TY - JOUR
AU1 - Smith-Greenaway,, Emily
AU2 - Yeatman,, Sara
AB - Abstract The rapid expansion of schooling across low-income countries, combined with intensive governmental and nongovernmental efforts to promote education, has encouraged youth in these contexts to form exceptionally high educational expectations, despite immense structural barriers to achieving them. Consequently, many young people’s educational expectations go unmet, driving concerns over the possible unintended consequences, including their elevated risk of mental health problems. At the same time, role transitions (e.g., marriage, parenthood) remain important elements of the transition to adulthood in many low-income countries, and may be a source of resilience—allowing youth to shift their identity away from education towards a new role. In this study, we explore the mental health implications of young women’s unmet educational expectations, and the possible buffering impact of motherhood, in a low-income community in Malawi, in southeast Africa. Analyses of six years of longitudinal data show that young women’s unmet expectations to continue school are associated with multiple indicators of mental health disadvantage across two points in time. In the short term, however, this is only true of young women who did not enter motherhood in the midst of their educational plans going unrealized: young women who became mothers—and thus achieved a key element of the transition to adulthood in this setting—experience fewer mental health disadvantages. The findings demonstrate the potential mental health consequences of young adults’ expectations exceeding their outcomes while also highlighting a source of resilience. Introduction Integral to understandings of human agency and social behavior is the idea that people are future-oriented and commonly anticipate and organize what is to come (Abbott 2001; Emirbayer and Mische 1998; Mische 2009; Tavory and Eliasoph 2013). As such, questions of how people anticipate and organize their futures (Bergmann 1992; Nowotny 1992), and what happens when those futures are not realized (Frank 1935), have long occupied social science. These long-standing questions are of continued relevance to today’s youth, many of whom are navigating adolescence and young adulthood in contexts that explicitly encourage them to form aspirational ideas about their future. The majority of youth now live in lower-income countries (United Nations 2015), and just as poor U.S. youth are highly ambitious (Baird, Burge and Reynolds 2008), youth in low-income countries similarly hold high aspirations and expectations (De Boeck and Honwana 2005; Leavy and Hossain 2014; Leavy and Smith 2010; Sumberg et al. 2017; Tadele and Gella 2012). The relatively recent expansion of formal education in many low-income countries, coupled with the explicit promotion by nongovernmental and governmental agencies of education as a way for youth to achieve a “good life”, likely contribute to this (Harper, Marcus and Moore 2003; Miller-Grandvaux, Welmond and Wolf 2002). Additionally, studies of select low-income contexts suggest educational campaigns have not only convinced youth and their parents to embrace education as a route to a better life, but also as a way to become a good, moral person (Leavy and Hossain 2014; Marshall 2016). Despite positive perceptions of, and strong commitment to, formal education, the unfavorable institutional and economic circumstances in low-income countries often interfere with youth realizing their high educational expectations.1 For instance, studies of youth across Africa show they commonly discontinue school far earlier than expected (Fortune, Ismail and Stephen 2014; Frye 2012; Kabiru et al. 2013). At the same time, mental health is a growing concern in the region (Bloom et al. 2012), raising questions of whether the common mismatch between young adults’ educational expectations and actual attainment elevates their psychological vulnerability (Fortune, Ismail and Stephen 2014; Jones and Chant 2009; Kabiru et al. 2013; Marshall 2016). Evidence that education is an integral part of African youth identity suggests unrealized educational plans may indeed harm their mental wellbeing, leading to minor mental health problems, like dissatisfaction with life and feelings of depression, stress, and isolation, or to more pressing concerns like clinical depression and anxiety. Yet, other research highlights young people’s resilience and flexibility in the wake of unrealized expectations (Martin and Gardner 2016; Reynolds and Baird 2010; Villarreal et al. 2015), and their ability to re-cast their identity when it is threatened (Norris 2016). In many African communities, role transitions (e.g., marriage, parenthood), continue to be central indicators of a successful transition to adulthood. In the wake of unrealized educational expectations, African youth may take refuge in adopting new roles and identities, ultimately buffering themselves from mental health problems. In this paper, we explore these possibilities by studying the experiences of young women in Malawi, in southeast Africa. We use longitudinal data from a representative cohort of young women living in a semi-rural community in Balaka, Malawi, where educational expectations are generally high despite substantial barriers to their achievement. The young women we study almost uniformly expect to continue further in school—an expectation that, two years later, only some achieve. Yet, of those who do not fulfill their educational plans, many experience a major role transition by becoming a mother. With information on their mental health at the time they express their educational expectations, and two and six years later, as well as their schooling and childbearing experiences, we analyze whether the young women whose educational expectations go unrealized experience elevated feelings of depression, loneliness, and stress and lower levels of contentment. Drawing on evidence that motherhood is a source of stability and happiness for young women in poor communities, we further assess whether this role transition buffers young women from any mental health disadvantages in the wake of unrealized educational expectations. Our findings speak to the potential public health risks associated with young people’s exceedingly high educational expectations in low-income countries, yet also attest to young people’s potential resilience when those expectations go unrealized. Background Education and Young Adulthood in Malawi The educational experiences of recent cohorts of youth in Malawi are, in many ways, similar to those of youth in numerous lower-income countries across the globe. In the 1990s, Malawi was at the forefront of progressive educational policies in sub-Saharan Africa, representing a context of immense educational potential in the region. In 1992, the Malawian government implemented a national program that distributed primary school fee waivers to girls (Mundy 2002).2 Soon after, Malawi became one of the first African countries to fully eliminate primary school fees (Stasavage 2005). This led to a dramatic increase in primary school participation. At the same time, the Malawian government and local institutions rolled out a far-reaching “ideological campaign” (Frye 2012:1577) to promote education—particularly for girls. Since the 1990s, the benefits of education for girls, and strategies for achieving success, have been recurrent themes in newspapers, radio shows, magazines, and school curricula in Malawi. These campaigns have cemented young Malawian women’s commitment to education as an integral part of their identity, convincing them education is the path to a “bright future” and success is within their reach, so long as they work hard (Frye 2012). Moreover, young women view educational aspirations and success as revealing key character traits, like good moral standing, the ability to work hard, and seriousness. Despite widespread institutional and cultural embrace of education in Malawi, in the ensuing decades, endemic poverty has hampered educational progress in the country. Against the backdrop of overcrowded classrooms, underpaid teachers, and poorly-resourced schools, most Malawians live on less than US$1.25 per day, and minor school expenses can severely strain household budgets (United Nations Development Program 2010). Thus, although young Malawian women commonly profess expectations to complete high levels of education (Frye 2012), the majority drop out prior to finishing—or even prior to attending—secondary school (Smith-Greenaway 2015). After discontinuing school, young Malawian women often profess a desire to re-enroll—attesting to the centrality of educational goals to young women’s visions for their futures (Deterding 2015; Frye 2012). Young Malawian women’s unwavering commitment to education is, in part, grounded in their desire to secure careers and financial stability (Frye 2012:1579), as in other low-income countries (Fortune, Ismail and Stephen 2014; Marshall 2016). The Malawian government explicitly features college-educated elites in their educational campaigns—placing education at the center of rural women’s understanding of how to achieve economic prosperity (Frye 2012). Like in other African countries (Locke and Lintelo 2012), in rural Malawi, opportunities for steady employment and salaried work are rare (Swidler and Watkins 2009). As a result, most rural Malawians—even the highly educated—engage in subsistence farming, petty trade, and/or temporary work. Unrealized Educational Expectations as Consequential: Identity Concerns and Stress Although mental health data on young Malawians is scarce (see discussion in “Measures” section), one study reports that depression is the fourth leading cause of disability in Malawi (Bowie 2006). The World Happiness Survey further suggests that poor mental health is common in Malawi. Relative to other countries in sub-Saharan Africa, which has the lowest levels of life satisfaction of all world regions, Malawi ranks 29th out of 44 countries for which there are data (Helliwell, Layard and Sachs 2017).3 Additionally, a study of depression and anxiety among older Malawians suggests that more than half of mid-50 year old women’s remaining life is spent depressed and/or anxious (Kohler et al. 2017). Together, these findings attest to the need to examine the possible role of unrealized educational expectations in negatively affecting Malawian women’s mental health early in the life course. Extending insights from social psychological research, one way that unmet educational expectations may compromise young Malawian women’s mental health is by negatively affecting their identity. Individuals’ social statuses (Pearlin 1999; Simon 1997; Thoits 1986), as well as the expectations and aspirations associated with those statuses (Emirbayer and Mische 1998; Hitlin and Johnson 2015), are key elements of identity. In Malawi (Frye 2012), like elsewhere (Deterding 2015; Nielsen 2015), individuals’ pursuit of education, and their future educational plans, are meaningful expressions of self and are a key component of identity. As a result, unachieved educational expectations could fracture young adults’ identity. A discrepancy between one’s idealized versus achieved self leads to feelings of dejection (Higgins 1987; Large and Marcussen 2000; Marcussen 2006), negative emotions (Higgins 1987, 1989), and depression (Sheeran and Abraham 1994; Sheeran and McCarthy 1990, 1992). As shown to be true of youth in other contexts (Silva 2013), young Malawians may experience anxiety as they try to re-create an identity. Moreover, in losing a primary sense of their identity as a student, young Malawian women may feel isolated from those in their peer network who are pursuing education, and thus at risk of social isolation and loneliness. Additional research suggests that unrealized educational expectations may make young women question themselves and their worth in fundamental ways, which could further take a toll on their self-esteem. Frye’s (2012) study—conducted with young women in the same community where data for the current study originate—reveals that young Malawians view educational success as signaling personal fortitude, whereas they associate discontinuing school with character flaws. This corresponds with evidence that Ethiopian youth perceive peers who have discontinued school as “lazy,” “disrespectful,” and “lacking manners” (Marshall 2016). Young Malawian women’s tendency to attribute educational failure to personal traits illustrates their strong sense that young adults have individual agency over their future trajectories. As a result, young Malawian women whose goals go unachieved may be vulnerable to self-blame and negative self-perceptions, and in turn, disappointment, discontentment, and depression. Young Malawian women who discontinue school earlier than planned may also experience stress as they face an unfavorable employment situation with less education than planned. Akin to Tavory’s (2009) description of flirtation, enrollment in school has a “future orientation” built into it—students can focus on the possibility of a future career and financial security in the abstract without having to navigate and sort its actualization. Conversely, being out of school ends young Malawian women’s “interstitial pause” (Johnson-Hanks 2014:28) and forces them to formulate a strategy for how to sustain themselves and their family. This may be especially stressful in economic contexts like rural Malawi, where young women have few options for achieving the life out of chronic poverty they envisioned an education would enable. Although educational success by no means guarantees occupational and economic success in this context, the young women who are out of school and navigating their next steps earlier than they expected may be particularly vulnerable to mental health problems. Unrealized Educational Expectations as Inconsequential: Sources of Resilience The likelihood that these two mechanisms—stress and issues of identity—lead young Malawian women to experience mental health problems in the wake of unrealized educational expectations may depend on other aspects of their circumstances when they discontinue school. Evidence from other contexts emphasizes young adults’ ability to adapt to the unexpected (Elder 1999; Mortimer et al. 2002) and to identify ways to buffer themselves from psychological harm (Cerulo 2008; Martin and Gardner 2016; McLeod and Almazan 2003; Mortimer et al. 2002; Reynolds and Baird 2010; Shanahan and Mortimer 1996; Thoits 1994; Villarreal et al. 2015). One life course development that may foster such a response in young women is their adopting a new role or identity (Gecas and Seff 1990; Norris 2016; Sieber 1974). Unlike in the United States, where the transition to adulthood is shifting away from empirical markers (Deterding 2015; Silva 2012, 2013, 2016), in Malawi, the transition to adulthood still hinges on achievement of key roles like parent and spouse. In Malawi—a pro-natal, high fertility context—motherhood is an especially prominent component of women’s identity and is key to attaining full adulthood (Evens et al. 2015; Yeatman and Trinitapoli 2013). Even educationally ambitious young women in Malawi value motherhood; thus, for women whose educational goals go unrealized, becoming a mother affords a new, socially valued identity that may help them detach from their identity as an aspiring student. Moreover, as it does for women in other contexts marked by poverty and uncertainty (Edin and Kefalas 2005; Nomaguchi and Milkie 2003), motherhood may provide young Malawian women a “chance to do something well—a new game with high stakes that they have not failed at yet” (Silva and Pugh 2010:614). Achieving the socially valued role and identity of mother may foster young women’s resilience in the face of educational expectations going unrealized. It is possible, however, that the psychological benefits tied to motherhood are concentrated in the short term. Just as Edin and Kefalas (2005:20) discuss the “euphoria” young couples experience immediately following a child’s birth in the United States, in Malawi, having a child may be an especially effective distraction in the immediate wake of other plans going unrealized. Over time, however, women may dwell on their desire for more education in order to provide for their children (Deterding 2015). The challenges of motherhood—especially with less education than initially envisioned—could even put women at risk of mental health problems over time. Data and Methods Sample To examine whether educational expectations are consequential for young women’s psychological health, and if so, under what circumstances, we analyzed data from Tsogolo la Thanzi (TLT),4 a study in Balaka—a town in southern Malawi. The first wave of data was collected between May and August 2009 among a simple random sample of 1,505 15- to 25-year-olds drawn from a sampling frame of the full population living within a seven-kilometer radius of the center of Balaka. This area includes residents in the town and the surrounding rural villages. TLT successfully enrolled approximately 96 percent of recruited respondents. The study collected eight waves of data at four-month intervals through 2011, and a follow-up wave in 2015. We constructed our target analytic sample from women who were 15 to 19 years old, unmarried, and childless at wave 2 when they were asked about their educational expectations (N = 502). This restriction enabled us to focus on the women most likely to expect more school and to observe their transition into motherhood. Note that this restriction excluded six unmarried, childless 15- to 19-year-olds who were interviewed in wave 1 but not in wave 2. In wave 2 (collected between October and December of 2009; hereafter “2009”), interviewers asked respondents a series of questions about their expectations for their future “in two years.” We examined educational outcomes two years later (in wave 8, collected between October and December of 2011; hereafter “2011”), to assess how unrealized educational expectations correspond with young women’s mental health at that time (in 2011), and then four years later (in 2015). Of the focal sample of 502 young women, 398 (79 percent of the target sample) were followed-up in 2011, at which time we observed their educational outcomes and mental health. 343 of the 398 women interviewed in both 2009 and 2011 were interviewed in 2015 (68 percent of the original target sample), at which time we again observed their mental health. This level of attrition is substantial, although comparable to other longitudinal studies of unrealized expectations in the United States (Reynolds and Baird 2010), and to other panel surveys conducted in Malawi (Kohler et al. 2014). Attrition does not substantially alter the sociodemographic profile of the analytic samples (see Appendix A). Nonetheless, a young woman’s attrition from the study could correspond with her propensity to respond negatively to unrealized educational expectations. More highly educated women are most likely to leave the study (Yeatman et al. 2019); however, migration could be a coping strategy for young women in the wake of unrealized educational expectations. If migration improves mental health, as shown elsewhere (Nauman et al. 2015), young women who are not followed-up may have better mental health. Alternatively, if migration worsens mental health (Warfa et al. 2006), young women who migrate in the wake of unrealized educational expectations may experience compounding risks of mental health, meaning the study results would be conservative estimates of the population average. It is important to keep these possibilities in mind when interpreting results. Measures Mental Health Outcomes Mental health problems are a major contributor to the global disease burden, and they are especially prevalent in low-income regions, including sub-Saharan Africa (Sankoh, Sevalie and Weston 2018; Vigo, Thornicroft and Atun 2016). Even so, there is a general lack of clinical research on mental health. This is especially true in Malawi, where there is a severe shortage of clinicians. Malawi’s entire public health sector reportedly had a single psychiatrist as recently as 2014 (Kim et al. 2014). As a result, few mental health scales have been adapted and locally validated for their psychometric properties. Only a handful of studies have implemented diagnostic tests in Malawi, including the Beck Depression Inventory (Kim et al. 2014) and the Perceived Stress Scale (PSS-4) (DeVylder et al. 2016). Thus, we examined young women’s responses to a series of three questions that address similar mental health symptoms as items on the Beck Depression Inventory, as well as the PSS-4, which has been shown to correspond with symptoms of depression and anxiety (Pereira-Morales, Adan and Forero 2017). In 2009 and 2011, young women reported whether, in the past month, they felt “not really,” “a little,” or “very much” depressed, lonely, or content.5 In 2015, interviewers asked young women similar questions, but they reported whether they “never,” “almost never,” “sometimes,” “fairly often,” or “very often” felt these ways. In 2015, interviewers also administered the PSS-4. Interviewers asked respondents whether, in the past month, they have “very often,” “fairly often,” “sometimes,” “almost never,” or “never” felt unable to control the important things in their life, confident about their ability to handle personal problems, believed things were going their way, or felt difficulties were piling up so high they could not overcome them.6 To obtain PSS-4 scores, we followed standard protocol of reverse coding the positive items (items 2 and 3) and then summing across all four items (Cohen and Williamson 1988; Cohen, Kamarck and Mermelstein 1983). Correlations of the mental health items are generally low, suggesting the indicators approximate different dimensions of young women’s psychological states, and that respondents’ mental health shifts over time. Expectations and Outcomes In 2009, young women answered questions about their educational expectations. Interviewers asked respondents: “Thinking of your life in two years, at age ___, where do you imagine yourself in terms of your education, if all goes as planned?” We distinguished between young women who expected to be in school (at the primary, secondary, or higher level) versus young women who expected to be out of school (including respondents who were unsure or expected to “take a break”). Overall, 90 percent of our sample expected to be in school, attesting to young Malawian women’s high educational expectations. In 2009 and 2011, interviewers asked women, “Are you currently enrolled in school?” Whereas 93 percent of the sample were in school in 2009, only 65 percent were enrolled in 2011.7 We combined respondents’ educational expectations with their school enrollment in 2011 to categorize young women into four groups: (1) expected to be “in school” and in school (reference group), (2) expected to be “in school” but out of school, (3) expected to be “out of school” but in school, and (4) expected to be “out of school” and out of school. In a second set of analyses, we assessed whether experiencing the transition to motherhood—a traditional marker of adulthood—buffers young women against any mental health disadvantage tied to unrealized educational expectations. We disaggregated women who expected to be in school but were out of school according to whether they had become a mother during the two-year period. In total, 56 percent of the young women who failed to realize their expectation to continue in school had a child between 2009 and 2011. Among the reference group of women who remained in school, none had children. This is expected given that pregnancy leads to school expulsion in this context.8 Prior Mental Health Reports, Sociodemographic Characteristics, and Additional Controls Neither expectations nor their achievement is random, but instead both systematically vary across women, which means that studying the consequences of unrealized expectations invariably comes with the challenges of unobserved heterogeneity. We leverage the many rich measures collected as part of TLT in an effort to isolate the impact of unrealized expectations. Beginning with the analyses of young women’s mental health outcomes in 2011, we included the corresponding mental health indicator (i.e., depressed, hopeless, content, lonely) that respondents reported in 2009, at the time they expressed their educational expectations. This accounts for the fact that some young women may be predisposed to mental health problems, which could influence both their education and subsequent mental health (Halpern-Manners et al. 2016). Moreover, given that final educational attainment—not necessarily the experience of one’s educational plans going off course—may be the real driver of mental health problems (Reynolds and Baird 2010), we included an indicator of respondents’ highest educational attainment (primary or secondary level) in 2009 and in 2011. To ensure that unfavorable economic conditions do not drive a spurious link between young women’s unachieved educational expectations and mental health, we controlled for respondents’ differential access to resources using a household asset index derived from a principal component analysis in 2011 (for more detail, see Bachan 2014).9 We also account for whether respondents experienced a food shortage over the study period. We accounted for additional shocks over the period that could influence young women’s educational outcomes and mental health, including whether respondents experienced the death or illness of a close relative (i.e., sibling, parent) or the death of a close friend. We included measures for respondents’ age and if they were married in 2011. We controlled for respondents’ number of siblings. Analytic Approach We start with a descriptive overview of the young women in our sample, including their educational expectations and outcomes, as well as their mental health outcomes in 2011 and 2015. We then demonstrate how respondents differed on key sociodemographic traits according to their expressed educational expectations and the outcomes they achieved. Next, we show the results for whether—among the full sample of young women—educational expectations and outcomes were associated with mental health outcomes in 2011. Because respondents’ reports of feeling depressed, hopeless, content, and lonely in 2011 are scaled (1 to 3), we estimate four ordinal logistic regression models.10 We then report models that assessed the persistence of these results over time and across an additional measure of mental health using data on the 343 young women followed-up in 2015. Because the mental health questions were measured using a five-point Likert scale in 2015, and because the stress measure (PSS-4) is continuously scaled, in these analyses, we estimated ordinary least squared regression models. In the analyses focused on young women’s mental health in 2015, we accounted for the corresponding mental health indicator, educational attainment, age, and number of siblings in 2009. We further controlled for respondents’ educational attainment, household asset index, marital status, and their number of children in 2015. We do not have data on respondents’ access to food or their exposure to bereavement or illness during the inter-survey period from 2011 to 2015, which prevented us from including these covariates in these analyses. Finally, to offer insight into whether any mental health implications of unrealized educational expectations in 2011 and/or 2015 varied according to whether respondents transitioned into motherhood, we report findings from models that disaggregated young women whose expectations went unrealized according to whether they had a child. Although we present only the full model results here, the key findings were stable to the exclusion of covariates. We also performed variance inflation factor (VIF) analyses to confirm that multicollinearity does not bias our results.11 Results Descriptive Findings Table 1 presents the distribution of young Malawian women according to their educational expectations, expressed in 2009, and outcomes observed in 2011. Overall, the majority of young women who expected to be in school were in school (71.7 percent of the 89.7 percent of young women who held this expectation); however, a sizeable proportion—nearly one in three (28.3 percent)—were not. Additional analyses confirm that most young women who discontinued school unexpectedly were consistently out of school as of wave 4, which was just eight months after they expressed their expectation. Of course, even for those in school, enrollment did not always mean they fully achieved their educational expectations: fewer than one in ten of respondents who realized their expectation to be in school were at a lower level of schooling than expected. Supplemental models confirmed the results were stable regardless of how we grouped these young women. Table 1. Distribution of educational expectations (expressed in 2009) and outcomes (observed in 2011) among young women 15 to 19 years old in Balaka, Malawi . Educational outcome in 2011 . . Expectation expressed in 2009 . In school . Out of school . Total . “Enrolled in school” 256 (71.7%) 101 (28.3%) 357 (89.7%) “Out of school” 8 (19.5%) 33 (80.5%) 41 (10.3%) . Educational outcome in 2011 . . Expectation expressed in 2009 . In school . Out of school . Total . “Enrolled in school” 256 (71.7%) 101 (28.3%) 357 (89.7%) “Out of school” 8 (19.5%) 33 (80.5%) 41 (10.3%) Source: Tsogolo La Thanzi; N = 398. Open in new tab Table 1. Distribution of educational expectations (expressed in 2009) and outcomes (observed in 2011) among young women 15 to 19 years old in Balaka, Malawi . Educational outcome in 2011 . . Expectation expressed in 2009 . In school . Out of school . Total . “Enrolled in school” 256 (71.7%) 101 (28.3%) 357 (89.7%) “Out of school” 8 (19.5%) 33 (80.5%) 41 (10.3%) . Educational outcome in 2011 . . Expectation expressed in 2009 . In school . Out of school . Total . “Enrolled in school” 256 (71.7%) 101 (28.3%) 357 (89.7%) “Out of school” 8 (19.5%) 33 (80.5%) 41 (10.3%) Source: Tsogolo La Thanzi; N = 398. Open in new tab Of the minority of respondents who expected to be out of school (10.3 percent), the vast majority were indeed not enrolled (80.5 percent); only 19.5 percent were in school despite not expecting to be. We kept this small group of women (those who were enrolled despite their expectations) in the sample because they also experienced unrealized expectations and offer an interesting comparison. It is important to interpret findings pertaining to these women cautiously, however, given the very small cell sizes. Table 2 provides a descriptive overview of the young women’s characteristics in 2009, 2011, and 2015. In general, women reported not, or only sometimes, feeling depressed and lonely; they more frequently reported feeling content. The distribution of their mental health outcomes was relatively stable over time (note that differences in means stem from the shift from three to five response options between 2011 and 2015). In 2015, the average woman scored 8.9 on the PSS-4, which ranges in value from 4 to 20, with higher values indicating more stress. Table 2. Descriptive statistics for young women 15 to 19 years old in Balaka, Malawi* . Mean (SD)/% . . 2009 . 2011 . •2015 . Mental health outcomes◊• Depressed 1.4 (0.7) 1.2 (0.5) 1.8 (1.2) Lonely 1.1 (0.4) 1.1 (0.3) 1.6 (1.1) Content 2.4 (0.7) 2.5 (0.8) 3.7 (1.3) Perceived Stress Scale (PSS-4) -- -- 8.9 (3.5) Socioeconomic factors Household goods index -- 0.7 (2.7) 0.4 (2.4) Experienced food shortage (2009–2011) -- 22.1 -- Educational attainment  Primary 61.6 34.2 35.4  Secondary/Tertiary 38.4 65.8 64.6 Additional Covariates Experienced close death/illness (2009–2011) -- 73.6 -- Married 0.0 25.1 55.7 Age 16.3 (1.2) -- -- Number of siblings 4.9 (2.6) -- -- Number of children -- -- 0.8 (0.7) . Mean (SD)/% . . 2009 . 2011 . •2015 . Mental health outcomes◊• Depressed 1.4 (0.7) 1.2 (0.5) 1.8 (1.2) Lonely 1.1 (0.4) 1.1 (0.3) 1.6 (1.1) Content 2.4 (0.7) 2.5 (0.8) 3.7 (1.3) Perceived Stress Scale (PSS-4) -- -- 8.9 (3.5) Socioeconomic factors Household goods index -- 0.7 (2.7) 0.4 (2.4) Experienced food shortage (2009–2011) -- 22.1 -- Educational attainment  Primary 61.6 34.2 35.4  Secondary/Tertiary 38.4 65.8 64.6 Additional Covariates Experienced close death/illness (2009–2011) -- 73.6 -- Married 0.0 25.1 55.7 Age 16.3 (1.2) -- -- Number of siblings 4.9 (2.6) -- -- Number of children -- -- 0.8 (0.7) Source: Tsogolo La Thanzi; N = 398. ◊2009 and 2011: 1 = not really; 2 = a little; 3 = very much; •2015:1 = never; 2 = almost never; 3 = sometimes; 4 = fairly often; 5 = very often. *N = 398 in 2009 and 2011; N = 343 in 2015. Open in new tab Table 2. Descriptive statistics for young women 15 to 19 years old in Balaka, Malawi* . Mean (SD)/% . . 2009 . 2011 . •2015 . Mental health outcomes◊• Depressed 1.4 (0.7) 1.2 (0.5) 1.8 (1.2) Lonely 1.1 (0.4) 1.1 (0.3) 1.6 (1.1) Content 2.4 (0.7) 2.5 (0.8) 3.7 (1.3) Perceived Stress Scale (PSS-4) -- -- 8.9 (3.5) Socioeconomic factors Household goods index -- 0.7 (2.7) 0.4 (2.4) Experienced food shortage (2009–2011) -- 22.1 -- Educational attainment  Primary 61.6 34.2 35.4  Secondary/Tertiary 38.4 65.8 64.6 Additional Covariates Experienced close death/illness (2009–2011) -- 73.6 -- Married 0.0 25.1 55.7 Age 16.3 (1.2) -- -- Number of siblings 4.9 (2.6) -- -- Number of children -- -- 0.8 (0.7) . Mean (SD)/% . . 2009 . 2011 . •2015 . Mental health outcomes◊• Depressed 1.4 (0.7) 1.2 (0.5) 1.8 (1.2) Lonely 1.1 (0.4) 1.1 (0.3) 1.6 (1.1) Content 2.4 (0.7) 2.5 (0.8) 3.7 (1.3) Perceived Stress Scale (PSS-4) -- -- 8.9 (3.5) Socioeconomic factors Household goods index -- 0.7 (2.7) 0.4 (2.4) Experienced food shortage (2009–2011) -- 22.1 -- Educational attainment  Primary 61.6 34.2 35.4  Secondary/Tertiary 38.4 65.8 64.6 Additional Covariates Experienced close death/illness (2009–2011) -- 73.6 -- Married 0.0 25.1 55.7 Age 16.3 (1.2) -- -- Number of siblings 4.9 (2.6) -- -- Number of children -- -- 0.8 (0.7) Source: Tsogolo La Thanzi; N = 398. ◊2009 and 2011: 1 = not really; 2 = a little; 3 = very much; •2015:1 = never; 2 = almost never; 3 = sometimes; 4 = fairly often; 5 = very often. *N = 398 in 2009 and 2011; N = 343 in 2015. Open in new tab One in five young women experienced a food shortage between 2009 and 2011. In 2009, 38.4 percent of respondents had attended secondary school; this increases to 65.8 percent by 2011. Note that only 64.6 percent of the subsample of young women followed-up in 2015 had attended secondary school, suggesting that respondents with more education were slightly less likely to be followed-up. One in four young women were married in 2011, and this increases to 55.7 percent by 2015. Table 3 shows how respondents differ according to their educational expectations and outcomes. For the purposes of this comparison, we combined women who expected to be out of school regardless of whether they did (N = 33) or did not (N = 8) achieve this expectation. Comparing the first and second columns, young women who expressed an expectation to be “out of school” in two years, regardless of whether they realized the expectation, were more likely to have attended secondary school compared to respondents who expected to continue further. These respondents also lived in households with lower asset index scores, and they were more likely to have married and/or become a mother by 2011, with the majority having married/had a child by 2015. Young women who realized their expectation to be in school in 2011 lived in households with higher asset index scores relative to their peers who did not realize their expectation to be in school. Additionally, none of these young women had children. As noted earlier, just over one-half of women who did not continue school as expected became a mother by 2011, with more than two-thirds doing so by 2015. Table 3. Comparison of young women 15 to 19 years old in Balaka, Malawi, by their educational expectations (expressed in 2009) and outcomes (observed in 2011) in 2009, 2011, and 2015 . Expect to be “out of school” (2009) . Expect to be “in school” (2009) . . . Total . In school (2011) . Out of school (2011) . 2009 ◊ Age 16.9 (1.3) 16.0 (1.2) 15.9 (1.1) 16.4 (1.3) % Secondary/tertiary education 48.8 37.3 34.8 43.6 2011 ◊ % Secondary/tertiary education 53.7 67.2 72.3 54.5 Household goods index −0.1 (2.5) 0.8 (2.7) 1.0 (2.8) 0.1 (2.4) Married 48.8 22.4 7.0 61.4 Ever experienced food shortage* 24.4 21.9 22.7 19.8 Experienced close death/illness* 70.7 73.9 74.6 72.3 % became mother 53.7 15.9 0.00 56.4 2015● % Secondary/tertiary education 55.9 65.6 74.7 42.7 Household goods index −0.4 (2.0) 0.1 (2.4) 0.6 (2.5) −0.4 (2.1) Married 70.6 54.2 48.7 67.4 % became mother 79.4 63.0 53.6 86.5 . Expect to be “out of school” (2009) . Expect to be “in school” (2009) . . . Total . In school (2011) . Out of school (2011) . 2009 ◊ Age 16.9 (1.3) 16.0 (1.2) 15.9 (1.1) 16.4 (1.3) % Secondary/tertiary education 48.8 37.3 34.8 43.6 2011 ◊ % Secondary/tertiary education 53.7 67.2 72.3 54.5 Household goods index −0.1 (2.5) 0.8 (2.7) 1.0 (2.8) 0.1 (2.4) Married 48.8 22.4 7.0 61.4 Ever experienced food shortage* 24.4 21.9 22.7 19.8 Experienced close death/illness* 70.7 73.9 74.6 72.3 % became mother 53.7 15.9 0.00 56.4 2015● % Secondary/tertiary education 55.9 65.6 74.7 42.7 Household goods index −0.4 (2.0) 0.1 (2.4) 0.6 (2.5) −0.4 (2.1) Married 70.6 54.2 48.7 67.4 % became mother 79.4 63.0 53.6 86.5 Source: Tsogolo la Thanzi; ◊N = 398; ●N = 343 *between 2009 and 2011. Open in new tab Table 3. Comparison of young women 15 to 19 years old in Balaka, Malawi, by their educational expectations (expressed in 2009) and outcomes (observed in 2011) in 2009, 2011, and 2015 . Expect to be “out of school” (2009) . Expect to be “in school” (2009) . . . Total . In school (2011) . Out of school (2011) . 2009 ◊ Age 16.9 (1.3) 16.0 (1.2) 15.9 (1.1) 16.4 (1.3) % Secondary/tertiary education 48.8 37.3 34.8 43.6 2011 ◊ % Secondary/tertiary education 53.7 67.2 72.3 54.5 Household goods index −0.1 (2.5) 0.8 (2.7) 1.0 (2.8) 0.1 (2.4) Married 48.8 22.4 7.0 61.4 Ever experienced food shortage* 24.4 21.9 22.7 19.8 Experienced close death/illness* 70.7 73.9 74.6 72.3 % became mother 53.7 15.9 0.00 56.4 2015● % Secondary/tertiary education 55.9 65.6 74.7 42.7 Household goods index −0.4 (2.0) 0.1 (2.4) 0.6 (2.5) −0.4 (2.1) Married 70.6 54.2 48.7 67.4 % became mother 79.4 63.0 53.6 86.5 . Expect to be “out of school” (2009) . Expect to be “in school” (2009) . . . Total . In school (2011) . Out of school (2011) . 2009 ◊ Age 16.9 (1.3) 16.0 (1.2) 15.9 (1.1) 16.4 (1.3) % Secondary/tertiary education 48.8 37.3 34.8 43.6 2011 ◊ % Secondary/tertiary education 53.7 67.2 72.3 54.5 Household goods index −0.1 (2.5) 0.8 (2.7) 1.0 (2.8) 0.1 (2.4) Married 48.8 22.4 7.0 61.4 Ever experienced food shortage* 24.4 21.9 22.7 19.8 Experienced close death/illness* 70.7 73.9 74.6 72.3 % became mother 53.7 15.9 0.00 56.4 2015● % Secondary/tertiary education 55.9 65.6 74.7 42.7 Household goods index −0.4 (2.0) 0.1 (2.4) 0.6 (2.5) −0.4 (2.1) Married 70.6 54.2 48.7 67.4 % became mother 79.4 63.0 53.6 86.5 Source: Tsogolo la Thanzi; ◊N = 398; ●N = 343 *between 2009 and 2011. Open in new tab Unrealized Educational Expectations and Mental Health Outcomes Table 4 reports coefficients from ordinal logistic regression models that estimate whether young women’s mental health outcomes in 2011 differed according to their educational expectations and outcomes. In each model, young women who realized their expectation to be in school are the reference group. As model 1 in Table 4 shows, young women who did not realize their expectation to be in school experienced significantly more intensive feelings of depression (odds ratio = 2.68). These young women were also lonelier relative to their peers who realized their educational expectations (model 2).12 Finally, young women who did not realize their expectation to be in school felt less content (model 3). Indeed, they have 52 percent lower odds of feeling content relative to their peers who realized their expectation to be in school. Table 4. Results from ordinal logistic regressions of mental health outcomes in 2011 on educational expectations and outcomes among young women 15 to 19 years old in Balaka, Malawi . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . Educational expectations and outcomes “Enrolled in school”, in school (reference) -- -- -- -- -- -- “Enrolled in school”, but out of school 0.99 (0.43) 2.68* 1.44 (0.64) 4.20* −0.74 (0.30) 0.48* “Out of school”, but in school 1.75 (0.78) 5.74* 1.45 (1.21) 4.26 −0.69 (0.77) 0.50 “Out of school”, out of school −1.09 (1.08) 0.34 -- −0.33 (0.46) 0.72 Corresponding mental health indicator (2009) 0.32 (0.20) 1.38 1.39 (0.44) 4.01 ** 0.19 (0.16) 1.21 Socioeconomic factors Educational attainment (2009) Primary -- -- -- -- -- -- Secondary/tertiary education −0.41 (0.43) 0.67 0.20 (0.82) 1.22 0.34 (0.30) 1.41 Experienced food shortage (2009–2011) 0.88 (0.35) 2.40* 0.66 (0.58) 1.94 −0.57 (0.25) 0.57* Household asset index (2011) 0.06 (0.06) 1.07 −0.07 (0.11) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011) Primary -- -- -- -- -- -- Secondary/tertiary education 0.87 (0.44) 2.38* −0.58 (0.79) 0.56 −0.11 (0.28) 0.90 Life Shocks Experienced death/illness (2009–2011) 1.12 (0.46) 3.06* 0.16 (0.62) 1.17 −0.56 (0.26) 0.57* Married (2011) −0.21 (0.44) 0.81 −0.82 (0.64) 0.44 0.21 (0.30) 1.24 Additional Covariates Age 0.01 (0.15) 1.01 0.57 (0.25) 1.77* 0.17 (0.11) 1.19 Number of siblings 0.05 (0.07) 1.05 0.16 (0.11) 1.17 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 32.75** 31.77*** 24.15* . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . Educational expectations and outcomes “Enrolled in school”, in school (reference) -- -- -- -- -- -- “Enrolled in school”, but out of school 0.99 (0.43) 2.68* 1.44 (0.64) 4.20* −0.74 (0.30) 0.48* “Out of school”, but in school 1.75 (0.78) 5.74* 1.45 (1.21) 4.26 −0.69 (0.77) 0.50 “Out of school”, out of school −1.09 (1.08) 0.34 -- −0.33 (0.46) 0.72 Corresponding mental health indicator (2009) 0.32 (0.20) 1.38 1.39 (0.44) 4.01 ** 0.19 (0.16) 1.21 Socioeconomic factors Educational attainment (2009) Primary -- -- -- -- -- -- Secondary/tertiary education −0.41 (0.43) 0.67 0.20 (0.82) 1.22 0.34 (0.30) 1.41 Experienced food shortage (2009–2011) 0.88 (0.35) 2.40* 0.66 (0.58) 1.94 −0.57 (0.25) 0.57* Household asset index (2011) 0.06 (0.06) 1.07 −0.07 (0.11) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011) Primary -- -- -- -- -- -- Secondary/tertiary education 0.87 (0.44) 2.38* −0.58 (0.79) 0.56 −0.11 (0.28) 0.90 Life Shocks Experienced death/illness (2009–2011) 1.12 (0.46) 3.06* 0.16 (0.62) 1.17 −0.56 (0.26) 0.57* Married (2011) −0.21 (0.44) 0.81 −0.82 (0.64) 0.44 0.21 (0.30) 1.24 Additional Covariates Age 0.01 (0.15) 1.01 0.57 (0.25) 1.77* 0.17 (0.11) 1.19 Number of siblings 0.05 (0.07) 1.05 0.16 (0.11) 1.17 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 32.75** 31.77*** 24.15* Source: Tsogolo La Thanzi; †p < 0.1; *p < 0.05; **p < 0.01 N = 398. Open in new tab Table 4. Results from ordinal logistic regressions of mental health outcomes in 2011 on educational expectations and outcomes among young women 15 to 19 years old in Balaka, Malawi . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . Educational expectations and outcomes “Enrolled in school”, in school (reference) -- -- -- -- -- -- “Enrolled in school”, but out of school 0.99 (0.43) 2.68* 1.44 (0.64) 4.20* −0.74 (0.30) 0.48* “Out of school”, but in school 1.75 (0.78) 5.74* 1.45 (1.21) 4.26 −0.69 (0.77) 0.50 “Out of school”, out of school −1.09 (1.08) 0.34 -- −0.33 (0.46) 0.72 Corresponding mental health indicator (2009) 0.32 (0.20) 1.38 1.39 (0.44) 4.01 ** 0.19 (0.16) 1.21 Socioeconomic factors Educational attainment (2009) Primary -- -- -- -- -- -- Secondary/tertiary education −0.41 (0.43) 0.67 0.20 (0.82) 1.22 0.34 (0.30) 1.41 Experienced food shortage (2009–2011) 0.88 (0.35) 2.40* 0.66 (0.58) 1.94 −0.57 (0.25) 0.57* Household asset index (2011) 0.06 (0.06) 1.07 −0.07 (0.11) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011) Primary -- -- -- -- -- -- Secondary/tertiary education 0.87 (0.44) 2.38* −0.58 (0.79) 0.56 −0.11 (0.28) 0.90 Life Shocks Experienced death/illness (2009–2011) 1.12 (0.46) 3.06* 0.16 (0.62) 1.17 −0.56 (0.26) 0.57* Married (2011) −0.21 (0.44) 0.81 −0.82 (0.64) 0.44 0.21 (0.30) 1.24 Additional Covariates Age 0.01 (0.15) 1.01 0.57 (0.25) 1.77* 0.17 (0.11) 1.19 Number of siblings 0.05 (0.07) 1.05 0.16 (0.11) 1.17 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 32.75** 31.77*** 24.15* . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . . Coeff . SE . Odds Ratio . Educational expectations and outcomes “Enrolled in school”, in school (reference) -- -- -- -- -- -- “Enrolled in school”, but out of school 0.99 (0.43) 2.68* 1.44 (0.64) 4.20* −0.74 (0.30) 0.48* “Out of school”, but in school 1.75 (0.78) 5.74* 1.45 (1.21) 4.26 −0.69 (0.77) 0.50 “Out of school”, out of school −1.09 (1.08) 0.34 -- −0.33 (0.46) 0.72 Corresponding mental health indicator (2009) 0.32 (0.20) 1.38 1.39 (0.44) 4.01 ** 0.19 (0.16) 1.21 Socioeconomic factors Educational attainment (2009) Primary -- -- -- -- -- -- Secondary/tertiary education −0.41 (0.43) 0.67 0.20 (0.82) 1.22 0.34 (0.30) 1.41 Experienced food shortage (2009–2011) 0.88 (0.35) 2.40* 0.66 (0.58) 1.94 −0.57 (0.25) 0.57* Household asset index (2011) 0.06 (0.06) 1.07 −0.07 (0.11) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011) Primary -- -- -- -- -- -- Secondary/tertiary education 0.87 (0.44) 2.38* −0.58 (0.79) 0.56 −0.11 (0.28) 0.90 Life Shocks Experienced death/illness (2009–2011) 1.12 (0.46) 3.06* 0.16 (0.62) 1.17 −0.56 (0.26) 0.57* Married (2011) −0.21 (0.44) 0.81 −0.82 (0.64) 0.44 0.21 (0.30) 1.24 Additional Covariates Age 0.01 (0.15) 1.01 0.57 (0.25) 1.77* 0.17 (0.11) 1.19 Number of siblings 0.05 (0.07) 1.05 0.16 (0.11) 1.17 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 32.75** 31.77*** 24.15* Source: Tsogolo La Thanzi; †p < 0.1; *p < 0.05; **p < 0.01 N = 398. Open in new tab Past work suggests lower educational attainment may drive a link between unfulfilled educational plans and poor mental health. However, unrealized educational expectations correspond with mental health disadvantage even after controlling for educational attainment. Moreover, additional analyses confirm that young women who expected to be out of school (and were not enrolled) felt less depressed (in 2011) relative to their peers who were also out of school but did not expect to be (p < 0.1). The very small number of women in this group requires we interpret these findings with caution; however, these results suggest that unrealized educational plans specifically, not just lower attainment or out-of-school status, are a precursor to mental health disadvantage. Models 1–3 in Table 4 further show that additional variables are associated with young women’s mental health outcomes. For instance, young women who experienced a food shortage over the study period were more depressed (model 1) and less content (model 3) relative to their peers. The death/illness of a close friend or relative also corresponds with young women feeling more depressed (model 1) and less content (model 3). To assess the persistence of results over time and across an additional measure of mental health, Table 5 shows results for the 343 young women whom TLT interviewed in 2015. In these analyses, we used ordinary least squared regression models because the mental health questions were measured using a five-point Likert scale in 2015, and because of the continuously scaled stress measure (PSS-4). As shown in Table 5, young women who expected to be in school in 2011 but were not enrolled experienced worse mental health four years later relative to their peers who realized their expectation to stay in school, including more frequent feelings of depression (model 1), less contentment (model 3), and higher stress scores (model 4). Additional analyses confirm that young women who expected to be out of school (and were not enrolled) were less frequently lonely (2015) relative to their peers who were also out of school but did not expect to be (p < 0.1). The very small number of women in this group requires us to interpret these findings with caution; however, the results lend support to the conclusion that unrealized educational plans specifically, not just lower attainment or out-of-school status, is a precursor to loneliness. The results additionally show that young women residing in better-resourced households reported significantly lower frequency of loneliness, more contentment, and lower scores on the stress measure. Table 5. Results from ordinary least squares regression models of mental health outcomes in 2015 on educational expectations and outcomes among young women 15 to 19 years old in Balaka, Malawi . (1) Depressed . . (2) Lonely . . (3) Content . . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . “Enrolled in school”, in school (reference) -- -- -- -- “Enrolled in school”, but out of school 0.38 (0.18)* 0.13 (0.16) −0.53 (0.19)** 1.10 (0.50)* “Out of school”, but in school 0.58 (0.46) 0.58 (0.41) −0.56 (0.49) 1.74 (1.32) “Out of school”, out of school 0.15 (0.27) −0.28 (0.24) −0.07 (0.29) 1.00 (0.77) Corresponding mental health indicator (2009) 0.44 (0.09)*** 0.01 (0.01) 0.07 (0.10) -- Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- --  Secondary/tertiary education 0.08 (0.19) 0.11 (0.17) −0.03 (0.20) −0.30 (0.53) Experienced food shortage (2009–2011) 0.08 (0.16) −0.13 (0.14) −0.21 (0.17) 0.85 (0.46)† Household goods index (2015) 0.00 (0.03) −0.08 (0.03)** 0.09 (0.04)* −0.45 (0.10)*** Educational attainment (2015)  Primary -- -- -- -- --  Secondary/tertiary education −0.09 (0.18) −0.11 (0.16) −0.08 (0.19) −0.51 (0.52) Life Shocks Experienced death/illness (2009–2011) 0.22 (0.15) 0.14 (0.13) −0.08 (0.15) 0.38 (0.42) Married (2015) −0.14 (0.16) −0.14 (0.15) 0.18 (0.17) −0.46 (0.46) Additional Covariates Age 0.00 (0.07) 0.16 (0.06)* −0.01 (0.07) 0.26 (0.20) Number of siblings 0.03 (0.03) 0.03 (0.02) 0.03 (0.03) −0.03 (0.08) Number of children −0.22 (0.12)† −0.03 (0.11) 0.16 (0.13) −0.74 (0.35)* Model fit (f-statistic) 3.08*** 1.97* 1.80* 4.34*** . (1) Depressed . . (2) Lonely . . (3) Content . . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . “Enrolled in school”, in school (reference) -- -- -- -- “Enrolled in school”, but out of school 0.38 (0.18)* 0.13 (0.16) −0.53 (0.19)** 1.10 (0.50)* “Out of school”, but in school 0.58 (0.46) 0.58 (0.41) −0.56 (0.49) 1.74 (1.32) “Out of school”, out of school 0.15 (0.27) −0.28 (0.24) −0.07 (0.29) 1.00 (0.77) Corresponding mental health indicator (2009) 0.44 (0.09)*** 0.01 (0.01) 0.07 (0.10) -- Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- --  Secondary/tertiary education 0.08 (0.19) 0.11 (0.17) −0.03 (0.20) −0.30 (0.53) Experienced food shortage (2009–2011) 0.08 (0.16) −0.13 (0.14) −0.21 (0.17) 0.85 (0.46)† Household goods index (2015) 0.00 (0.03) −0.08 (0.03)** 0.09 (0.04)* −0.45 (0.10)*** Educational attainment (2015)  Primary -- -- -- -- --  Secondary/tertiary education −0.09 (0.18) −0.11 (0.16) −0.08 (0.19) −0.51 (0.52) Life Shocks Experienced death/illness (2009–2011) 0.22 (0.15) 0.14 (0.13) −0.08 (0.15) 0.38 (0.42) Married (2015) −0.14 (0.16) −0.14 (0.15) 0.18 (0.17) −0.46 (0.46) Additional Covariates Age 0.00 (0.07) 0.16 (0.06)* −0.01 (0.07) 0.26 (0.20) Number of siblings 0.03 (0.03) 0.03 (0.02) 0.03 (0.03) −0.03 (0.08) Number of children −0.22 (0.12)† −0.03 (0.11) 0.16 (0.13) −0.74 (0.35)* Model fit (f-statistic) 3.08*** 1.97* 1.80* 4.34*** Source: Tsogolo la Thanzi 1 (2009–2011) & Tsogolo la Thanzi 2 (2015); †p < 0.1; *p < 0.05; **p < 0.01; N = 343. Models 1–3 outcomes coded as 1 “never,” 2 “almost never,” 3 “sometimes,” 4 “fairly often,” and 5 “very often.” Model 4 outcome ranges from 4 (lowest stress score) to 20 (highest stress score). ○Perceived Stress Scale-4. Open in new tab Table 5. Results from ordinary least squares regression models of mental health outcomes in 2015 on educational expectations and outcomes among young women 15 to 19 years old in Balaka, Malawi . (1) Depressed . . (2) Lonely . . (3) Content . . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . “Enrolled in school”, in school (reference) -- -- -- -- “Enrolled in school”, but out of school 0.38 (0.18)* 0.13 (0.16) −0.53 (0.19)** 1.10 (0.50)* “Out of school”, but in school 0.58 (0.46) 0.58 (0.41) −0.56 (0.49) 1.74 (1.32) “Out of school”, out of school 0.15 (0.27) −0.28 (0.24) −0.07 (0.29) 1.00 (0.77) Corresponding mental health indicator (2009) 0.44 (0.09)*** 0.01 (0.01) 0.07 (0.10) -- Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- --  Secondary/tertiary education 0.08 (0.19) 0.11 (0.17) −0.03 (0.20) −0.30 (0.53) Experienced food shortage (2009–2011) 0.08 (0.16) −0.13 (0.14) −0.21 (0.17) 0.85 (0.46)† Household goods index (2015) 0.00 (0.03) −0.08 (0.03)** 0.09 (0.04)* −0.45 (0.10)*** Educational attainment (2015)  Primary -- -- -- -- --  Secondary/tertiary education −0.09 (0.18) −0.11 (0.16) −0.08 (0.19) −0.51 (0.52) Life Shocks Experienced death/illness (2009–2011) 0.22 (0.15) 0.14 (0.13) −0.08 (0.15) 0.38 (0.42) Married (2015) −0.14 (0.16) −0.14 (0.15) 0.18 (0.17) −0.46 (0.46) Additional Covariates Age 0.00 (0.07) 0.16 (0.06)* −0.01 (0.07) 0.26 (0.20) Number of siblings 0.03 (0.03) 0.03 (0.02) 0.03 (0.03) −0.03 (0.08) Number of children −0.22 (0.12)† −0.03 (0.11) 0.16 (0.13) −0.74 (0.35)* Model fit (f-statistic) 3.08*** 1.97* 1.80* 4.34*** . (1) Depressed . . (2) Lonely . . (3) Content . . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . “Enrolled in school”, in school (reference) -- -- -- -- “Enrolled in school”, but out of school 0.38 (0.18)* 0.13 (0.16) −0.53 (0.19)** 1.10 (0.50)* “Out of school”, but in school 0.58 (0.46) 0.58 (0.41) −0.56 (0.49) 1.74 (1.32) “Out of school”, out of school 0.15 (0.27) −0.28 (0.24) −0.07 (0.29) 1.00 (0.77) Corresponding mental health indicator (2009) 0.44 (0.09)*** 0.01 (0.01) 0.07 (0.10) -- Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- --  Secondary/tertiary education 0.08 (0.19) 0.11 (0.17) −0.03 (0.20) −0.30 (0.53) Experienced food shortage (2009–2011) 0.08 (0.16) −0.13 (0.14) −0.21 (0.17) 0.85 (0.46)† Household goods index (2015) 0.00 (0.03) −0.08 (0.03)** 0.09 (0.04)* −0.45 (0.10)*** Educational attainment (2015)  Primary -- -- -- -- --  Secondary/tertiary education −0.09 (0.18) −0.11 (0.16) −0.08 (0.19) −0.51 (0.52) Life Shocks Experienced death/illness (2009–2011) 0.22 (0.15) 0.14 (0.13) −0.08 (0.15) 0.38 (0.42) Married (2015) −0.14 (0.16) −0.14 (0.15) 0.18 (0.17) −0.46 (0.46) Additional Covariates Age 0.00 (0.07) 0.16 (0.06)* −0.01 (0.07) 0.26 (0.20) Number of siblings 0.03 (0.03) 0.03 (0.02) 0.03 (0.03) −0.03 (0.08) Number of children −0.22 (0.12)† −0.03 (0.11) 0.16 (0.13) −0.74 (0.35)* Model fit (f-statistic) 3.08*** 1.97* 1.80* 4.34*** Source: Tsogolo la Thanzi 1 (2009–2011) & Tsogolo la Thanzi 2 (2015); †p < 0.1; *p < 0.05; **p < 0.01; N = 343. Models 1–3 outcomes coded as 1 “never,” 2 “almost never,” 3 “sometimes,” 4 “fairly often,” and 5 “very often.” Model 4 outcome ranges from 4 (lowest stress score) to 20 (highest stress score). ○Perceived Stress Scale-4. Open in new tab To summarize, young women whose expectations went unrealized experienced elevated susceptibility to feelings of depression and less contentment both in the immediate wake of their plans going unrealized and multiple years later. Additionally, they are more susceptible to loneliness soon after their expectations went unrealized, and they scored higher on the stress measure multiple years later. Unrealized Educational Expectations and Mental Health Outcomes: Variable According to the Transition to Motherhood? In the next set of analyses, we assessed whether becoming a mother buffered young women from elevated feelings of depression, loneliness, discontentment, and stress tied to their unrealized educational expectations. The results in Table 6 show that the negative mental health outcomes in the immediate wake of unrealized educational expectations were almost entirely concentrated among young women who were out of school but did not become mothers. Young women who discontinued school unexpectedly, and did not adopt a new role as mother, were more depressed (model 1), lonelier (model 2), and less content relative to their childless peers who were in school as expected. Conversely, young women whose plans to continue school went unachieved, but who assumed a new role as mother over the same time horizon, had comparable mental health reports as their peers who realized their educational plans to continue school, aside from a marginally significant increase in loneliness (p < 0.1). Alternating the reference group to women who did not transition to motherhood confirms that their peers who became mothers were less depressed (not shown; p < 0.05). These findings suggest that achieving a traditional marker of adulthood can buffer young women from negative mental health outcomes in the immediate wake of unrealized educational expectations. Table 6. Results from ordinal logistic regressions of mental health outcomes in 2011 on educational and childbearing experiences among subsample of young women 15 to 19 years old in Balaka, Malawi . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . In school and childless -- -- -- -- -- -- Out of school and childless 1.58 (0.50) 4.84** 1.52 (0.72) 4.59* −0.87 (0.37) 0.42* Out of school and had baby 0.57 (0.57) 1.77 1.31 (0.76) 3.69† −0.53 (0.38) 0.59 Corresponding mental health indicator (2009) 0.36 (0.21) 1.44† 1.34 (0.44) 3.82** 0.18 (0.16) 1.20 Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- -- --  Secondary/tertiary −0.41 (0.44) 0.66 0.26 (0.82) 1.30 0.27 (0.32) 1.31 Experienced food shortage (2009–2011) 0.77 (0.38) 2.17* 0.76 (0.59) 2.13 −0.58 (0.27) 0.56* Household asset index (2011) 0.04 (0.07) 1.04 −0.07 (0.12) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011)  Primary -- -- -- -- -- --  Secondary/Tertiary 1.16 (0.47) 3.20* −0.54 (0.79) 0.58 −0.02 (0.30) 0.98 Life shocks Experienced death/illness (2009–2011) 1.72 (0.62) 5.56** 0.04 (0.63) 1.04 −0.71 (0.28) 0.49* Married (2011) −0.28 (0.49) 0.76 −0.72 (0.66) 0.49 0.14 (0.34) 1.15 Additional covariates Age −0.02 (0.17) 0.98 0.49 (0.26) 1.63 † 0.19 (0.12) 1.21 Number of siblings 0.05 (0.07) 1.06 0.14 (0.11) 1.16 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 34.51*** 28.81*** 24.00* . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . In school and childless -- -- -- -- -- -- Out of school and childless 1.58 (0.50) 4.84** 1.52 (0.72) 4.59* −0.87 (0.37) 0.42* Out of school and had baby 0.57 (0.57) 1.77 1.31 (0.76) 3.69† −0.53 (0.38) 0.59 Corresponding mental health indicator (2009) 0.36 (0.21) 1.44† 1.34 (0.44) 3.82** 0.18 (0.16) 1.20 Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- -- --  Secondary/tertiary −0.41 (0.44) 0.66 0.26 (0.82) 1.30 0.27 (0.32) 1.31 Experienced food shortage (2009–2011) 0.77 (0.38) 2.17* 0.76 (0.59) 2.13 −0.58 (0.27) 0.56* Household asset index (2011) 0.04 (0.07) 1.04 −0.07 (0.12) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011)  Primary -- -- -- -- -- --  Secondary/Tertiary 1.16 (0.47) 3.20* −0.54 (0.79) 0.58 −0.02 (0.30) 0.98 Life shocks Experienced death/illness (2009–2011) 1.72 (0.62) 5.56** 0.04 (0.63) 1.04 −0.71 (0.28) 0.49* Married (2011) −0.28 (0.49) 0.76 −0.72 (0.66) 0.49 0.14 (0.34) 1.15 Additional covariates Age −0.02 (0.17) 0.98 0.49 (0.26) 1.63 † 0.19 (0.12) 1.21 Number of siblings 0.05 (0.07) 1.06 0.14 (0.11) 1.16 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 34.51*** 28.81*** 24.00* Source: Tsogolo la Thanzi; †p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001; N = 356. Open in new tab Table 6. Results from ordinal logistic regressions of mental health outcomes in 2011 on educational and childbearing experiences among subsample of young women 15 to 19 years old in Balaka, Malawi . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . In school and childless -- -- -- -- -- -- Out of school and childless 1.58 (0.50) 4.84** 1.52 (0.72) 4.59* −0.87 (0.37) 0.42* Out of school and had baby 0.57 (0.57) 1.77 1.31 (0.76) 3.69† −0.53 (0.38) 0.59 Corresponding mental health indicator (2009) 0.36 (0.21) 1.44† 1.34 (0.44) 3.82** 0.18 (0.16) 1.20 Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- -- --  Secondary/tertiary −0.41 (0.44) 0.66 0.26 (0.82) 1.30 0.27 (0.32) 1.31 Experienced food shortage (2009–2011) 0.77 (0.38) 2.17* 0.76 (0.59) 2.13 −0.58 (0.27) 0.56* Household asset index (2011) 0.04 (0.07) 1.04 −0.07 (0.12) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011)  Primary -- -- -- -- -- --  Secondary/Tertiary 1.16 (0.47) 3.20* −0.54 (0.79) 0.58 −0.02 (0.30) 0.98 Life shocks Experienced death/illness (2009–2011) 1.72 (0.62) 5.56** 0.04 (0.63) 1.04 −0.71 (0.28) 0.49* Married (2011) −0.28 (0.49) 0.76 −0.72 (0.66) 0.49 0.14 (0.34) 1.15 Additional covariates Age −0.02 (0.17) 0.98 0.49 (0.26) 1.63 † 0.19 (0.12) 1.21 Number of siblings 0.05 (0.07) 1.06 0.14 (0.11) 1.16 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 34.51*** 28.81*** 24.00* . (1) Depressed . (2) Lonely . (3) Content . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . . Coeff . SE . Odds ratio . In school and childless -- -- -- -- -- -- Out of school and childless 1.58 (0.50) 4.84** 1.52 (0.72) 4.59* −0.87 (0.37) 0.42* Out of school and had baby 0.57 (0.57) 1.77 1.31 (0.76) 3.69† −0.53 (0.38) 0.59 Corresponding mental health indicator (2009) 0.36 (0.21) 1.44† 1.34 (0.44) 3.82** 0.18 (0.16) 1.20 Socioeconomic factors Educational attainment (2009)  Primary -- -- -- -- -- --  Secondary/tertiary −0.41 (0.44) 0.66 0.26 (0.82) 1.30 0.27 (0.32) 1.31 Experienced food shortage (2009–2011) 0.77 (0.38) 2.17* 0.76 (0.59) 2.13 −0.58 (0.27) 0.56* Household asset index (2011) 0.04 (0.07) 1.04 −0.07 (0.12) 0.93 −0.05 (0.05) 0.95 Educational attainment (2011)  Primary -- -- -- -- -- --  Secondary/Tertiary 1.16 (0.47) 3.20* −0.54 (0.79) 0.58 −0.02 (0.30) 0.98 Life shocks Experienced death/illness (2009–2011) 1.72 (0.62) 5.56** 0.04 (0.63) 1.04 −0.71 (0.28) 0.49* Married (2011) −0.28 (0.49) 0.76 −0.72 (0.66) 0.49 0.14 (0.34) 1.15 Additional covariates Age −0.02 (0.17) 0.98 0.49 (0.26) 1.63 † 0.19 (0.12) 1.21 Number of siblings 0.05 (0.07) 1.06 0.14 (0.11) 1.16 0.03 (0.05) 1.03 Model fit (likelihood ratio; chi-square) 34.51*** 28.81*** 24.00* Source: Tsogolo la Thanzi; †p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001; N = 356. Open in new tab Table 7 shows that the young women who became mothers as their educational plans went unrealized remained buffered from most mental health disadvantages, aside from elevated depression four years later. That is, the young women who became mothers around the same time their educational plans went off course between four and six years earlier were more depressed relative to their peers who continued school as expected (model 1), and relative to their peers who remained childless even as they did not continue school as expected (as of 2011). Yet, young women who remained childless in the wake of their educational plans going unrealized experienced persistent feelings of discontentment and elevated stress (relative to peers who achieved their expectation to continue school). Additionally analyses confirmed that having a child in the wake of unrealized educational plans corresponds with less loneliness than not having a child (not shown; p < .1). Table 7. Results from ordinary least squares regression models of mental health outcomes in 2015 on educational expectations and outcomes among the subsample of young women 15 to 19 years old in Balaka, Malawi, disaggregated by motherhood status . (1) Depressed . (2) Lonely . (3) Content . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . In school, childless (reference) -- -- -- -- -- Out of school, childless 0.16 (0.23) 0.25 (0.21) −0.73 (0.24)*** 1.12 (0.64)† Out of school, had child 0.70 (0.23)** −0.14 (0.21) −0.26 (0.24) 0.99 (0.64) Corresponding mental health indicator (2009) 0.44 (0.10)*** 0.01 (0.01) 0.03 (0.11) -- Socioeconomic factors Educational attainment (2009) Primary (reference) -- -- -- -- -- Secondary/tertiary education 0.06 (0.20) 0.09 (0.18) −0.21 (0.21) 0.04 (0.55) Experienced food shortage (2009–2011) 0.17 (0.17) −0.02 (0.16) −0.23 (0.19) 1.18 (0.49)* Household goods index (2015) 0.01 (0.04) −0.07 (0.03)* 0.10 (0.04)* −0.48 (0.10)*** Educational attainment (2015)  Primary (reference) -- -- -- -- --  Secondary/tertiary education −0.07 (0.19) −0.10 (0.18) −0.07 (0.20) −0.43 (0.54) Life Shocks Experienced death/illness (2009–2011) 0.24 (0.16) 0.08 (0.14) −0.06 (0.17) 0.46 (0.44) Married (2015) −0.07 (0.17) −0.16 (0.16) 0.19 (0.18) −0.60 (0.49) Additional covariates Age 0.03 (0.08) 0.18 (0.07)* 0.06 (0.08) 0.10 (0.22) Number of siblings 0.03 (0.03) 0.03 (0.03) 0.02 (0.03) 0.00 (0.08) Number of children −0.31 (0.14)* 0.09 (0.13) 0.09 (0.15) −0.60 (0.39) Model fit (f-statistic) 3.49*** 1.86* 2.02* 4.96*** . (1) Depressed . (2) Lonely . (3) Content . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . In school, childless (reference) -- -- -- -- -- Out of school, childless 0.16 (0.23) 0.25 (0.21) −0.73 (0.24)*** 1.12 (0.64)† Out of school, had child 0.70 (0.23)** −0.14 (0.21) −0.26 (0.24) 0.99 (0.64) Corresponding mental health indicator (2009) 0.44 (0.10)*** 0.01 (0.01) 0.03 (0.11) -- Socioeconomic factors Educational attainment (2009) Primary (reference) -- -- -- -- -- Secondary/tertiary education 0.06 (0.20) 0.09 (0.18) −0.21 (0.21) 0.04 (0.55) Experienced food shortage (2009–2011) 0.17 (0.17) −0.02 (0.16) −0.23 (0.19) 1.18 (0.49)* Household goods index (2015) 0.01 (0.04) −0.07 (0.03)* 0.10 (0.04)* −0.48 (0.10)*** Educational attainment (2015)  Primary (reference) -- -- -- -- --  Secondary/tertiary education −0.07 (0.19) −0.10 (0.18) −0.07 (0.20) −0.43 (0.54) Life Shocks Experienced death/illness (2009–2011) 0.24 (0.16) 0.08 (0.14) −0.06 (0.17) 0.46 (0.44) Married (2015) −0.07 (0.17) −0.16 (0.16) 0.19 (0.18) −0.60 (0.49) Additional covariates Age 0.03 (0.08) 0.18 (0.07)* 0.06 (0.08) 0.10 (0.22) Number of siblings 0.03 (0.03) 0.03 (0.03) 0.02 (0.03) 0.00 (0.08) Number of children −0.31 (0.14)* 0.09 (0.13) 0.09 (0.15) −0.60 (0.39) Model fit (f-statistic) 3.49*** 1.86* 2.02* 4.96*** Source: Tsogolo la Thanzi 1 (2009–2011) & Tsogolo la Thanzi 2 (2015); † p < 0.1; *p < 0.05; **p < 0.01; N = 298. Models 1–3 outcomes coded as 1 “never,” 2 “almost never,” 3 “sometimes,” 4 “fairly often,” and 5 “very often.” Model 4 outcome ranges from 4 (lowest stress score) to 20 (highest stress score). ○Perceived Stress Scale-4. Open in new tab Table 7. Results from ordinary least squares regression models of mental health outcomes in 2015 on educational expectations and outcomes among the subsample of young women 15 to 19 years old in Balaka, Malawi, disaggregated by motherhood status . (1) Depressed . (2) Lonely . (3) Content . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . In school, childless (reference) -- -- -- -- -- Out of school, childless 0.16 (0.23) 0.25 (0.21) −0.73 (0.24)*** 1.12 (0.64)† Out of school, had child 0.70 (0.23)** −0.14 (0.21) −0.26 (0.24) 0.99 (0.64) Corresponding mental health indicator (2009) 0.44 (0.10)*** 0.01 (0.01) 0.03 (0.11) -- Socioeconomic factors Educational attainment (2009) Primary (reference) -- -- -- -- -- Secondary/tertiary education 0.06 (0.20) 0.09 (0.18) −0.21 (0.21) 0.04 (0.55) Experienced food shortage (2009–2011) 0.17 (0.17) −0.02 (0.16) −0.23 (0.19) 1.18 (0.49)* Household goods index (2015) 0.01 (0.04) −0.07 (0.03)* 0.10 (0.04)* −0.48 (0.10)*** Educational attainment (2015)  Primary (reference) -- -- -- -- --  Secondary/tertiary education −0.07 (0.19) −0.10 (0.18) −0.07 (0.20) −0.43 (0.54) Life Shocks Experienced death/illness (2009–2011) 0.24 (0.16) 0.08 (0.14) −0.06 (0.17) 0.46 (0.44) Married (2015) −0.07 (0.17) −0.16 (0.16) 0.19 (0.18) −0.60 (0.49) Additional covariates Age 0.03 (0.08) 0.18 (0.07)* 0.06 (0.08) 0.10 (0.22) Number of siblings 0.03 (0.03) 0.03 (0.03) 0.02 (0.03) 0.00 (0.08) Number of children −0.31 (0.14)* 0.09 (0.13) 0.09 (0.15) −0.60 (0.39) Model fit (f-statistic) 3.49*** 1.86* 2.02* 4.96*** . (1) Depressed . (2) Lonely . (3) Content . (4) PSS-4○ . . Coeff . SE . . Coeff . SE . . Coeff . SE . . Coeff . SE . In school, childless (reference) -- -- -- -- -- Out of school, childless 0.16 (0.23) 0.25 (0.21) −0.73 (0.24)*** 1.12 (0.64)† Out of school, had child 0.70 (0.23)** −0.14 (0.21) −0.26 (0.24) 0.99 (0.64) Corresponding mental health indicator (2009) 0.44 (0.10)*** 0.01 (0.01) 0.03 (0.11) -- Socioeconomic factors Educational attainment (2009) Primary (reference) -- -- -- -- -- Secondary/tertiary education 0.06 (0.20) 0.09 (0.18) −0.21 (0.21) 0.04 (0.55) Experienced food shortage (2009–2011) 0.17 (0.17) −0.02 (0.16) −0.23 (0.19) 1.18 (0.49)* Household goods index (2015) 0.01 (0.04) −0.07 (0.03)* 0.10 (0.04)* −0.48 (0.10)*** Educational attainment (2015)  Primary (reference) -- -- -- -- --  Secondary/tertiary education −0.07 (0.19) −0.10 (0.18) −0.07 (0.20) −0.43 (0.54) Life Shocks Experienced death/illness (2009–2011) 0.24 (0.16) 0.08 (0.14) −0.06 (0.17) 0.46 (0.44) Married (2015) −0.07 (0.17) −0.16 (0.16) 0.19 (0.18) −0.60 (0.49) Additional covariates Age 0.03 (0.08) 0.18 (0.07)* 0.06 (0.08) 0.10 (0.22) Number of siblings 0.03 (0.03) 0.03 (0.03) 0.02 (0.03) 0.00 (0.08) Number of children −0.31 (0.14)* 0.09 (0.13) 0.09 (0.15) −0.60 (0.39) Model fit (f-statistic) 3.49*** 1.86* 2.02* 4.96*** Source: Tsogolo la Thanzi 1 (2009–2011) & Tsogolo la Thanzi 2 (2015); † p < 0.1; *p < 0.05; **p < 0.01; N = 298. Models 1–3 outcomes coded as 1 “never,” 2 “almost never,” 3 “sometimes,” 4 “fairly often,” and 5 “very often.” Model 4 outcome ranges from 4 (lowest stress score) to 20 (highest stress score). ○Perceived Stress Scale-4. Open in new tab To summarize, young women who discontinued school unexpectedly but became mothers experienced no elevated risk of negative mental health in the short term, but young women who did not experience this traditional marker of adulthood reported feeling more depressed, lonely and less content. An additional four years later, however, these women experienced only marginally higher stress and only marginally lower contentment relative to their peers who realized their educational goals. Young women who became mothers following their educational plans going unrealized were more depressed. These results suggest that although motherhood can contribute to young women’s psychological resilience in the immediate wake of unrealized educational expectations, over time, it corresponds with adverse mental health outcomes. Discussion In low-income contexts across the globe, governments, nongovernmental organizations, and ordinary families place considerable stock in formal education as a means to empower young people and improve their life chances (Esson 2013; Frye 2012; Hervish and Clifton 2012; Kabiru et al. 2013; Mseleku 2015). Youth in poor contexts not only perceive of education as a means to another end, but also as a way to make themselves better, morally-upright, respectable, and enlightened people (Deterding 2015; Frye 2012; Marshall 2016). Young adults’ commitment to education, and their perceived agency over their educational futures even in some of the poorest countries in Africa, raises fundamental questions of whether the frequent mismatch between what youth expect and what they experience bears consequences for their well-being. Focusing specifically on the case of young women in Balaka, Malawi, this study highlights youth vulnerability to mental health problems. Young Malawian women whose expectations to continue further in school went unmet experienced worse mental health than did their peers who realized their expectations, both in the immediate aftermath of their plans going unrealized and, to some degree, even an additional four years later. Specifically, young women’s unrealized expectations to continue school were associated with their feeling more depressed and less content both immediately and four years after their plans failed—at which time they also experienced more stress. Corresponding with evidence of youth resilience in the face of unrealized plans (Martin and Gardner 2016; Reynolds and Baird 2010; Villarreal et al. 2015), our findings further demonstrate that assuming a new role as a mother can psychologically buffer young women from unfulfilled educational expectations. The negative mental health outcomes associated with unrealized educational expectations in the short term were generally not observed among young women who became mothers—a key element of the transition to adulthood—but instead were concentrated among those who remained childless. Four years later, however, young women who had a child around the time their educational plans failed were more depressed relative to their peers who achieved their educational expectations. This implies that the challenges associated with charting a life course one did not envision can produce negative feelings that accumulate over time. Although we are unable to confirm the mechanisms driving the findings shown here, evidence that motherhood generally buffers against the mental health implications of failed educational plans points to issues of identity. Young women who did not realize their educational expectations and did not become mothers—and thus did not adopt a new, socially-valued identity even as they lost their identity as an aspiring student—were generally at greatest risk of negative mental health, particularly in the short term. The psychological protection associated with motherhood counters evidence of the adverse emotional impact of having children, found in other regions (Margolis and Myrskylä 2011) and in other African countries (Conzo, Fuochi and Mencarini 2017). Instead, this protection aligns closely with a vast literature on the psychological benefits of motherhood for women in poor communities (Edin and Kefalas 2005; Nomaguchi and Milkie 2003; Silva and Pugh 2010). Distinct from research emphasizing the importance for young women in poor countries to delay motherhood (Clark 2004; Koski 2016), our findings demonstrate why young women take refuge in motherhood as one of the only life course achievements over which they have control. Yet, just as Edin and Kefalas (2005:20) discuss the “euphoria” and optimism immediately following a child’s birth in poor communities in the United States, in Malawi, having a child offers the most psychological benefit in the short term. The fact that these women were more depressed over time could reflect the challenges of motherhood, especially for women who have less education than envisioned. Importantly, the outcome measures we study are not diagnosable mental health disorders. With the survey items available, we cannot assess whether respondents met a clinical threshold for diagnosis of a specific mental health problem, like depression. Nonetheless, because subthreshold symptoms predict lapses into diagnosable disorders (Judd et al. 1998), the results offer a valuable look at how unrealized expectations affect young women’s mental health. By studying women’s reports of other emotional reactions, including their feelings of loneliness or general (dis)contentment with life, our analysis offers a broader sense of the multiple, even if minor, ways that unrealized plans can negatively influence young women psychologically. Even minor differences in the mental well-being of young Malawian women who did versus did not achieve their educational expectations is notable. As in much of sub-Saharan Africa, day-to-day life in Malawi is unpredictable, spanning issues from the more mundane, such as whether there will be electricity, to the more dire, such as one’s own mortality in the face of exceptionally high death rates (Johnson-Hanks 2006). Growing up in an uncertain environment primes young Malawians to change course readily as circumstances shift (Trinitapoli and Yeatman 2018). Moreover, life in rural Malawi is difficult. Approximately one-fourth of our sample experienced a food shortage over just a two-year period, and three-fourths endured the death of a close acquaintance. The fact that young women’s educational plans going off course is associated with any observable differences in their mental well-being is a testament to the importance of education to these young women. With that said, although our study design features many strengths, including longitudinal data from a group of young women who reside in the same community, we are unable to account for all possible differences between the young women who did versus did not achieve their educational expectations. These unobserved, and possibly unobservable, differences could contribute to the mental health inequality we document. Moreover, we could not follow-up with more than one-fourth of the original sample. The prime source of attrition was migration out of the study area, which could have either a buffering or a compounding influence on young women’s mental health responses to unrealized educational expectations. Our findings could thus be upwardly or downwardly biased estimates of the total population average of the relationships we study here. Furthermore, in our study, very few women expected to be out of school, which required that we compare women according to where they end up—not what they expect. Another valuable comparison would be young women who end up in the same circumstance but who anticipated it to varying degrees. We do find that young women who expected to be out of school generally have better mental health, but with so few of these women in our data, we must interpret these findings cautiously. Additionally, by focusing only on women’s educational expectations, and studying them over a relatively short time horizon, we capture only a narrow window into young women’s visions of their futures, and what it means for those visions to be left unrealized. Even the young women who achieved their educational expectations may have subsequently experienced other failed expectations, including occupational and economic ones. This may be why some mental health differences between young women who did versus did not achieve their educational expectations attenuated over time. Research that simultaneously studies young adults’ educational, occupational, and economic expectations could confirm that realizing educational expectations is of little psychological benefit if opportunities for realizing one’s career and economic expectations are unavailable. Along with raising these additional questions, our study provides valuable insight into the possible downside of the often exceedingly high educational expectations of youth in low-income countries. Over the coming decades, formal education will become more accessible to youths in these contexts, and education will likely feature prominently in their plans to achieve a more prosperous life than that of their parents. High expectations can certainly translate into higher attainment for some youth (Halleröd 2011; Villarreal et al. 2015; Vuolo, Staff and Mortimer 2012). Yet, many young people in poor countries, as well as those in disadvantaged communities in the world’s richest countries, will be left behind due to worsening structural inequality, and will perhaps blame themselves for their underachievement (Aronson 2017; Silva 2013). The answer need not be to discourage youth from aiming high, but rather to develop social programs and policies that support youth in translating their high aspirations into achievements. Doing so would ensure that young people’s profound sense of agency over their futures is not a façade and their optimism not wasted. Footnotes 1 In this paper, we focus explicitly on expectations—individuals’ accounts of what their futures are likely to hold (Hanson 1994). 2 The Malawian education system consists of eight levels of primary school (i.e., standard) and four levels of secondary school (i.e., form). 3 For the full report, see http://worldhappiness.report/ed/2017/. 4 Tsogolo la Thanzi means “healthy futures” in Chichewa, Malawi’s national language. Tsogolo la Thanzi was designed by Jenny Trinitapoli and Sara Yeatman and funded by grants R01-HD058366 and R01-HD077873 from the National Institute of Child Health and Human Development. Project details are described at https://tsogololathanzi.uchicago.edu and in Yeatman et al. (2019). 5 Interviewers asked women whether the following statements were true in the past month: “Munali okhumudwa” (I have felt depressed); “Mumaona ngati bola mudakangofa” (I have felt life was not worth living); “Munali okwanilitsidwa” (I have felt content); “Mumangokhala nokha” (I have felt lonely). 6 Interviewers asked respondents, “Kodi zinthu izi zidakuchitikirani mowirikiza bwanji mwezi watha” (How true are the following statements for you in the last month?). They then asked: (1) “Ndalephere kuwongolela zinthu zofunika pa moyo wanga” (I have felt unable to control the important things in my life); (2) “Ndimazikhulupilira kuti ndili ndi kuthekera kothana ndi mavuto anga” (I have felt confident about my ability to handle my personal problems); (3) “Ndimazimva kuti zinthu zimayenda kumbali yanga” (I have felt that things were going my way); (4) “Ndimazimwa kuti mavuto amachulukirachulukira moti ndima lephera kuthana nawo” (I have felt difficulties were piling up so high that I could not overcome them). Respondents reported if they felt this way “Pafupipafupi kwambiri” (very often), “Pafupipafupi” (fairly often), “Nthawizina” (sometimes), “Pafupifupi sizinachitikepo” (almost never), or “Sizinandichitikilepo” (never). 7 We code women as “enrolled” in school if they were enrolled during at least one of the three interviews in 2011—approximately two years after the 2009 interview. In supplemental models, we reclassified women as enrolled if they were (1) continuously enrolled at each 2011 interview (waves 6-8) or (2) enrolled at the interview (wave 8) administered almost exactly 24 months from the date of the 2009 interview, at which time women reported their expectations for their lives in two years. Results were consistent across codings. 8 Only 14 young women expected to be a mother despite also professing an expectation to continue school. The low prevalence of “incompatible” answers in our data is likely because pregnancy leads to school expulsion in this setting. 9 This approach to measuring socioeconomic status has been validated by previous research (Filmer and Pritchett 2001; Howe, Hargreaves, and Huttly 2008). The linear asset index comprises nine durable goods (a bed with a mattress, a television, a radio, a landline or mobile phone, a refrigerator, a bicycle, a motorcycle, an animal-drawn cart, and an automobile) and one household asset (electricity). The focus on a household’s durable goods better captures economic fluctuation over the relatively short time-span of the study. A principal components analysis calculates weights following the same procedure used to construct the Demographic Health Survey wealth index. The resulting index places households on a continuous scale relative to the sample (Filmer and Pritchett 2001; Howe, Hargreaves, and Huttly 2008). To ensure factorial invariance, the weights are consistent across waves. 10 Because of our small sample sizes, as well as the fact that young women in our sample generally report good mental health, in some instances, our outcome measures have small cell counts. To ensure this is not influencing our multivariable estimates, in additional analyses we adopted a penalized likelihood estimation method—the Firth method (1993)—and re-estimated the models as Firth logistic regression models (using firthlogit command in Stata). The Firth approach penalizes the likelihood, and allows convergence to finite estimates, thereby helping reduce the bias that can occur due to small cell sizes. These models, which are available upon request, produce results consistent with the model results we present. 11 In supplementary analyses, we accounted for physical health in 2009 and in 2011 using a self-rated indicator and HIV status. Of the subsample of respondents whom TLT tested for HIV, only 11 were HIV positive. Supplementary models confirmed that including these measures does not alter the key results shown here. 12 Note that we excluded the small group of women who reported that they expected to be out of school and were out of school by 2011 from these analyses given the instability of estimates. Appendix Appendix A. Characteristics of analytic sample of 15- to 19-year-olds in Balaka, Malawi◊ . Baseline survey in 2009 . Followed through 2011 . Followed through 2015 . . Mean (SD)/% . Mean (SD)/% . Mean (SD)/% . Age 16.19 (1.23) 16.13 (1.21) 16.10 (1.18) Educational attainment % Primary 60.87 61.56 63.19 % Secondary/Tertiary 38.46 38.44 36.81 Household asset index 0.47 (2.68) 0.28 (2.56) 0.23 (2.53) Married 0.00 0.00 0.00 Has child 0.00 0.00 0.00 N 502 398 343 . Baseline survey in 2009 . Followed through 2011 . Followed through 2015 . . Mean (SD)/% . Mean (SD)/% . Mean (SD)/% . Age 16.19 (1.23) 16.13 (1.21) 16.10 (1.18) Educational attainment % Primary 60.87 61.56 63.19 % Secondary/Tertiary 38.46 38.44 36.81 Household asset index 0.47 (2.68) 0.28 (2.56) 0.23 (2.53) Married 0.00 0.00 0.00 Has child 0.00 0.00 0.00 N 502 398 343 Source: Tsogolo la Thanzi. ◊All variables measured in 2009. Open in new tab Appendix A. Characteristics of analytic sample of 15- to 19-year-olds in Balaka, Malawi◊ . Baseline survey in 2009 . Followed through 2011 . Followed through 2015 . . Mean (SD)/% . Mean (SD)/% . Mean (SD)/% . Age 16.19 (1.23) 16.13 (1.21) 16.10 (1.18) Educational attainment % Primary 60.87 61.56 63.19 % Secondary/Tertiary 38.46 38.44 36.81 Household asset index 0.47 (2.68) 0.28 (2.56) 0.23 (2.53) Married 0.00 0.00 0.00 Has child 0.00 0.00 0.00 N 502 398 343 . Baseline survey in 2009 . Followed through 2011 . Followed through 2015 . . Mean (SD)/% . Mean (SD)/% . Mean (SD)/% . Age 16.19 (1.23) 16.13 (1.21) 16.10 (1.18) Educational attainment % Primary 60.87 61.56 63.19 % Secondary/Tertiary 38.46 38.44 36.81 Household asset index 0.47 (2.68) 0.28 (2.56) 0.23 (2.53) Married 0.00 0.00 0.00 Has child 0.00 0.00 0.00 N 502 398 343 Source: Tsogolo la Thanzi. ◊All variables measured in 2009. Open in new tab About the Authors Emily Smith-Greenaway is Assistant Professor of Sociology at the University of Southern California. Her research centers on understanding health inequality in low-income communities. Her current research seeks to understand how key transitions in young adults’ lives can influence their own, and their children’s, future health. Her research has appeared in Demography, Journal of Marriage and Family, Population and Development Review, and Social Science Research. Sara Yeatman is Associate Professor of Health and Behavioral Sciences at the University of Colorado Denver and Faculty Affiliate at the CU Population Center. Her research focuses broadly on reproductive health and specifically on the social causes and consequences of unintended fertility and HIV in sub-Saharan Africa and in the United States. Her research has been published in American Sociological Review, Demography, Journal of Marriage and Family, and Population and Development Review. References Abbott , Andrew . 2001 . Time Matters: On Theory and Method . University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Aronson , Pamela . 2017 . “ Contradictions in the American Dream: High Educational Aspirations and Perceptions of Deteriorating Institutional Support .” International Journal of Psychology 52 ( 1 ): 49 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat Bachan , Lauren K . 2014 . “ Anticipatory Child Fostering and Household Economic Security in Malawi .” Demographic Research 30 : 1157 . Google Scholar Crossref Search ADS PubMed WorldCat Baird , Chardie L , Stephanie W. Burge , and John R. Reynolds . 2008 . “ Absurdly Ambitious? Teenagers’ Expectations for the Future and the Realities of Social Structure .” Sociology Compass 2 ( 3 ): 944 – 62 . Google Scholar Crossref Search ADS WorldCat Bergmann , Werner . 1992 . “ The Problem of Time in Sociology an Overview of the Literature on the State of Theory and Research on Thesociology of Time’, 1900–82 .” Time & Society 1 ( 1 ): 81 – 134 . Google Scholar Crossref Search ADS WorldCat Bloom , David E. , Elizabeth Cafiero , Eva Jané-Llopis , Shafika Abrahams-Gessel , Lakshmi Reddy Bloom , Sana Fathima , Andrea B. Feigl , Tom Gaziano , Ali Hamandi , Danny O'Farrell, and Emre . 2012 . “The Global Economic Burden of Noncommunicable Diseases.” PGDA Working Papers 8712, Program on the Global Demography of Aging. Bowie , Cameron . 2006 . “ The Burden of Disease in Malawi .” Malawi Medical Journal: The Journal of Medical Association of Malawi 18 ( 3 ): 103 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Cerulo , Karen A . 2008 . Never Saw It Coming: Cultural Challenges to Envisioning the Worst . University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Clark , Shelley . 2004 . “ Early Marriage and Hiv Risks in Sub‐Saharan Africa .” Studies in Family Planning 35 ( 3 ): 149 – 60 . Google Scholar Crossref Search ADS PubMed WorldCat Cohen , S , and G Williamson . 1988 . “Perceived Stress in a Probability Sample of the Us in S. Spacapam & S. Oskamp (Eds.), the Social Psychology of Health: Claremont Symposium on Applied Social Psychology (Pp. 31–67).” Newbury Park, CA: Sage. Cohen , Sheldon , Tom Kamarck , and Robin Mermelstein . 1983 . “ A Global Measure of Perceived Stress .” Journal of Health and Social Behavior 24: 385 – 96 . OpenURL Placeholder Text WorldCat Conzo , Pierluigi , Giulia Fuochi , and Letizia Mencarini . 2017 . “ Fertility and Life Satisfaction in Rural Ethiopia .” Demography 54 ( 4 ): 1331 – 51 . Google Scholar Crossref Search ADS PubMed WorldCat De Boeck , Filip , and Alcinda Honwana . 2005 . “ Children and Youth in Africa: Agency, Identity and Place .” Africa e Mediterraneo; Cultura e Società 51 : 42 – 51 . OpenURL Placeholder Text WorldCat Deterding , Nicole M . 2015 . “ Instrumental and Expressive Education College Planning in the Face of Poverty .” Sociology of Education 88 ( 4 ): 284 – 301 . Google Scholar Crossref Search ADS WorldCat DeVylder , Jordan E. , Ai Koyanagi , Jay Unick , Hans Oh , Boyoung Nam , and Andrew Stickley . 2016 . “ Stress Sensitivity and Psychotic Experiences in 39 Low-and Middle-Income Countries .” Schizophrenia Bulletin 42: sbw044 . OpenURL Placeholder Text WorldCat Edin , Kathryn , and Maria Kefalas . 2005 . Promises I Can Keep: Why Poor Women Put Motherhood before Marriage . University of California Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Elder , Glen H . 1999 . “The Life Course and Aging: Some Reflections.” Distinguished Scholar Lecture 10. Emirbayer , Mustafa , and Ann Mische . 1998 . “ What Is Agency? ” American journal of Sociology 103 ( 4 ): 962 – 1023 . Google Scholar Crossref Search ADS WorldCat Esson , James . 2013 . “ A Body and a Dream at a Vital Conjuncture: Ghanaian Youth, Uncertainty and the Allure of Football .” Geoforum; Journal of Physical, Human, and Regional Geosciences 47 : 84 – 92 . OpenURL Placeholder Text WorldCat Evens , Emily , Elizabeth Tolley , Jennifer Headley , Donna R. McCarraher , Miriam Hartmann , Vuyelwa T. Mtimkulu , Kgahlisho Nozibele Manenzhe , Gloria Hamela , and Fatima Zulu , SBC FEM-PrEP . 2015 . “ Identifying Factors That Influence Pregnancy Intentions: Evidence from South Africa and Malawi .” Culture, Health & Sexuality 17 ( 3 ): 374 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat Filmer , Deon , and Lant H. Pritchett . 2001 . “ Estimating Wealth Effects without Expenditure Data—or Tears: An Application to Educational Enrollments in States of India .” Demography 38 ( 1 ): 115 – 32 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Fortune , Francis , Olawale Ismail , and Monica Stephen . 2014 . “Rethinking Youth, Livelihoods, and Fragility in West Africa: One Size Doesn’t Fit All.” Fragility, Conflict, and Violence Group, World Bank, Washington, DC. Frank , Jerome D . 1935 . “ Some Psychological Determinants of the Level of Aspiration .” The American Journal of Psychology 47 ( 2 ): 285 – 93 . Google Scholar Crossref Search ADS WorldCat Frye , Margaret . 2012 . “ Bright Futures in Malawi’s New Dawn: Educational Aspirations as Assertions of Identity .” American Journal of Sociology 117 ( 6 ): 1565 . Google Scholar Crossref Search ADS WorldCat Gecas , Viktor , and Monica A. Seff . 1990 . “ Social Class and Self-Esteem: Psychological Centrality, Compensation, and the Relative Effects of Work and Home .” Social Psychology Quarterly 53: 165 – 73 . OpenURL Placeholder Text WorldCat Halleröd , Björn . 2011 . “ What Do Children Know About Their Futures: Do Children’s Expectations Predict Outcomes in Middle Age? ” Social Forces 90 ( 1 ): 65 – 83 . Google Scholar Crossref Search ADS WorldCat Halpern-Manners , Andrew , Landon Schnabel , Elaine M. Hernandez , Judy L. Silberg , and Lindon J. Eaves . 2016 . “ The Relationship between Education and Mental Health: New Evidence from a Discordant Twin Study .” Social Forces 95 ( 1 ): 107 – 31 . Google Scholar Crossref Search ADS WorldCat Hanson , Sandra L . 1994 . “ Lost talent: Unrealized educational aspirations and expectations among US youths .” Sociology of Education , 159 – 183 . OpenURL Placeholder Text WorldCat Harper , Caroline , Rachel Marcus , and Karen Moore . 2003 . “ Enduring Poverty and the Conditions of Childhood: Lifecourse and Intergenerational Poverty Transmissions .” World Development 31 ( 3 ): 535 – 54 . Google Scholar Crossref Search ADS WorldCat Helliwell , J. , R. Layard , and J. Sachs . 2017 . “World Happiness Report 2017”. New York: Sustainable Development Solutions Network. Hervish , Alexandra , and Donna Clifton . 2012 . “Status Report: Adolescents and Young People in Sub-Saharan Africa. Opportunities and Challenges.” Higgins , E. Tory . 1987 . “ Self-Discrepancy: A Theory Relating Self and Affect .” Psychological Review 94 ( 3 ): 319 . Google Scholar Crossref Search ADS PubMed WorldCat ——— . 1989 . “ Self-Discrepancy Theory: What Patterns of Self-Beliefs Cause People to Suffer .” Advances in Experimental Social Psychology 22 : 93 – 136 . OpenURL Placeholder Text WorldCat Hitlin , Steven , and Monica Kirkpatrick Johnson . 2015 . “ Reconceptualizing Agency within the Life Course: The Power of Looking Ahead .” American Journal of Sociology 120 ( 5 ): 1429 . Google Scholar Crossref Search ADS WorldCat Howe , Laura D. , James R. Hargreaves , and Sharon R.A. Huttly . 2008 . “ Issues in the Construction of Wealth Indices for the Measurement of Socio-Economic Position in Low-Income Countries .” Emerging Themes in Epidemiology 5 ( 1 ): 3 . Google Scholar Crossref Search ADS PubMed WorldCat Johnson-Hanks , Jennifer . 2006 . Uncertain Honor: Modern Motherhood in an African Crisis . University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC ——— . 2014 . “ Waiting for the Start .” Ethnographies of Youth and Temporality: Time Objectified 11 : 23 . OpenURL Placeholder Text WorldCat Jones , Gareth A. , and Sylvia Chant . 2009 . “ Globalising Initiatives for Gender Equality and Poverty Reduction: Exploring ‘Failure’with Reference to Education and Work among Urban Youth in the Gambia and Ghana .” Geoforum; Journal of Physical, Human, and Regional Geosciences 40 ( 2 ): 184 – 96 . OpenURL Placeholder Text WorldCat Judd , Lewis L. , Hagop S. Akiskal , Jack D. Maser , Pamela J. Zeller , Jean Endicott , William Coryell , Martin P. Paulus , Jelena L. Kunovac , Andrew C. Leon , and Timothy I. Mueller . 1998 . “ Major Depressive Disorder: A Prospective Study of Residual Subthreshold Depressive Symptoms as Predictor of Rapid Relapse .” Journal of Affective Disorders 50 ( 2 ): 97 – 108 . Google Scholar Crossref Search ADS PubMed WorldCat Kabiru , Caroline W. , Sanyu A. Mojola , Donatien Beguy , and Chinelo Okigbo . 2013 . “ Growing up at the “Margins”: Concerns, Aspirations, and Expectations of Young People Living in Nairobi’s Slums .” Journal of Research on Adolescence 23 ( 1 ): 81 – 94 . Google Scholar Crossref Search ADS PubMed WorldCat Kim , Maria H. , Alick C. Mazenga , Akash Devendra , Saeed Ahmed , Peter N. Kazembe , Xiaoying Yu , Chi Nguyen , and Carla Sharp . 2014 . “ Prevalence of Depression and Validation of the Beck Depression Inventory-Ii and the Children’s Depression Inventory-Short Amongst Hiv-Positive Adolescents in Malawi .” Journal of the International AIDS Society 17 ( 1 ): 18965. OpenURL Placeholder Text WorldCat Kohler , Hans-Peter , Susan C. Watkins , Jere R. Behrman , Philip Anglewicz , Iliana V. Kohler , Rebecca L. Thornton , James , Mkandawire , Hastings Honde , Augustine Hawara , and Ben Chilima . 2014 . “ Cohort Profile: The Malawi Longitudinal Study of Families and Health (Mlsfh) .” International Journal of Epidemiology 54: dyu049 . OpenURL Placeholder Text WorldCat Kohler , Iliana V. , Collin F. Payne , Chiwoza Bandawe , and Hans-Peter Kohler . 2017 . “ The Demography of Mental Health among Mature Adults in a Low-Income, High-Hiv-Prevalence Context .” Demography 54 ( 4 ): 1529 – 58 . Google Scholar Crossref Search ADS PubMed WorldCat Koski , Alissa . 2016 . “Child Marriage in Sub-Saharan Africa: Trends, Effects on Health, and Efforts to Limit the Practice.” McGill University Libraries. Large , Michael D. , and Kristen Marcussen . 2000 . “ Extending Identity Theory to Predict Differential Forms and Degrees of Psychological Distress .” Social Psychology Quarterly 63: 49 – 59 . OpenURL Placeholder Text WorldCat Leavy , Jennifer , and Naomi Hossain . 2014 . “ Who Wants to Farm? Youth Aspirations, Opportunities and Rising Food Prices .” IDS Working Papers 2014 ( 439 ): 1 – 44 . Google Scholar Crossref Search ADS WorldCat Leavy , Jennifer , and Sally Smith . 2010 . “ Future Farmers: Youth Aspirations, Expectations and Life Choices .” Future Agricultures Discussion Paper 13 : 1 – 15 . OpenURL Placeholder Text WorldCat Locke , Catherine , and Dolf J.H. Lintelo . 2012 . “ Young Zambians ‘Waiting’for Opportunities and ‘Working Towards’living Well: Lifecourse and Aspiration in Youth Transitions .” Journal of International Development 24 ( 6 ): 777 – 94 . Google Scholar Crossref Search ADS WorldCat Marcussen , Kristen . 2006 . “ Identities, Self-Esteem, and Psychological Distress: An Application of Identity-Discrepancy Theory .” Sociological Perspectives 49 ( 1 ): 1 – 24 . Google Scholar Crossref Search ADS WorldCat Margolis , Rachel , and Mikko Myrskylä . 2011 . “ A Global Perspective on Happiness and Fertility .” Population and Development Review 37 ( 1 ): 29 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat Marshall , Lydia . 2016 . “ ‘Going to School to Become Good People’: Examining Aspirations to Respectability and Goodness among Schoolchildren in Urban Ethiopia .” Childhood (Copenhagen, Denmark) 23 ( 3 ): 423 – 37 . OpenURL Placeholder Text WorldCat Martin , Anne , and Margo Gardner . 2016 . “ College Expectations for All? The Early Adult Outcomes of Low-Achieving Adolescents Who Expect to Earn a Bachelor’s Degree .” Applied Developmental Science 20 ( 2 ): 108 – 20 . Google Scholar Crossref Search ADS WorldCat McLeod , Jane D. , and Elbert P. Almazan . 2003 . Handbook of the Life Course . New York : Kluwer/Plenum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Miller-Grandvaux , Yolande , Michel Welmond , and Joy Wolf . 2002 . “Evolving Partnerships: The Role of Ngos in Basic Education in Africa.” Mische , Ann . 2009 . “Projects and Possibilities: Researching Futures in Action.” In Sociological forum , Vol. 24 , Pp. 694 – 704 . Wiley Online Library . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mortimer , Jeylan T , Melanie J. Zimmer-Gembeck , Mikki Holmes , and Michael J. Shanahan . 2002 . “ The Process of Occupational Decision Making: Patterns During the Transition to Adulthood .” Journal of Vocational Behavior 61 ( 3 ): 439 – 65 . Google Scholar Crossref Search ADS WorldCat Mseleku , Zethembe . 2015 . “Aspirations for Higher Education: Evidence from Youth Living in Kenneth Gardens Municipal Housing Estate (Durban).” School of Built Environment and Development Studies, Faculty of Humanities, University of KwaZulu-Natal, Durban, South Africa. Mundy , Karen E . 2002 . “ Retrospect and Prospect: Education in a Reforming World Bank .” International Journal of Educational Development 22 ( 5 ): 483 – 508 . Google Scholar Crossref Search ADS WorldCat Nauman , Elizabeth , Mark VanLandingham , Philip Anglewicz , Umaporn Patthavanit , and Sureeporn Punpuing . 2015 . “ Rural-to-Urban Migration and Changes in Health among Young Adults in Thailand .” Demography 52 ( 1 ): 233 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat Nielsen , Kelly . 2015 . “ ‘Fake It’til You Make It”Why Community College Students’ Aspirations “Hold Steady’ .” Sociology of Education 88 ( 4 ): 265 – 83 . Google Scholar Crossref Search ADS WorldCat Nomaguchi , Kei M. , and Melissa A. Milkie . 2003 . “ Costs and Rewards of Children: The Effects of Becoming a Parent on Adults’ Lives .” Journal of Marriage and Family 65 ( 2 ): 356 – 74 . Google Scholar Crossref Search ADS WorldCat Norris , Dawn R . 2016 . Job Loss, Identity, and Mental Health . Rutgers University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nowotny , Helga . 1992 . “ Time and Social Theory Towards a Social Theory of Time .” Time & Society 1 ( 3 ): 421 – 54 . Google Scholar Crossref Search ADS WorldCat Pearlin , Leonard I. , Howard B. Kaplan , Carol S. Aneshensel , and Jo C. Phelan . 1999 . The Stress Process Revisited , pp. 77 – 415 . Cham : Springer International Publishing , doi:10.1007/0-387-36223-1_19 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pereira-Morales , Angela J. , Ana Adan , and Diego A. Forero . 2017 . “ Perceived Stress as a Mediator of the Relationship between Neuroticism and Depression and Anxiety Symptoms .” Current Psychology 38: 1 – 9 . OpenURL Placeholder Text WorldCat Reynolds , John R. , and Chardie L. Baird . 2010 . “ Is There a Downside to Shooting for the Stars? Unrealized Educational Expectations and Symptoms of Depression .” American Sociological Review 75 ( 1 ): 151 – 72 . Google Scholar Crossref Search ADS WorldCat Sankoh , Osman , Stephen Sevalie , and Mark Weston . 2018 . “ Mental Health in Africa .” The Lancet Global Health 6 ( 9 ): e954 – e55 . Google Scholar Crossref Search ADS PubMed WorldCat Shanahan , Michael J. , and Jeylan T. Mortimer . 1996 . “ Understanding the Positive Consequences of Psychosocial Stress .” Advances in Group Processes 13 : 189 – 209 . OpenURL Placeholder Text WorldCat Sheeran , Paschal , and Charles Abraham . 1994 . “ Unemployment and Self‐Conception: A Symbolic Interactionist Analysis .” Journal of Community & Applied Social Psychology 4 ( 2 ): 115 – 29 . Google Scholar Crossref Search ADS WorldCat Sheeran , Paschal , and Eunice McCarthy . 1990 . “ The Impact of Unemployment Upon Self-Conception: Evaluation, Affection, Consistency and Involvement Dimensions .” Social Behaviour . OpenURL Placeholder Text WorldCat ——— . 1992 . “ Social Structure, Self‐Conception and Well‐Being: An Examination of Four Models with Unemployed People .” Journal of Applied Social Psychology 22 ( 2 ): 117 – 33 . Google Scholar Crossref Search ADS WorldCat Sieber , Sam D . 1974 . “ Toward a Theory of Role Accumulation .” American Sociological Review , 567 – 78 . OpenURL Placeholder Text WorldCat Silva , Jennifer M . 2012 . “ Constructing Adulthood in an Age of Uncertainty .” American Sociological Review 77 ( 4 ): 505 – 22 . Google Scholar Crossref Search ADS WorldCat ——— . 2013 . Coming up Short: Working-Class Adulthood in an Age of Uncertainty . Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC ——— . 2016 . “ High Hopes and Hidden Inequalities: How Social Class Shapes Pathways to Adulthood .” Emerging Adulthood 4 ( 4 ): 239 – 41 . Google Scholar Crossref Search ADS WorldCat Silva , Jennifer M. , and Allison J. Pugh . 2010 . “ Beyond the Depleting Model of Parenting: Narratives of Childrearing and Change .” Sociological Inquiry 80 ( 4 ): 605 – 27 . Google Scholar Crossref Search ADS PubMed WorldCat Simon , Robin W . 1997 . “ The Meanings Individuals Attach to Role Identities and Their Implications for Mental Health .” Journal of Health and Social Behavior 38: 256 – 74 . OpenURL Placeholder Text WorldCat Smith-Greenaway , Emily . 2015 . “ Are Literacy Skills Associated with Young Adults’ Health in Africa? Evidence from Malawi .” Social Science & Medicine 127 : 124 – 33 . Google Scholar Crossref Search ADS WorldCat Stasavage , David . 2005 . “ Democracy and Education Spending in Africa .” American journal of political science 49 ( 2 ): 343 – 58 . Google Scholar Crossref Search ADS WorldCat Stryker , Sheldon . 1980 . Symbolic Interactionism: A Social Structural Version . Benjamin-Cummings Publishing Company . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sumberg , James , Thomas Yeboah , Justin Flynn , and Nana Akua Anyidoho . 2017 . “ Young People’s Perspectives on Farming in Ghana: A Q Study .” Food Security 9 ( 1 ): 151 – 61 . Google Scholar Crossref Search ADS WorldCat Swidler , Ann , and Susan Cotts Watkins . 2009 . “ ‘Teach a Man to Fish’: The Sustainability Doctrine and Its Social Consequences .” World Development 37 ( 7 ): 1182 – 96 . Google Scholar Crossref Search ADS PubMed WorldCat Tadele , Getnet , and Asrat Ayalew Gella . 2012 . “ ‘A Last Resort and Often Not an Option at All’: Farming and Young People in Ethiopia .” IDS Bulletin 43 ( 6 ): 33 – 43 . Google Scholar Crossref Search ADS WorldCat Tavory , Iddo . 2009 . “ The Structure of Flirtation: On the Construction of Interactional Ambiguity .” Studies in Symbolic Interaction 33 : 59 – 74 . Google Scholar Crossref Search ADS WorldCat Tavory , Iddo , and Nina Eliasoph . 2013 . “ Coordinating Futures: Toward a Theory of Anticipation1 .” American Journal of Sociology 118 ( 4 ): 908 – 42 . Google Scholar Crossref Search ADS WorldCat Thoits , Peggy A . 1986 . “ Multiple Identities: Examining Gender and Marital Status Differences in Distress .” American Sociological Review 35: 259 – 72 . OpenURL Placeholder Text WorldCat ——— . 1994 . “ Stressors and Problem-Solving: The Individual as Psychological Activist .” Journal of Health and Social Behavior , 143 – 60 . OpenURL Placeholder Text WorldCat Trinitapoli , Jenny , and Sara Yeatman . 2018 . “ The flexibility of fertility preferences in a context of uncertainty .” Population and Development Review 44 ( 1 ): 87 . Google Scholar Crossref Search ADS PubMed WorldCat United Nations . 2015 . “Young People.” United Nations Department of Economic and Social Affairs; Population Division. United Nations Development Program . 2010 . Human Development Report . New York : United Nations . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Vigo , Daniel , Graham Thornicroft , and Rifat Atun . 2016 . “ Estimating the True Global Burden of Mental Illness .” The Lancet Psychiatry 3 ( 2 ): 171 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat Villarreal , Brandilynn J. , Jutta Heckhausen , Jared Lessard , Ellen Greenberger , and Chuansheng Chen . 2015 . “ High-School Seniors’ College Enrollment Goals: Costs and Benefits of Ambitious Expectations .” Journal of Adolescence 45 : 327 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat Vuolo , Mike , Jeremy Staff , and Jeylan T. Mortimer . 2012 . “ Weathering the Great Recession: Psychological and Behavioral Trajectories in the Transition from School to Work .” Developmental Psychology 48 ( 6 ): 1759 . Google Scholar Crossref Search ADS PubMed WorldCat Warfa , Nasir , Kamaldeep Bhui , Tom Craig , Sarah Curtis , Salaad Mohamud , Stephen Stansfeld , Paul McCrone , and Graham Thornicroft . 2006 . “Post-Migration Geographical Mobility, Mental Health and Health Service Utilisation among Somali Refugees in the Uk: A Qualitative Study.” In Health & Place , 12 , 503 – 15 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Yeatman , Sara , and Jenny Trinitapoli . 2013 . “ “‘I Will Give Birth But Not Too Much’: HIV-Positive Childbearing in Rural Malawi.” In Women, Motherhood and Living with HIV/AIDS , pp. 93 – 109 . Dordrecht: Springer . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Yeatman , Sara , et al. 2019 . “ Tsogolo la Thanzi: A Longitudinal Study of Young Adults Living in Malawi’s HIV Epidemic .” Studies in family planning . OpenURL Placeholder Text WorldCat Author notes This research was supported by the U.S. National Institute of Child Health and Human Development (NICHD) (R01‐HD058366, R01‐HD077873, and R03-HD 097360). We thank Nina Eliasoph, Paul Lichterman, Sanyu Mojola, Jeremy Staff and Jeremy Staff Susan Watkins for providing detailed comments on earlier drafts. Earlier versions of this article were presented at colloquia held at the University of Southern California, University of Chicago, and Penn State University, and the 2017 Psychosocial Workshop in Chicago, Illinois. Direct correspondence to Emily Smith-Greenaway, Department of Sociology, University of Southern California, 851 Downey Way, Hazel and Stanley Hall, Office 309, Los Angeles, CA 90089, USA; e-mail: smithgre@usc.edu © The Author(s) 2019. Published by Oxford University Press on behalf of the University of North Carolina at Chapel Hill. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Unrealized Educational Expectations and Mental Health: Evidence from a Low-Income Country
JF - Social Forces
DO - 10.1093/sf/soz021
DA - 2020-02-10
UR - https://www.deepdyve.com/lp/oxford-university-press/unrealized-educational-expectations-and-mental-health-evidence-from-a-Go09DjXpGF
SP - 1112
VL - 98
IS - 3
DP - DeepDyve
ER -