that with all of the fighting and violence, “I was like, I would rather hook school then to be in here with all this other stuff. So, I would hook school and smoke with my friends and stuff like that.” Although this strategy took her out of the chaos she perceived in her high school, it also meant that graduating from high school became a more distant possibility. Kelly was subsequently expelled in tenth grade for chronic absenteeism. Kelly reflected on this period, saying, “Now, I look back on it—it wasn’t good.”
Selling Drugs to Earn Money for Family
Turning to the street, selling drugs, or committing crimes was the exception rather than the norm in the study, with only twenty-six of 150 youth ever reporting such activities. In fact, it was the least desirable way to end up, according to the youth in our study. While rare, Jayden’s case provides an example of why one might do it to survive. Jayden, twenty-two, began selling drugs when he was in his mid-teens, after his stepfather recruited him, and “sat me down and told me everything, as far as the prices and what was, how much you can make off certain drugs and everything.” He was tired of never having any money and was influenced by a friend who always had money from selling. When asked the best job he ever had, he responded, “selling weed,” because, he explained, “More money. I don’t never had to worry about nothing, like a bill had to be paid, it wasn’t nothin’. Just touch what you saved and—yeah. All that, I saved up a lot of money, selling marijuana.”
Jayden mostly stopped dealing when his daughter was born, saying that he’d “rather work for a paycheck” than getting locked up or killed. Two years before he was first interviewed, his grandmother passed away, and his mother was committed to the psychiatric unit for eight months for depression, hallucinations, and detoxing from alcoholism. He drained his savings during this time because he was solely in charge of the household, and he paid the hospital bills for his mother’s stay. He said that once he was caring for his daughter and mother, that was the point at which he felt like an adult. Yet he was struggling to get the kind of job with which he could adequately support them like one. He hated his job at Kentucky Fried Chicken, saying it made him “feel like a loser” and that he longed for an office job. He applied for other jobs but said that mostly he was not called back. When asked why, he said he assumed it was how he looked, “Probably my hair. But I go to a job, I make sure I get my hair done, get it re-twisted.”
As Jayden explained, most of the better jobs want “college.” He said, “I think about going back to school a lot. I keep telling myself, by the time I’m a certain age, I want a certain amount of money saved up. But when I told you I went to ITT [for-profit school], I never paid them, so they’re threatening me with garnishing my taxes and stuff, so a bank account right now is out of the question. But if I put my little $500 paycheck in there and they take it from me, I’m going to be sick.” He was frustrated with his current work options and considered getting involved in drugs more seriously again. He explained, “I don’t want to hustle, but if, if I have enough money to invest in a friend that I know that’s getting money, I will invest. I’m not gonna say I’m not. If I give him five hundred and he can turn it into a thousand, I’m gonna do it.” When the research team last spoke to him, he was selling weed on a small scale to people he knew well.
A Model of “Surviving Versus Thriving”
The first response category, whereby adversity can derail individuals’ lives, suggests a possible lack of agency. The second and third categories suggest agency in the form of active decision-making or strategizing in the face of adversity. We cast such responses to adversity as potential investments in human capital. We emphasize the third category, characterized by a trade-off between surviving now and thriving in the future, because it constitutes a novel addition to traditional views of human capital accumulation—and yet is a straightforward extension of basic frameworks. As we show, the model easily accommodates the second category of responses and, moreover, can provide an interpretation of the first category. We also discuss possible extensions to the theory along with ways we could use data to calibrate or to estimate model parameters, though we caution that the qualitative data used in this project may include too few observations to accomplish this effectively. We therefore discuss how larger-N data sets, such as the National Education Longitudinal Study or the National Longitudinal Survey of Youth 1997, could be explored for this purpose.
The Basic Setup
To begin and to fix ideas, we focus on “surviving” and envision a two-period model and a single decision. In the first period, an agent decides whether to make an investment or to engage in a strategy G∈{0,1} that is potentially costly (where the cost is denoted c) but raises the probability of survival from π(0)∈(0,1) to π(1)∈(0,1). If the agent does not survive until the next period, he receives utility U(N), which is normalized to zero. If he survives until the second period, he receives the utility of surviving, denoted U(S), which is distinguished from that of thriving, denoted U(T), where we assume that U(T) > U(S) > 0. The second period is discounted by factor β∈(0,1).
The model could easily accommodate different kinds of strategies distinguished by different costs c and different probabilities of survival π(G). These factors could also be a function of the kinds of adversity an individual faces. Moreover, the model could be extended to incorporate a menu of potential strategies rather than a binary choice, each option distinguished by different costs and probabilities. Finally, an individual’s perceptions of the strategy could influence their choice to engage in a strategy; for example, an individual may not have a correct belief of π(G). However, rather than extending the model in these directions, we offer a highly stylized version that allows the individual with correct perceptions of π(G) to engage in a strategy G or not.
The lifetime values (the sum of utility over two periods) for choosing G or not, denoted V(G = 1) and V(G = 0), respectively, are given by the following two equations:
The individual will choose G = 1 if V(G = 0) < V(G = 1), which is equivalent to the following inequality:
This expression means that the agent will engage in strategy (G = 1) when the future benefits of doing so exceed the upfront cost. The benefits are the discounted marginal increase in the likelihood of obtaining the utility from surviving by engaging in the action. To fix ideas, a person would not engage in a costly strategy if they heavily discounted the future (β close to zero), if doing so increased the likelihood of survival only slightly ([π(1) – π(0)] close to zero), or if the utility of surviving were sufficiently low relative to not surviving (U(S) close to zero). Notice, the individuals in the first category (Anna and Marco) may not see the point of making investments because the degree or complexity of the adversity they face means that they see few options that could have an appreciable impact on their likelihood of avoiding a continued downward trajectory toward some of the worst outcomes, which would mean investing has little payoff.
The model allows for the possibility that engaging in G is enjoyable as some strategies may be, for example, joining a sports team or engaging in extracurricular or other personally fulfilling activities that generate utility. An important example is an identity project. If this is the case, the cost c is negative, which means that inequality (3) always holds given that the right side consists of the product of three positive expressions and is thus always positive.
Extending the Model to Capture When Surviving and Thriving Are at Odds
Having exposited a version of the model where survival is the only concern, we now incorporate a distinction between surviving and thriving as we discussed earlier. The key addition to the model is allowing engagement in strategy G to have different effects on the likelihood of surviving versus of thriving in a potentially countervailing manner. The earlier notation remains unchanged. However, we add the probability of thriving θ(G). In this case, both the probability of surviving and of thriving are endogenous to the choice of G, which distinguished this model from those that allow investments in human capital to be a function of exogenous shifts to survival. The lifetime value of G = 1 versus G = 0 is now as follows:
The individual will choose G = 1 if V(G = 0) < V(G = 1), which is equivalent to the following inequality:
This somewhat more complicated expression is interpreted similarly to inequality (3): the individual will choose G = 1 if the benefits exceed the upfront costs. As before, the benefits include the discounted increase in the likelihood of enjoying the utility of surviving, which is the first expression on the right side of the inequality. However, there is now a second expression, which is the additional increase in utility from thriving (which is of course contingent on surviving, which is why this expression also includes survival probability). Notice, if the likelihood of thriving is zero (θ(1) = θ(0) = 0), inequality (6) collapses to inequality (3). Likewise, if there is no utility difference between surviving and thriving (U(T) – U(S) = 0), there is no additional benefit to choosing G = 1 and, again, the agent faces the same cost-benefit analysis as captured by inequality (3). Hence, the key addition to the model is the second expression on the right side, which is the impact of choosing G=1 on thriving versus surviving.
The crucial expression in inequality (6) is π(1)θ(1) – π(0)θ(0), which captures the interplay between the likelihood of surviving and, conditional on survival, on thriving. To fix ideas, suppose the agent faces little serious adversity and so survival is assured whether he engages in G. If so, the probabilistic expression is simply θ(1) – θ(0), which means that choice to engage in G is driven by the additional probability of thriving. This appears to be the case for no individual in our sample. Alternatively, suppose that by choosing G the agent can affect survival but has little influence on thriving. In that case, θ(1) = θ(0) ≡ θ in which case we can write the probabilistic expression as θ[π(1) – π(0)]. This means that the agent faces increased incentives to engage in strategy G = 1 because it not only raises survival but also means that the agent, if he survives, may also enjoy the higher utility generated by thriving even though the choice of G has no direct impact on the likelihood of thriving. This situation seems consistent with the individuals who responded to adversity by engaging in activities that will keep them safe in the short term and also help them succeed in the long term; consider Tony and his church or internship or Jessica and her tightknit Circle of Success.
An interesting case arises when π(1)θ(1) – π(0)θ(0) > 0. This occurs if the same strategy G raises the likelihood of survival, but decreases the likelihood of thriving so much so that it lowers the total benefit in inequality (6). For example, consider strategies that raise the likelihood of survival, such as withdrawal from activities, self-isolation, and caring for family in a way that crowds out other opportunities. These strategies might help ensure a basic level of utility and avoidance of some of the worst circumstances, that is, survival but might also essentially preclude thriving. In this situation, the agent faces a terrible trade-off between surviving and thriving. If he chooses G = 1, he survives, but has little chance of thriving once he survives. Alternatively, if he chooses G = 0, he is more likely to thrive if he survives (which would generate U(T) > U(S)), but is also less likely to survive, which means he may receive zero. More simply, whereas some strategies mean that agents raise the likelihood of receiving U(S) and perhaps U(T), other strategies mean that the agent chooses between a high likelihood of simply surviving U(S) versus a high likelihood of thriving, but also of not surviving at all, that is, a higher probability of either zero or U(T).
The model as outlined assumes that the agent is fully aware of the potential trade-offs between surviving and thriving when choosing G. It is also possible (likely) that agents are not fully aware of this trade-off, in which case survival strategies can backfire. For example, when facing inequality (3), the agent may assume that the strategy increases survival probability and has no impact on the likelihood of thriving, which increases their incentives to choose G = 1 since θ > 0 and U(T) > U(S). However, in reality, the second expression on the right side of equation (3) could be negative. If the costs are high enough, an agent may choose G = 1 even though they would optimally choose not to if they were fully aware that doing so might diminish the likelihood of thriving.
The albeit very simple and stylized model nonetheless highlights how in formally distinguishing between surviving and thriving, we are able to capture several potentially important reactions to adversity. For many people, survival (broadly construed to include avoidance of some of the worst outcomes) is all but guaranteed and the relevant trade-off is between the cost of investments or strategies and their positive impact on thriving. This possibility aligns with typical economic models of human capital accumulation via costly investments among youth who are not severely disadvantaged or facing substantial adversity. The model also allows for situations where actions have little bearing on survival or on thriving, which means that the payoff to investments is small. This would be observationally equivalent to assuming the agent has little to no agency. The model also captures the possibility that both survival and thriving are a concern, but where a disadvantaged agent can do little to influence the likelihood of thriving and only faces options that raise the likelihood of survival. It also allows for options whereby surviving and thriving go hand in hand. Finally, the model accommodates a possibility articulated by several of the respondents earlier: strategies, investments and actions that increase survival probabilities can reduce the likelihood of thriving, so much so that agents face a trade-off between the two. Understanding the relevant trade-offs can also offer lessons for policy. For example, school-based enrichment programs can be a helpful lifeline for disadvantaged students, but will be of limited value if students do not feel safe at school and thus avoid them. Notice, once again, that the model does not require individual irrationality or specific pressure from a peer group or a community to invest little in one’s human capital. Instead, the central focus of the model is a tension between surviving and thriving such that lower investments in human capital are an optimal response to a brutal trade-off.
CONCLUSION
In this article, we examine heterogeneity in disruptive events and adverse conditions among a sample of disadvantaged youth growing up in high-poverty and racially isolated communities in Baltimore. Although many studies might group these respondents together based on their similarity along dimensions used to explain variation in human capital outcomes (they are all Black and from low-income families), we provide evidence of rich heterogeneity along several important dimensions, including what specific kinds of adversity respondents reported and in how they perceived their adversity. Importantly, we provide evidence of differences in how individuals responded to adverse events. Some were essentially derailed by traumatic events such as homelessness, deaths in the family, or parental incarceration. In the face of stark adversity, deliberate and thoughtful choices might not be expected to influence outcomes and so individuals made sometimes dangerous and harmful reactive choices. Others appeared to make more deliberate decisions in the face of adversity with both short-term and long-term implications. Some choices not only kept them safe in the short-term, but also seemed likely to serve them in the long run, akin to making deeper investments in human capital. We also find evidence that some youth, when developing strategies in response to adversity, appeared to face a trade-off between surviving (interpreted broadly to incorporate avoiding some of the most costly outcomes) and thriving.
We formalize this trade-off in a model, which extends basic human capital theory to allow for the possibility that some actions that might be thought of as good decisions in the sense that they increase the likelihood of future success (thriving) do so at the cost of also increasing the likelihood of never getting to the future (not surviving). For example, leaving school to make money to support a family is not a recipe for success, but it is also a way to avoid a disastrous short-term outcome, such as not being able to pay the hospital bills for your mother and further burdening her. The model not only provides some guidance for further empirical research; it also illustrates that a straightforward addition to a basic dynamic model of human capital accumulation allows it to capture a type of dynamic trade-off that may be largely irrelevant for middle- or high-income youth, but may nonetheless be pervasive in the lives of the disadvantaged youth in this study.
These ideas can lead to two changes in how we approach adversity in the social sciences: improvements to data and improvements to models. The former point is simply that larger-N data collection of factors we examine (such as coping strategies) would allow us to draw more conclusions about their role in predicting variation in trajectories. The latter point is that models of decision-making that omit trade-offs relevant to respondents who face adversity should be modified to capture how adversity can perpetuate inequality in part through the adoption of strategies that allow people to survive but dampen their ability to thrive. Indeed, an implication of the model is that we need not focus on suboptimal decision-making or cultural factors or social norms to understand why individuals make choices that undercut future thriving.
The model provides a starting point for more elaborate models that could be used to develop hypotheses that could be tested using larger-N data sets or, if matched to data (such as using structural econometric estimation techniques), to simulate how policy interacts with decisions surrounding adversity to drive inequality. Indeed, better data coupled with models that incorporate such channels could be used to explore policies that specifically target individuals forced to contend with adversity. The ultimate goal is to explore, devise, and design policy interventions that relax this particular trade-off so that individuals who are born into disadvantage and face various forms of adversity need not be put into the unfortunate and untenable position of choosing between surviving and thriving. Certainly, some of these policies must confront larger upstream structural sources of racial and economic inequality. But even many of these policies lack an understanding of how youth make decisions within unequal contexts, which we believe is crucial to better policies and better outcomes.
A natural extension of the current study would be to consider how different categories of strategies map to different outcomes. Unfortunately, data collection ended just as the individuals in the sample were beginning adulthood and thus it would be premature to draw strong conclusions. Moreover, we must be mindful of additional limitations to the current study. We focus on a small sample of Black youth in one city. Data from another context (such as non-Black or rural youth) could provide different lessons, as we know literature points to challenges faced by families in rural areas (Edin, Shaefer, and Nelson 2023) and among youth from indigenous communities, for example. Another useful exercise, which we leave to future research, would be to use larger-N data sets to more carefully examine investments in human capital that could put individuals at immediate risk of losing something critical (such as attending school despite violence at school). In general, larger data sets could be used to corroborate and further explore the degree to which the types of strategies and trade-offs we observe and analyze here are more general feature of adversity among disadvantaged youth that helps to explain heterogeneous responses to it.
FOOTNOTES
↵1. A natural extension of the analysis would be to relate adversity and strategies in the sample to long-term outcomes, but the data are not well suited to this exercise. Most of the youth were just entering adulthood when they were last interviewed, and follow-up fieldwork was conducted with only a subset of respondents up to two years later. It would therefore be premature to make strong claims about the long-term impacts of varied responses to adversity.
↵2. Different methods are not always used in tandem for practical reasons as well. For example, structural econometric models are computationally burdensome, which limits how many variables or mechanisms can be incorporated if the model is to remain tractable. Relatedly, qualitative data that generate rich narratives and could help inform conceptual models are not always well suited for causal inference and are labor intensive and expensive to collect as well.
↵4. The MTO experiment is not a focal part of this article; rather, participants whose families were part of the initial MTO experiment make up a population of economically disadvantaged individuals from which we use the data to understand adverse events in the lives of youth. While the experiment, which reduced exposure to neighborhood poverty, also reduced exposure to adverse events for some youth (namely, victimization and safety issues in the neighborhood, see DeLuca, Clampet-Lundquist, and Edin 2016), the extent and frequency of adversity experienced by youth in the study is still significant.
↵5. The more detailed sample distribution is as follows: fifteen to sixteen years old (14 percent); seventeen to eighteen years old (17 percent); nineteen to twenty years old (27 percent); twenty-one to twenty-two years old (34 percent); and twenty-three to twenty-four years old (8 percent).
↵6. By the end of our study, only one youth had completed a bachelor’s degree.
↵7. We also found a range of experiences, decisions, and outcomes that are harder to cleanly categorize as adversities per se, rather than as consequences of adversities or one’s own decisions; we consider these “gray areas” and describe them in the appendix to be comprehensive.
↵8. For about one-third of the sample, we could not identify a response to adversity in the interview data, or the response was hard to categorize.
↵9. For the full sample, of those for whom a response could be identified in the interviews, 10 percent reported derailed or reactive coping responses, 64 percent reported protective or promotive strategies, 17 percent reported tough trade-offs, and 9 percent reported other types of responses not easily categorized. Some gender differences appeared in the proportion of the sample who reported any response to adverse events during the interviews: 81 percent of the male respondents and 61 percent of the female respondents. These differences are likely due to the types of responses reported—more male young adults reported protective social avoidance strategies as well as more severe social isolation, for example, which combined were also the modal subcategories of responses. Conversely, reporting responses differed little by age; 70 percent of those nineteen to twenty-three years old reported responses and 74 percent of those fifteen to eighteen years old did. However, we caution against interpreting these rates of response and gender differences as reflective of all responses youth ever had to adversity; as noted in the methods section, we did not ask about these experiences directly, and therefore assume what we report here to be an undercount of the true prevalence of adversity as well as responses to adversity.
↵10. The literature in gerontology and economics on the negative labor-supply consequences of caregiving, especially among women, is huge (see Burton 2007).
- © 2024 Russell Sage Foundation. DeLuca, Stefanie, Nicholas W. Papageorge, and Joseph L. Boselovic. 2024. “Exploring the Trade-Off Between Surviving and Thriving: Heterogeneous Responses to Adversity and Disruptive Events Among Disadvantaged Black Youth.” RSF: The Russell Sage Foundation Journal of the Social Sciences 10(1): 103–31. DOI: 10.7758/RSF.2024.10.1.05. We thank Kathryn Edin and Susan Clampet-Lundquist, the co–principal investigators with Stefanie DeLuca on the MTO Q10 Transition to Adulthood Study in Baltimore, which provided the interview data we use in this article. We are grateful for the generous support of the Russell Sage Foundation (#1808-07819, ROR: https://ror.org/02yh9se80) and the William T. Grant Foundation (#9031). We also acknowledge the excellent research support of Paige Ackman, Lidie Ataoğuz, Olivia Cigarroa, Jamie Chan, Courtney Colwell, Kendall Dorland, Thelonious Goerz, Matt Gonzalez, Min-Seo Kim, Hannah Lee, Olivia Morse, Lauren Ricci, Jasmine Sausedo, Jane Scinta, Claire Smith, Margaret Tydings, Oscar Volpe, Yuwen Wang, and Eliza Zimmerman. We also thank the editors of this volume, the external reviewers, and the participants at the June 2022 workshop for this volume for their valuable guidance on improving the manuscript. Direct correspondence to: Stefanie DeLuca, at sdeluca{at}jhu.edu, 3400 N. Charles Street Baltimore, MD 21218, United States.
Open Access Policy: RSF: The Russell Sage Foundation Journal of the Social Sciences is an open access journal. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
REFERENCES
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.