Content uploaded by Colleen M. Heflin
Author content
All content in this area was uploaded by Colleen M. Heflin on Mar 25, 2021
Content may be subject to copyright.
Exposure to Food Insecurity during
Adolescence and Educational Attainment
Colleen Heflin
1
, Rajeev Darolia
2
and Sharon Kukla-Acevedo
3
1
Syracuse University,
2
University of Kentucky, and
3
Central Michigan University
ABSTRACT
Previous research has documented the negative consequences of exposure to food inse-
curity over the early childhood period in terms of health and cognitive and behavioral out-
comes, but less research has explored the consequences of exposure to food insecurity at
other points in childhood. We examine the association between food insecurity during ado-
lescence and educational attainment. We begin by exploring a conceptual framework for
the potential mechanisms that might lead adolescents who experience food insecurity to
have differential educational outcomes. Then, we use descriptive and regression analysis to
see whether food insecurity is associated with lower educational attainment using data from
the Panel Study of Income Dynamics Transition to Adulthood Survey. We find that expo-
sure to food insecurity during adolescence predicts lower levels of educational attainment
by reducing college attendance.
KEYWORDS:food insecurity; educational attainment; poverty; adolescence; transition to
adulthood.
Childhood food insecurity in America, defined as the “limited or uncertain availability of
nutritionally-adequate and safe foods or limited or uncertain ability to acquire acceptable foods in
socially-acceptable ways,” is a persistent social problem affecting approximately 6.5 million children
in 2016 (USDA 2019). Levels of household food insecurity increased with the Great Recession from
approximately 11 percent in 2005–2007 to 14.6 percent in 2008–2010 before falling over time to
11.1 percent in 2018 (Coleman-Jensen et al. 2019). Previous research has documented the negative
consequences of exposure to food insecurity over the early childhood period in terms of health and
cognitive and behavioral outcomes (Bhattacharya, Currie, and Haider 2004;Cook and Frank 2008;
Morgane et al. 1993;Pollit, 1994;Scholl and Johnson 2000). However, the current literature has fo-
cused less often on the consequences of adolescent exposure to household food insecurity. This is an
important omission given that Current Population Survey data document the age profile over child-
hood for food insecurity and show that among all food insecure households with children, 26 percent
contain an adolescent between the ages of 13–15, while only 17 percent contain a child of 5–8, and
Supported by the U.S. Department of Health and Human Services, Administration for Children and Families (grant number
9OPD0277 to Heflin), this article represents the views of the authors and not the U.S. Department of Health and Human Services.
This research uses public-use, de-identified, secondary data and is, therefore, not considered human subjects research as defined by
the Syracuse University Institutional Review Board. Please direct correspondence to the first author at Department of Public
Administration and International Affairs, The Maxwell School of Citizenship and Public Affairs, 426 Eggers Hall, Syracuse University,
Syracuse, New York 13244; telephone: (315) 443-9042; email: cmheflin@maxwell.syr.edu.
V
CThe Author(s) 2020. Published by Oxford University Press on behalf of the Society for the Study of Social Problems.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
1
Social Problems, 2020, 0, 1–17
doi: 10.1093/socpro/spaa036
Article
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
only 12 percent contain a children of age 4 or younger (Nord 2009).
1
Moffitt and Ribar (2018) pro-
vide additional evidence that adolescence is a period of special concern. Specifically, when food inse-
curity occurs in households with multiple children at different ages, families with very few economic
resources prioritize allocating food to younger children to the disadvantage of adolescent children.
We propose a conceptual framework postulating that food insecurity at adolescence is a key period
that affects adolescents’ cognitive functioning, mental health, and family functioning, which, in turn,
can reduce educational attainment. Thus, one contribution of our work is the theoretical and empiri-
cal focus on exposure to food insecurity during the adolescent period specifically.
Our study also contributes to the literature that links material disadvantage to educational dispar-
ities through the presence of shared risk factors associated with poverty. There is evidence that stu-
dents’ family income levels positively correspond to educational attainment and outputs such as test
scores. For example, Bailey and Dynarski (2011) show large and growing gaps in college entry among
students of different income levels when they were on the cusp of adulthood (ages 15–18). Other
researchers show that socioeconomic status positively predicts higher test scores (e.g., Duncan and
Magnuson 2011;Reardon 2011). Michelmore and Dynarski (2017) stress the importance of the per-
sistence of disadvantage, showing that that the number of years a child is eligible for subsidized meals
(where eligibility is largely based on income) predicts lower test scores, and also that the timing of
the disadvantage matters. Chetty, Friedman, and Rockoff (2014) link test scores in grades 3–8 (about
ages 8 to 13) to longer term academic outcomes (e.g., college going); this suggests that the early aca-
demic achievement gaps attributed to differences in socioeconomic status, as in Reardon (2011) and
Duncan and Magnuson (2011), are likely to manifest in long-term educational disparities. Finally,
there has also been an increased interest in material hardship experienced among students while in
college, with researchers finding notable rates of food insecurity among college students and linking
food insecurity to poor academic performance and health (e.g., Blagg et al. 2017;Broton and
Goldrick-Rab 2018;Knol et al. 2017;Wood and Harris 2018).
Drawing on prior research from a range of disciplines that touch on what might lead adolescents
who experience food insecurity to have differential educational outcomes, we test our framework us-
ing data on food insecure and food secure adolescents from the Panel Study of Income Dynamics
(PSID). Based on unadjusted means, those who report adolescent household food insecurity had
lower high school graduation rates, lower rates of attending college, and lower college graduation
rates than those who were food secure during adolescence. When we control for exposure to food in-
security during other moments in the life cycle, the level and variance of permanent income, and de-
mographic characteristics, results attenuate in magnitude, but the college going outcomes remain
large and statistically significant. In sum, our results suggest a link between exposure to food insecu-
rity during adolescence and reductions in young adult educational outcomes. Results from this study
have implications for the organization of the safety net and discussions of the U.S. Farm Bill, which
provides funding for U.S. Department of Agriculture food and nutrition programs.
BACKGROUND
Life Course Perspective
This project uses a theoretical framework that rests on a life course perspective (Alwin 2012;Elder
1994), combined with insights from both family systems and human capital theory. In essence, family
systems theory views the nuclear family during adolescence as a cultural frame within which to view
future opportunities and investment decisions. Under human capital theory, resources used for edu-
cation can be viewed as investments; one implication is that financial expenditures by the family are
expected to yield future benefits through labor market activity. Life course perspective theory situates
1 Furthermore, Nord (2009) reports this age-specific pattern occurs independent of the number of children in the household and
the greater food needs of older children.
2Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
the experience of food insecurity in time and allows us to evaluate the relative importance of food in-
security at different points during childhood. When household food insecurity occurs, it is likely expe-
rienced as a period of family stress. The life course perspective suggests that the consequences of this
stress will be different depending on when during childhood the stress is experienced. Recent evi-
dence from Chetty and colleagues (2016,2018), who focus on the timing of residential moves during
childhood on later life outcomes, lends empirical support to this argument.
It is important to examine the differential consequences of exposure to food insecurity across the
life course, because children cycle in and out of food insecurity over childhood. Hamersma and Kim
(2016) document that food insecurity is a dynamic, frequently changing experience for teenagers and
that one-year transitions in food insecurity were much more common than consistent reports of food
insecurity for two years in a row. Whereas, only about four percent of all teenagers were food inse-
cure across a two-year period, over twice as many teenagers were identified as food insecure only
once over the two-year period (teenagers were about equally likely to transition from being food se-
cure to food insecure and vice versa).
Humans are more malleable to social and policy circumstances during significant developmental
periods, rendering these periods particularly interesting to scholars and decision makers.
Developmental psychologists broadly agree that adolescent development is second only to early
childhood development in regard to significant individual biological and psychological transforma-
tions (Eccles 1999;Schwartz et al. 2012).
The school setting (and by extension educational outcomes) is especially important to adolescents
for two significant reasons. First, the school setting is where adolescents spend the majority of their
time outside of the home, and those social interactions are particularly important to their develop-
ment (Eccles and Roeser 2011). Second, in the education model, “skill begets skill,” where earlier ca-
pabilities affect the later educational productivity potential of an individual (Cunha et al. 2006). If
food insecurity affects skill formation from age 12–15 during this developmentally sensitive period,
then those effects will be evident later, especially during the transition to adulthood.
Food Insecurity and Educational Attainment
While the consequences of exposure to poverty during adolescence are well-studied (Teachman et al.
1997), very little is currently known about how exposure to food insecurity during adolescence affects
the life course, and while the two are related, they remain distinct both statistically and conceptually.
In terms of poverty, previous research using the National Longitudinal Study of Youth finds that ex-
posure to poverty during adolescence is not associated with high school graduation, college atten-
dance, and years of school obtaining, once controls are included such as parental education, family
structure, and IQ (Teachman et al. 1997). Here we hypothesize that household food insecurity expe-
rienced during adolescence, which we define as ages 12 to 15, generates short-term changes in family
functioning, adolescent mental health, and cognitive functioning that reduces educational attainment.
More specifically, as outcomes of interest, we focus on the following outcomes as adolescents transi-
tion to adulthood: high school completion, college enrollment, college degree attainment, and college
matriculation directly from high school.
Our conceptual framework draws from Heflin, Kukla-Acevedo, and Darolia (2019) and posits that
food insecurity during the adolescent period is associated with educational attainment through the
mechanisms of reduced cognitive functioning, family functioning, and mental health (see Figure 1).
In essence, our conceptual framework proposes a model in which food insecurity is the observable
manifestation of household scarcity—and the subsequent resulting changes in adolescents’ cognitive
functioning, mental health, and family functioning reduce educational attainment.
Anandi Mani and colleagues argue that food scarcity reduces the cognitive “bandwidth” available
for other cognitive functions, as daily energy and time are directed toward securing food in a food
insecure environment. Studies demonstrate that this tunneling effect reduces cognitive ability
Food Insecurity 3
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
(Mani et al. 2013;Mullainathan and Shafir 2014). In one study, participants were randomly reminded
of financial constraints and asked a series of questions. Invoking the financial constraint resulted in a “loss
of bandwidth” that had a cognitive impact comparable to a loss of 13 IQ points (on a scale with a 100-
point mean and standard deviation of 15) or an entire night’s sleep (Mani et al. 2013). Based on these
studies, we argue that adolescents dealing with food insecurity may face reduced levels of cognitive func-
tioning that impair their academic performance and subsequent educational attainment.
Additionally, food insecurity may alter family functioning in ways that are detrimental for adoles-
cents’ school engagement. First, family meal consumption is related to a reduced risk of depression and
suicide, fewer high-risk behaviors (such as violence and substance use), as well as increased school suc-
cess (Fulkerson et al. 2006;Harrison et al. 2015). However, in food insecure households with older
children, families eat together less often (Widome et al. 2009). Second, using a sample of 5,366 12–14
year olds from the 2002 National Survey of American Families and controlling for the household pov-
erty level, Ashiabi and O’Neal (2007) found that the relationship between food insecurity and adoles-
cent emotional problems and distress was mediated by diminished parenting quality. Third, Goldrick-
Rab (2016)extendsStack’s (1975) theories about the interdependence of low-income family members
to argue that adolescents in families facing economic hardship may fill adult family roles in a process
called “adultification” (Burton 2007). In these ways, food insecurity potentially harms parent-child inter-
actions, which also negatively affect adolescent well-being and educational attainment.
Finally, there is strong evidence that, conditional on income, food insecurity during adolescence is
associated with a number of internalizing (emotional distress, anxiety, depression) and externalizing
behavioral challenges (substance use and suicidal ideation) (Alaimo, Olson and Frongillo 2001,2002;
McIntyre et al. 2013;McLaughlin et al. 2012). Additional evidence suggests food insecurity may im-
pact “soft skills” that help adolescents navigate the school environment. For instance, food insecure
children have lower levels of self-control, attentiveness, and task persistence (Howard 2011). In turn,
the mental health challenges that are more frequently exhibited among food insecure adolescents may
lead to lower attendance rates, lower rates of high school completion, and reduced motivation to apply
for college. Food insecure adolescents may experience more school suspensions (Alaimo, Olson, and
Frongillo 2001) and higher labor force participation (Hamersma and Kim 2016) than other food se-
cure adolescents. School suspensions and early labor force participation are both associated with less
desirable high school outcomes, including lower graduation rates (e.g., Goldschmidt and Wang 1999;
Greenberger and Steinberg 1986;Marsh and Kleitman 2005;Rothstein 2007).
2
Instability
Food insecurity
during
adolescence
Outcomes
•high school
compleon
•college
enrollment
•
directly to college
•college
compleon
Family funconing
Adolescent cognive
funconing
Adolescent mental
health
Decisions,
acons,
behaviors
Figure 1. Conceptual Model
2 On the other hand, early work experience can be associated with positive labor market outcomes because it can improve voca-
tionally related soft skills (e.g., Baum and Ruhm 2016;Meyer and Wise, 1982).
4Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
DATA AND METHODS
Panel Study of Income Dynamics Transition into Adulthood
We analyze data from the Panel Study of Income Dynamics (PSID) to study the effects of food inse-
curity bouts during adolescence on educational outcomes during the transition to adulthood. The
PSID is a nationally representative sample of families and their children in 1968, and provides infor-
mation on families’ social, economic, and environmental circumstances over the past fifty years. It
was collected annually from 1968–1997, and biennially thereafter (see PSID http://psidonline.isr.
umich.edu).
In 1997, to better understand early human capital formation, the PSID began to collect data on
children aged 0–12 and their parents, called the Child Development Supplement (CDS). CDS in-
cluded 3,563 children in the original sample and followed them over three waves – 1997, 2002, and
2007. Since the PSID collects the most detailed individual information on household heads and
wives, however, and many children present in the CDS did not transition to head/wife status for
many years, a potential gap emerged in the life course of data collection for the CDS children. In or-
der to bridge this gap, the PSID began to collect additional information on these individuals’ transi-
tion into young adulthood in 2005, termed the Transition into Adulthood Study (TA).
Inclusion in the TA requires that children: (1) have participated in the CDS; (2) have turned age
18; (3) are no longer attending high school; and (4) had families still active in PSID. The TA survey
was conducted in 2005 (n¼745); 2007 (n¼1,115); 2009 (n¼1,554); 2011 (n¼1,907); 2013
(n¼1,804); and 2015 (n¼1,887), a time period which spans the Great Recession (PSID Transition
into Adulthood Supplement 2015).
Analytic Sample
To study the effects of food insecurity during adolescence (ages 12–15) on educational outcomes in
early adulthood (18–23), we use TA participants as our primary sample. The main variable of interest
is equal to one if the household reports moderate to severe food insecurity (according to the U.S.
Department of Agriculture’s food security scale variable) at any point between the ages 12–15. We
identified 4,426 respondents who lived in households that reported information on food insecurity at
different points in their childhood (collected in the 1997, 1999, 2001, and 2003 PSID family sample).
We restrict our sample to respondents where household food insecurity information during adoles-
cence is observed, resulting in a sample of 1,790 individuals, 182 of which we consider food insecure
during adolescence.
3
Though PSID is longitudinal, the outcomes we analyze are not repeated (e.g., did the student
graduate from high school?). Therefore, we create an analytic data set at the individual level that
includes the measure of adolescent food security previously described, along with characteristics that
are time invariant in our data (race and gender) and static measures of permanent income (and its
standard deviation) during childhood, as well as including exposure to food insecurity prior to adoles-
cence (ages 0 to 11) and afterwards (ages 16 to 25).
Measures
Because we are primarily interested in the potential relationship between exposure to food insecurity
during adolescence and educational attainment, we analyze a cross-sectional data set including infor-
mation from each individual’s last recorded survey. We measure educational attainment by measuring
whether the individual (1) completed high school; (2) completed some college (defined as having
earned college credits but not having formally earned a degree); and (3) earned a college degree
(associate’s or above). We also measure whether a youth went to college soon after finishing high
school, which we define as being enrolled in college while the individual was 18, 19, or 20 years old.
3 We dropped five individuals who were food secure at adolescence but for whom we lacked background information in the data.
Food Insecurity 5
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Research suggests that delayed entry may lower students’ future income according to a life-cycle
model of earnings, since entering school earlier can allow students to more quickly accrue income
gains associated with education (Ben-Porath 1967). Delayed entry has also been found to be associ-
ated with dropping out of college (Stratton, O’Toole, and Wetzel 2008).
The primary variable of interest is an indicator equal to one when the respondent lived in a house-
hold that reports marginal to very low food security when she was between the ages of 12–15 and
zero if the respondent’s household was food secure. We constructed the dichotomous variable from
the PSID’s household food security scale variable, which was measured in 1997 as part of the Child
Development Supplement and then in 1999, 2001, and 2003 in the household file (PSID 2017). The
Food Security Module is an eighteen-item scale that produces scores from zero to eighteen as per the
U.S. Department of Agriculture’s convention to express the full range of severity of food insecurity as
observed in U.S. households. Our measure includes a range of responses to the availability of food at
the household level from an increase in stress or anxiety about the ability to stretch available food
stores to the more extreme instances in which adults and children in the household experience re-
peated and extensive reductions in food intake (Bickel et al. 2000;Cook et al. 2013;Gundersen,
Kreider, and Pepper 2011;Gundersen and Ziliak 2014;Ziliak, Haist, and Gundersen 2008).
4
We account for the respondents’ age at time of survey, gender (binary male or female), and indica-
tors for African American/black, Hispanic/Latino, and other race/ethnicity. In our preferred specifi-
cations, we also account for various other measures of resources during childhood. As noted above,
income is an important correlate of food insecurity. As such, the analyses control for the level and
variation in permanent household income by accounting for the average annual income and the stan-
dard deviation of annual income. The PSID’s annual total family income (pre-tax) for the years in
which the child was 0–18 years old serves as the base variable for these measures. It includes the tax-
able income and cash transfers of all adults in the household. Prior to variable creation, we used the
U.S. consumer price index to inflate all annual incomes to 2013 levels.
5
The final set of controls helps to more carefully control for the effect of adolescent food insecurity
and the likely covariation present from early childhood and late adolescent food insecurity. Food in-
security early in life is likely to be correlated with food insecurity later in life. To address these issues,
we create an early food insecurity indicator variable equal to one, if a respondent lived in a household
that reported food insecurity at any point until they were eleven years old. We also control for house-
hold food insecurity reported in young adulthood (ages 16–25); in a similar manner, we create a late
food insecurity indicator variable equal to one if a respondent’s household reported food insecurity at
any point between the ages of 16 and 25. The scarcity argument described above contributes to our
expectation that concurrent young adult food insecurity and outcomes will interact in ways that are
detrimental to young adult outcomes. In other words, food insecure young adults experience the
tunneling effect toward securing food, which will harm educational outcomes (Mullainathan and
Shafir 2014). To address this concern, we estimate models with and without controls for young adult
food insecurity.
Unfortunately, because of the structure and timing of the way the data were collected, there are a
non-trivial number of missing observations for the early and late food insecurity variables. About 25
percent of respondents in our sample with otherwise complete data on measures outside of food inse-
curity had missing values for the early food insecurity variable. Nearly all of this missing data are arti-
facts of the survey construction interacting with our designation of adolescence (ages 12–15). Those
respondents who were twelve years old in 1997 are designated as adolescents, and, therefore, their
food insecurity measure is recorded as adolescent food insecurity. Those respondents will not have a
4 Work by Connell et al. (2005) and Fram et al. (2011) suggests that this measure may underestimate the number of children who
are aware of food insecurity in their household.
5 In results not shown but available upon request, we explored the use of permanent income at different stages of childhood.
Specifically, we used different measures for average and standard deviation of permanent income in three age groups: 5–10,
11–14, and 15–18. Results are qualitatively unchanged from the estimates in the paper.
6Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
recorded early food insecurity measure. The rate of missing data for early food security is not statisti-
cally different among the adolescent food secure and food insecure samples (p-value ¼0.81 from a
t-test on the equality of means). Moreover, we estimate whether observed factors predict unknown
early food security and do not find any statistically significant predictors, and all point estimates are
substantively small (the exception is the point estimate for age, as would be expected given the timing
of data collection; see the Appendix). There is a corollary issue with observing late food insecurity,
where this measure is not recorded for nearly half of our sample. Again, these missing data are be-
cause of the timing of data collection, as those who were up to nine years old in 1997 were too young
to have their late adolescent (ages 15–26) food insecurity experiences recorded once the CDS data
collection stopped in 2003. Similar to the estimates for missing early food insecurity, estimates shown
in the Appendix suggest that no observed factors substantively or statistically significantly predict
missing late food security in our data sample. Nonetheless, we estimate models with and without
early/late food insecurity controls and our results are robust to these different specifications.
6
Our goal is to understand whether food insecurity at adolescence predicts later educational out-
comes. We estimate outcome Yfor individual iin year tusing the following equation:
Yit ¼b0þcAFIiþb1X1iþb2X2iþait þhtþeit
Adolescent food insecurity, AFI, is an indicator equal to one if individual iis food insecure during
adolescence; therefore, our relationship of interest is captured by the estimated parameter on this
term, c. All of our outcome variables are dichotomous, and we estimate the above equation using a
logit model and report marginal effects. From these results, we interpret cas the conditional marginal
increase in the predicted probability of the outcome when the individual was food insecure at adoles-
cence as compared to individuals who were observably food secure at adolescence. In all estimates
we control for vectors of age, ait ,
7
and year, ht, dummies at the time of adult survey, and for demo-
graphics (race/ethnicity and gender) in the X
1
-vector. In additional specifications, we control for
available financial factors in the X
2
-vector that are likely related to later outcomes, all measured prior
to adulthood. Specifically, we account for average permanent income, which we define as the average
income of the youth’s family when the youth was 0–18, and standard deviation of permanent income,
which we define as the standard deviation of the youth’s family while the youth was 0–18.
8
We also
include indicators for being food insecure before adolescence (from ages 0–11) or after adolescence
(from ages 16–18). As previously discussed, these measures are not available for every individual, so
we also separately control for unknown food insecurity before and after adolescence in the model.
9
RESULTS
Descriptive Statistics
We present summary statistics for the sample in Table 1 for the pooled sample of 1,790 respondents
split by those who were food secure (adolescent food security) and food insecure (adolescent food
insecurity) at adolescence. 182 individuals, about ten percent of our sample, are considered food
6 We also estimate subsample analyses where we estimate our models using only samples of respondents who had valid food inse-
curity data before adolescence, after adolescence, and both before and after adolescence. Generally, results are directionally con-
sistent than our main results presented in the paper, though sample sizes are smaller, standard errors are larger, and results
attenuate slightly.
7 Because of sparse volume at some ages, we group together respondents aged 18–21 and respondents aged 26–28.
8 In results available upon request, we conduct sensitivity analysis across income groups. Results for subgroups, including those be-
low the median income or in the bottom quartile of income, are similar to our pooled estimates reported in the paper.
9 There are other available measures in the data that could influence educational outcomes including marital status, number of chil-
dren, and being the head of one’s own household. However, we do not include these factors in our estimates of education be-
cause they are potentially simultaneously determined or come after educational outcomes for many respondents. Nonetheless,
inference on our variables of interest are similar when including these factors in our estimates (available upon request).
Food Insecurity 7
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
insecure at ages 12–15. As compared to the adolescent food security group, those who were food in-
secure at adolescence were also more likely to be food insecure before or after adolescence. We also
observe that food insecurity is not permanent throughout one’s life cycle. 44 percent of those food in-
secure at adolescence were in food secure households before 12–15, and 31 percent were food secure
after adolescence. On the other hand, only three percent and 2 percent of those who were food se-
cure at adolescence were in food insecure households before and after that period, respectively.
These numbers are likely underestimated, since food security status is unknown before and/or after
adolescence for a substantial number of respondents; however, we reiterate that the pattern of un-
known food security does not systematically differ between the adolescent food insecurity and adoles-
cent food security groups as evidenced by inability to reject the null hypothesis of differences in the
level of missing food security data before and after adolescence for the two groups.
We do not find statistically significant differences in respondents’ gender across groups, but the
food insecure group is more likely to identify as Hispanic/Latino, African American/Black, or other
race (as compared to White). Not surprisingly, the food insecure group grew up in households with
lower average incomes (about $32K in the adolescent food insecurity group as compared to about
$73K in the adolescent food security group), and with higher variance relative to mean income.
Table 1. Sample Summary Statistics
Adolescent Food Insecurity Adolescent Food Security
Mean SD Mean SD
A. Food Insecurity & Financial Measures
Adolescent food insecurity (%) 100 0 0 0*
Adolescent food security (%) 0 0 100 0*
Food insecurity before adolescence (%) 30 46 3 18*
Food security before adolescence (%) 44 50 72 45*
Unknown food insecurity before adolescence (%) 26 44 25 43
Food insecurity after adolescence (%) 26 44 2 15*
Food security after adolescence (%) 31 46 48 50*
Unknown food insecurity after adolescence (%) 43 50 50 50
Mean Permanent Income (000s) 32.32 19.96 72.61 70.13*
SD Permanent Income (000s) 19.89 10.78 36.41 51.08*
B. Demographic Characteristics
Male 45 50 49 50
Female 55 50 51 50
Latino 18 39 10 30*
White 20 40 46 50*
Black 56 50 42 49*
Other race 5 23 2 15*
C. Educational Attainment Outcomes
Finished high school (%) 85 36 93 26*
Ever went to college (%) 55 50 74 44*
Went to college directly from high school, if college (%) 61 49 76 43*
College Degree
þ
(%) 14 35 32 47*
College Degree (if College)
þ
(%) 25 44 44 50*
Individuals 182 1608
Notes: * indicates statistically significant differences between the adolescent food insecurity and adolescent food security groups at the 95%
confidence level.
þ
Comparisons for “College Degree” and “College Degree (if College)” include only respondents who are at least 22 years
old at the time of their last survey.
8Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Summary Outcomes
In Table 2 we display the summary measures for the outcomes we analyze. Members of the adoles-
cent FI group are about eight percent less likely to have completed high school, over a quarter less
likely to go to college, and about half as likely to have obtained a college degree. Among those who
went to college, adolescent food insecurity students are about one-fifth less likely to have gone to col-
lege without a delay from high school and about half as likely to obtain a college degree. These stark
differences in educational attainment are not surprising, however, given the observed differences in
permanent income and other demographic characteristics in Table 1. As a consequence, we next turn
to regression adjusted outcomes that examine the association between educational attainment and ad-
olescent food insecurity controlling for observed differences.
Regression Adjusted Estimates
Our primary goal in this analysis is to understand whether food insecurity in adolescence predicts
later educational attainment outcomes. The results suggest that the experience of adolescent food in-
security, separately from permanent income (and its variance), might be related to outcomes in early
adulthood that are indicative of later life success. However, because of data limitations, we do not
make the claim that our analysis yields conclusively causal links between food insecurity at adoles-
cence and later outcomes.
We present estimates of secondary school educational attainment and college enrollment in
Tables 2 and 3. Overall, our estimates indicate that adolescent food insecurity corresponds to lower
educational attainment. Also, consistent with prior literature (e.g., Bailey and Dynarski, 2011;Ryan
and Bauman 2016), lower levels of permanent household income while growing up and Black race
are generally related to lower educational attainment. Female gender is positively related to higher
educational attainment in the sample, which is also consistent with recent prior empirical work (e.g.,
Conger and Long 2013;Ryan and Bauman 2016). Surprisingly, conditional on other covariates, food
insecurity before or after adolescence significantly predicts educational attainment in only limited
cases; however, this result should be interpreted with caution, as this finding could be a function of
the large number of respondents who have unknown food insecurity during those periods.
Adolescent food insecurity individuals (see the first column of Table 2) are less likely to finish
high school than their adolescent food secure peers in Model 1, which controls for gender and race
alone. However, the adolescent food insecurity and high school completion relationship is not statisti-
cally significant when accounting for childhood income and food insecurity outside of adolescence in
Model 2 and in the models accounting for earlier or later food insecurity (the magnitude of the point
estimates is about 1–2 percent of the adolescent food secure group mean). The coefficients on the
race variables attenuate once permanent income and other markers of childhood advantage are in-
cluded in the model, suggesting that high school completion may be largely determined by perma-
nent income and its unobserved correlates which are separate from adolescent food insecurity,
though most of the students in the PSID sample earn a high school credential (>90 percent).
However, adolescent food insecurity is strongly related to lower college enrollment in the next set
of estimates in Table 2. In column 5 with the sparsest set of controls, adolescent food insecurity is as-
sociated with a 15.6 percentage point reduction in college attendance; this point estimate attenuates
to about 7–8 percentage points once permanent income and food insecurity outside of adolescence
are included in the model (a magnitude of about 10 percent of the food secure mean). With appro-
priate caveats about missing data, household food insecurity reported before and after adolescence
does not precisely predict educational outcomes, and point estimates are smaller than those for ado-
lescent food security, suggesting that there is something particularly potent about adolescent house-
hold food insecurity.
In the final sets of columns 9–12, we present the results for the outcome indicating whether stu-
dents enrolled in college before turning age 21. This sample includes only those who were observed
Food Insecurity 9
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Table 2. Estimates of Secondary School Educational Attainment and College Enrollment
Finished HS Attended college Directly to college (if college)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
Adolescent food insecurity 0.048** 0.014 0.016 0.015 0.156** 0.068* 0.073* 0.080* 0.092* 0.067 0.062 0.046
(0.017) (0.016) (0.018) (0.018) (0.031) (0.029) (0.031) (0.033) (0.036) (0.036) (0.038) (0.041)
Female 0.045** 0.046** 0.047** 0.047** 0.136** 0.136** 0.137** 0.138** 0.005 0.005 0.004 0.003
(0.013) (0.013) (0.013) (0.013) (0.020) (0.019) (0.019) (0.019) (0.021) (0.021) (0.021) (0.021)
Latino 0.060** 0.022 0.023 0.023 0.020 0.143** 0.144** 0.144** 0.012 0.038 0.036 0.034
(0.023) (0.024) (0.024) (0.024) (0.037) (0.037) (0.037) (0.037) (0.037) (0.040) (0.040) (0.040)
Black 0.090** 0.020 0.021 0.020 0.139** 0.009 0.009 0.009 0.095** 0.054* 0.054* 0.055*
(0.017) (0.017) (0.017) (0.017) (0.021) (0.023) (0.023) (0.023) (0.022) (0.025) (0.025) (0.025)
Other race 0.032 0.022 0.032 0.032 0.000 0.098 0.105 0.106 0.018 0.058 0.058 0.056
(0.045) (0.044) (0.046) (0.046) (0.068) (0.065) (0.066) (0.066) (0.077) (0.078) (0.078) (0.078)
Mean Permanent Income 0.002** 0.002** 0.002** 0.004** 0.004** 0.004** 0.001** 0.001** 0.001**
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
SD Permanent Income 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.000) (0.000) (0.000)
Food insecurity before adolescence 0.005 0.005 0.013 0.015 0.021 0.026
(0.023) (0.023) (0.040) (0.040) (0.056) (0.056)
Unknown food insecurity before adolescence 0.138 0.139 0.290 0.294 1.728 1.808
(0.087) (0.088) (0.198) (0.201) (92.725) (89.983)
Food insecurity after adolescence 0.006 0.029 0.052
(0.025) (0.045) (0.049)
Unknown food insecurity after adolescence 0.895 0.048 0.244*
(59.383) (0.123) (0.117)
Observations 1,790 1,790 1,790 1,790 1,790 1,790 1,790 1,790 1,295 1,295 1,295 1,295
Notes: Omitted race is white and omitted gender is male. Marginal effects from a logit estimation reported for dichotomous outcomes. Estimates control for age at time of latest survey. Permanent income is measured in
thousands. Sample in estimates in columns 9-12 is just those in the sample that attended college.
**p <0.01, *p <0.05.
10 Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Table 3. Estimates of College Educational Attainment
College Degree College Degree (if College)
(1) (2) (3) (4) (5) (6) (7) (8)
Adolescent food insecurity 0.188** 0.107* 0.100* 0.092 0.158** 0.091 0.084 0.069
(0.047) (0.045) (0.046) (0.049) (0.060) (0.059) (0.060) (0.065)
Female 0.093** 0.091** 0.090** 0.090** 0.075* 0.077** 0.077** 0.076*
(0.024) (0.023) (0.023) (0.023) (0.030) (0.030) (0.030) (0.030)
Latino 0.092* 0.037 0.042 0.042 0.116* 0.013 0.020 0.021
(0.040) (0.041) (0.041) (0.041) (0.051) (0.053) (0.053) (0.054)
Black 0.200** 0.090** 0.091** 0.091** 0.208** 0.114** 0.114** 0.114**
(0.024) (0.026) (0.026) (0.026) (0.031) (0.034) (0.034) (0.034)
Other race 0.029 0.062 0.067 0.065 0.037 0.060 0.057 0.057
(0.078) (0.074) (0.075) (0.075) (0.101) (0.099) (0.099) (0.099)
Mean Permanent Income 0.003** 0.003** 0.003** 0.003** 0.003** 0.003**
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
SD Permanent Income 0.001** 0.001** 0.001** 0.002** 0.002** 0.002**
(0.000) (0.000) (0.000) (0.000) (0.001) (0.001)
Food insecurity before adolescence 0.064 0.065 0.064 0.065
(0.082) (0.082) (0.105) (0.105)
Unknown food insecurity before adolescence 2.430 2.407 2.930 2.883
(126.842) (125.904) (170.708) (168.564)
Food insecurity after adolescence 0.023 0.044
(0.058) (0.076)
Unknown food insecurity after adolescence 0.071 0.093
(0.144) (0.185)
Observations 1,360 1,360 1,360 1,360 980 980 980 980
Notes: Omitted race is white and omitted gender is male. Marginal effects from a logit estimation reported for dichotomous outcomes. Estimates control for age at time of latest survey. Permanent income is measured in
thousands. Sample includes only respondents who are at least 22 years old at the time of their last survey. Sample in estimates in columns 5-8 includes only those in the sample that attended college.
**p <0.01, *p <0.05.
Food Insecurity 11
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
to enroll in college, and, therefore, the sample size is smaller (n¼1295). We find point estimates of
6.7–9.2 percentage points (about 8–12 percent of the food secure group mean) for going directly to
college for adolescent food insecurity respondents, but the point estimate is not precisely estimated.
We display results for college degree attainment in Table 3. We restrict this sample to those who
were at least 22 years old by their last survey.
10
The result for college attendance is mirrored in the
results for successfully obtaining a college degree. In the sparse model that controls for only race and
gender, adolescent food insecurity is associated with an 18.8 percentage point reduction in the proba-
bility of receiving a college degree. In other specifications that control permanent income and food
insecurity before and after adolescence, the association between adolescent food insecurity and the
probability of obtaining a college degree reduces to 9.2–10.7 percentage points (about 30 percent of
the food secure group mean, though the parameter estimate in column 4 is statistically significant at
the 90 percent confidence level). Similar to the college enrollment estimates, food insecurity outside
of adolescence is not observed to have precisely estimated predictive power in determining college
degree completion. When restricting to just those who attended college, we observe large and nega-
tive point estimates in all specifications (magnitudes of about 16–36 percent of the food secure sam-
ple mean), but these results are not statistically significant when controlling for food insecurity before
or after adolescence.
Taken together, our results suggest that food insecurity at adolescence negatively predicts young
adult educational outcomes. Using our preferred specifications, we find that adolescent food insecu-
rity individuals were substantially less likely to attend college or earn a college degree than those in
the adolescent food security group. Point estimates are consistently negative for other outcomes (fin-
ishing high school, going directly to college from high school, or earning a college degree conditional
on enrollment), though they are not precisely estimated. The magnitude of the parameter estimates
for high school completion is practically small; however, while we do not want too strongly to rely
on results that are not precisely estimated, the estimates for attending college soon after high school
and for college degree attainment among those who attend college are of a notably large magnitude.
The latter result suggests that food security can not only inhibit students’ access to college, but is also
consistent with extant literature finding insidious effects of food insecurity on college students who
make it to college – our results suggest that such difficulties may have roots, at least partially, in
adolescence.
DISCUSSION AND CONCLUSIONS
While the bulk of research on the consequences of food insecurity has focused on the early childhood
period, this study examines potential educational outcomes associated with experiences of adolescent
food insecurity. We begin by theoretically motivating the potential mechanisms that might link expo-
sure to food insecurity during adolescence to educational outcomes. Then, we use the PSID to see if
there is evidence of the presence of differential outcomes for adolescents based on their food security
status in models that control for permanent income and exposure to food insecurity during other
periods of the life course. We find consistent evidence that adolescent food insecurity is associated
with lower educational attainment across college matriculation measures. However, given the issues
related to missing data on food insecurity at different points in the life course, researchers should in-
terpret these results, which merit further investigation, with caution.
This study contributes to the literature on the consequences of exposure to food insecurity during
childhood by providing evidence that adolescent exposure may be particularly detrimental to second-
ary school completion and higher education outcomes. There are several limitations worth noting,
however. First and foremost, this study is descriptive only and is not able to definitively say that ado-
lescent food insecurity is causally related to educational outcomes. The PSID allows us to control for
10 Results are similar (available upon request) if we do not restrict the sample based on age or if we restrict the sample to include
only older individuals.
12 Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
a wide range of factors in a longitudinal manner, including permanent income and its variance, in ad-
dition to food insecurity at other points in childhood. However, food insecurity is only sporadically
measured, and there are many missing points of observation, so there is ample reason for caution
when interpreting these results. This area of research could certainly benefit from further study using
a variety of methods and data sources to properly explore the causal relationship between exposure
to food insecurity during different points of childhood and later life outcomes, much in the same way
that Greg Duncan and others (Duncan and Brooks-Gunn 1999) have done for poverty exposure. Of
course, one of the specific challenges for this work is identifying the effect of food insecurity directly
and separately from the low–income, family dysfunction and other potential factors that might be re-
lated to the episode of food insecurity. Additionally, while the PSID is a unique dataset with many at-
tractive features, it does not contain detailed information with sufficient variation and power at the
point of the life course we are interested in that would allow us to test the specific mechanisms that
we believe might link food insecurity to educational outcomes. Indeed, we know of no dataset that
provided detailed longitudinal data and sufficient power among food insecure households that would
allow us to directly test our conceptual model. Therefore, we provide a set of possible explanations,
but much work and data collection are needed for researchers fully to understand how exposure to
food insecurity during adolescence changes educational attainment.
There are several potential implications of these results that future research may address. First,
while early childhood may, indeed, be a point of childhood that is particularly important, it may be
that adolescence is also worthy of research and policy attention. While there are several food and nu-
trition programs that focus on the early childhood period (WIC and the Child and Adult Care Food
Program), none are crafted for the adolescent period specifically when non-participation among eligi-
ble populations in school lunch programs is very high. It might be worth thinking about creating tar-
geted interventions for this age specifically, particularly if there are potential benefits in terms of later
life outcomes such as educational attainment. Additionally, to the extent that our food and nutrition
policies and the practices of nonprofit organizations do not support the ability of adolescents to assist
family members (both adults and younger siblings) by providing or redistributing food to household
members, restrictions of benefits based on age or household head status should be reconsidered.
Second, there is evidence from the work of Fram et al. (2011) and Connell et al. (2005) that ado-
lescents in food insecure households may participate in family coping strategies, such as shielding
younger siblings, eating away from home or working, and that these strategies may have significant
implications for their socio-emotional development, family dynamics, and school success. While we
have posited several mechanisms through which exposure to food insecurity may be associated with
educational outcomes, future work should continue to develop and test theories as well as document
behaviors that relate exposure to food insecurity to differential life course attainment through a vari-
ety of mechanisms.
Finally, given that many more children were exposed to household food insecurity during the
Great Recession than in previous measured periods, our study suggests that children who were ado-
lescents may have lower lifetime human capital levels as a consequence. Future research should cer-
tainly continue to explore the implications of exposure to food insecurity over the life course for this
cohort who are now young adults.
Food Insecurity 13
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
REFERENCES
Alaimo, Katherine, Christine M. Olson, and Edward A. Frongillo. 2001. “Food Insufficiency and American School Aged
Children’s Cognitive, Academic, and Psychosocial Development.” Pediatrics 108(1):44–53.
Alaimo, Katherine, Christine M. Olson, and Edward A. Frongillo. 2002. “Family Food Insufficiency, But Not Low
Family Income, Is Positively Associated with Dysthymia and Suicide Symptoms in Adolescents.” The Journal of
Nutrition 132(4):719–725.
Alwin, Duane F. 2012. “Integrating Varieties of Life Course Concepts.” The Journals of Gerontology: Series
B67(2):206–220.
Ashiabi, Godwin S., and Keri K. O’Neal. 2007. “Children’s Health Status: Examining the Associations among Income
Poverty, Material Hardship, and Parental Factors.” Plos One 2(9):e940. https://journals.plos.org/plosone/article/
file?type¼printable&id¼10.1371/journal.pone.0000940.
Bailey, Martha J., and Susan M. Dynarski. 2011. “Inequality in Postsecondary Education.” Pp. 117–132 in Whither
Opportunity? Rising Inequality, Schools, and Children’s Life Chances, edited by G. J. Duncan and R. J. Murnane. New
York: Russell Sage.
Baum, Charles L., and Christopher J. Ruhm. 2016. “The Changing Benefits of Early Work Experience.” Southern
Economic Journal 83(2):343–363.
Ben-Porath, Yoram. 1967. “The Production of Human Capital and the Life Cycle of Earnings.” Journal of Political
Economy 75(4, Part 1):352–365.
Bhattacharya, Jayanta, Janet Currie, and Steven Haider. 2004. “Poverty, Food Insecurity, and Nutritional Outcomes in
Children and Adults.” Journal of Health Economics 23(4):839–862.
Bickel, Gary, Mark Nord, Cristofer Price, William Hamilton, and John Cook. 2000. “Guide to Measuring Household
Food Security, Revised March 2000.” U.S. Department of Agriculture, Food and Nutrition Service. Washington, DC:
USDA.
Blagg, Kristin, Craig Gundersen, Diane W. Schanzenbach, and James P. Ziliak. 2017. “Assessing Food Insecurity on
Campus: A National Look at Food Insecurity among America’s College Students.” Washington, DC: Urban Institute.
https://www.urban.org/research/publication/assessing-food-insecurity-campus.
Appendix Estimates of Missing Food Insecurity before or after Adolescence
(1) (2)
Before Adolescence FI Missing After Adolescence FI Missing
Female 0.003 0.007
(0.003) (0.005)
Latino 0.007 0.004
(0.005) (0.009)
Black 0.002 0.004
(0.004) (0.006)
Other race 0.009 0.007
(0.006) (0.017)
Mean Permanent Income 0.000 0.000
(0.000) (0.000)
SD Permanent Income 0.000 0.000
(0.000) (0.000)
Age 0.038** 0.075**
(0.009) (0.009)
Observations 1,790 1,790
Notes: Omitted race is white and omitted gender is male. Marginal effects from a logit estimation reported. We include age as a linear variable
in these estimates. Permanent income is measured in thousands.
**p <0.01, *p <0.05.
14 Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Broton, Katharine M., and Sara Goldrick-Rab. 2018. “Going without: An Exploration of Food and Housing Insecurity
Among Undergraduates.” Educational Researcher 47(2):121–133.
Burton, Linda. 2007. “Childhood Adultification in Economically Disadvantaged Families: A Conceptual
Model.” Family Relations, 56(4):329–345. www.jstor.org/stable/4541675.
Chetty, Raj, John N. Friedman, and Jonah E. Rockoff. 2014. “Measuring the Impacts of Teachers II: Teacher Value-
Added and Student Outcomes in Adulthood.” American Economic Review 104(9):2633–79.
Chetty, Raj, and Nathaniel Hendren. 2018. “The Impacts of Neighborhoods on Intergenerational Mobility I:
Childhood Exposure Effects.” The Quarterly Journal of Economics 133(3): 1107–1162.
Chetty, Raj, Nathaniel Hendren, and Lawrence F. Katz. 2016. “The Effects of Exposure to Better Neighborhoods on
Children: New Evidence from The Moving to Opportunity Experiment.” American Economic
Review 106(4):855–902.
Cohen, Alexandra O., and Betty J. Casey. 2014. “Rewiring Juvenile Justice: The Intersection of Developmental
Neuroscience and Legal Policy.” Trends in Cognitive Sciences 18(2):63–65.
Coleman-Jensen, Alisha, Matthew P. Rabbitt, Christian A. Gregory, and Anita Singh. 2017. Household Food Security in
the United States in 2016. ERR-237. Washington, DC: USDA, Economic Research Service. https://www.ers.usda.
gov/webdocs/publications/84973/err-237.pdf?v¼9062.8.
Coleman-Jensen, Alisha, Matthew P. Rabbitt, Christian A. Gregory, and Anita Singh. 2019. Household Food Security in
the United States in 2018. ERR-270. Washington, DC: USDA, Economic Research Service. https://www.ers.usda.
gov/webdocs/publications/94849/err-270.pdf?v¼9274.2.
Conger, Dylan, and Mark C. Long. 2013. “Gender Gaps in College Enrollment: The Role of Gender Sorting across
Public High Schools.” Educational Researcher 42(7):371–380.
Connell, Carol L., Kristi L. Lofton, Kathy Yadrick, and Timothy A. Rehner. 2005. “Children’s Experiences of Food
Insecurity Can Assist in Understanding Its Effect on Their Well-Being.” The Journal of Nutrition 135(7):1683–90.
https://doi.org/10.1093/Jn/135.7.1683.
Cook, John T., Maureen Black, Mariana Chilton, Diana Cutts, Stephanie Ettinger De Cuba, Timothy C. Heeren, Ruth
Rose-Jacobs et al 2013. “Are Food Insecurity’s Health Impacts Underestimated in the US Population? Marginal
Food Security Also Predicts Adverse Health Outcomes in Young US Children and Mothers.” Advances in
Nutrition 4(1):51–61.
Cook, John T., and Deborah A. Frank. 2008. “Food Security, Poverty, and Human Development in the United
States.” Annals of the New York Academy of Sciences 1136(1):193–209.
Cunha, Flavio, James J. Heckman, Lance Lochner, and Dimitriy V. Masterov. 2006. “Interpreting the Evidence on Life
Cycle Skill Formation.” Handbook of the Economics of Education 1:697–812.
Duncan, Greg J., and Jeanne Brooks-Gunn. 1999. Consequences of Growing Up Poor. New York: Russell Sage
Foundation,
Duncan, Greg J., and Katherine Magnuson. 2011. “The Nature and Impact of Early Achievement Skills, Attention
Skills, and Behavior Problems.” Pp.47–70 in Whither Opportunity? Rising Inequality, Schools, and Children’s Life
Chances, edited by G. J. Duncan and R. J. Murnane. New York: Russell Sage.
Eccles, Jacquelynne S. 1999. “The Development of Children Ages 6 To 14.” The Future of Children 9(2):30–44.
Eccles, Jacquelynne S., and Robert W. Roeser. 2011. “Schools as Developmental Contexts during Adolescence.” Journal
of Research on Adolescence 21(1):225–241.
Elder, Glen H., Jr. 1994. “Time, Human Agency, and Social Change: Perspectives on the Life Course.” Social Psychology
Quarterly 57(1):4–15.
Fram, Maryah Stella, Edward A. Frongillo, Sonya J. Jones, Roger C. Williams, Michael P. Burke, Kendra P. Deloach,
and Christine E. Blake. 2011. “Children Are Aware of Food Insecurity and Take Responsibility for Managing Food
Resources.” The Journal of Nutrition 141(6):1114–1119.
Fulkerson, Jayne A., Mary Story, Alison Mellin, Nancy Leffert, Dianne Neumark-Sztainer, and Simone A. French. 2006.
“Family Dinner Meal Frequency and Adolescent Development: Relationships with Developmental Assets and
High-Risk Behaviors.” Journal of Adolescent Health 39(3):337–345.
Goldrick-Rab, Sara. 2016. Paying the Price: College Costs, Financial Aid, and the Betrayal of the American Dream. Chicago,
IL: University of Chicago Press.
Goldschmidt, Pete, and Jia Wang. 1999. “When Can Schools Affect Dropout Behavior? A Longitudinal Multilevel
Analysis.” American Educational Research Journal 36(4):715–738.
Greenberger, Ellen, and Laurence D. Steinberg. 1986. When Teenagers Work: The Psychological and Social Costs of
Adolescent Employment. New York: Basic Books, Inc.
Gundersen, Craig, Brent Kreider, and John Pepper. 2011. “The Economics of Food Insecurity in the United
States.” Applied Economic Perspectives and Policy 33(3):281–303.
Food Insecurity 15
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Gundersen, Craig, and James P. Ziliak. 2014. “The Future of Children: Research Report: Childhood Food Insecurity in
the US: Trends, Causes, and Policy Options.” Princeton-Brookings. https://futureofchildren.princeton.edu/sites/
futureofchildren/files/media/childhood_food_insecurity_researchreport-fall2014.pdf.
Hamersma, Sarah, and Matthew Kim. 2016. “Food Security and Teenage Labor Supply.” Applied Economic Perspectives
and Policy 38(1):73–92.
Harrison, Megan E., Mark L. Norris, Nicole Obeid, Maeghan Fu, Hannah Weinstangel, and Margaret Sampson. 2015.
“Systematic Review of the Effects of Family Meal Frequency on Psychosocial Outcomes in Youth.” Canadian
Family Physician 61(2):e96–e106.
Heflin, Colleen, Sharon Kukla-Acevedo, and Rajeev Darolia. 2019. “Adolescent Food Insecurity and Risky Behaviors
and Mental Health During the Transition to Adulthood.” Children and Youth Services Review 105:104416. https://
doi.org/10.1016/j.childyouth.2019.104416.
Howard, Larry L. 2011. “Transitions between Food Insecurity and Food Security Predict Children’s Social Skill
Development During Elementary School.” British Journal of Nutrition 105(12):1852–1860.
Knol, Linda L., Cliff A. Robb, Erin M. Mckinley, and Mary Wood. 2017. “Food Insecurity, Self-Rated Health, and
Obesity among College Students.” American Journal of Health Education 48(4):248–255.
Mani, Anandi, Sendhil Mullainathan, Eldar Shafir, and Jiaying Zhao. 2013. “Poverty Impedes Cognitive Function.”
Science 341(6149):976–980. http://doi.org/10.1126/science.1238041
Marsh, Herbert W., and Sabina Kleitman. “Consequences of Employment During High School: Character Building,
Subversion of Academic Goals, or a Threshold?.” American Educational Research Journal 42(2):331–369.
McIntyre, Lynn, Jeanne V. A. Williams, Dina H. Lavorato, and Scott Patten. 2013. “Depression and Suicide Ideation in
Late Adolescence and Early Adulthood are an Outcome of Child Hunger.” Journal of Affective
Disorders 150(1):123–129.
McLaughlin, Katie A., Jennifer Greif Green, Margarita Alegrıa, E. Jane Costello, Michael J. Gruber, Nancy A. Sampson,
and Ronald C. Kessler. 2012. “Food Insecurity and Mental Disorders in a National Sample of US Adolescents.”
Journal of the American Academy of Child & Adolescent Psychiatry 51(12):1293–1303.
Meyer, Robert H., and David A. Wise. 1982. “High School Preparation and Early Labor Force Experience.” Pp. 247–
348 in the Youth Labor Market Problem: Its Nature, Causes, and Consequences, edited by R. B. Freeman, and D. A.
Wise, Chicago, IL: The University of Chicago Press.
Michelmore, Katherine, and Susan Dynarski. 2017. “The Gap Within the Gap: Using Longitudinal Data to Understand
Income Differences in Educational Outcomes.” AERA Open 3(1):1–18.
Moffitt, Robert A., and David C. Ribar. 2018. “Child Age and Gender Differences in Food Security in a Low-Income
US Inner-City Population.” European Economic Review 109: 23–41.
Morgane, Peter J., Robert Austin-Lafrance, Joseph Bronzino, John Tonkiss, Sofia Diaz-Cintra, L. Cintra, Tom Kemper,
and Janina R. Galler. 1993. “Prenatal Malnutrition and Development of the Brain.” Neuroscience & Biobehavioral
Reviews 17(1):91–128.
Mullainathan, Sendhil, and Eldar Shafir. 2014. Scarcity: The New Science of Having Less and How It Defines Our Lives.
London: Picador Paper.
Nord, Mark. 2009. “Food Security in Households Children: Prevalence, Severity, and Household Characteristics.” U.S.
Department of Agriculture. Economic Research Service. Economic Information Bulletin Number 56. Washington,
DC: USDA.
Panel Study of Income Dynamics. 2017. “User’s Guide.” https://Psidonline.Isr.Umich.Edu/CDS/CDS2014_
Userguide.Pdf.
Pollit, Ernesto. 1994. “Poverty and Child Development: Relevance of Research in Developing Countries to the United
States.” Child Development 65:997S–1001S.
PSID Transition into Adulthood Supplement. 2015. “User Guide.” https://Psidonline.Isr.Umich.Edu/CDS/TAS15_
Userguide.Pdf.
Reardon, Sean F. 2011. “The Widening Academic Achievement Gap between the Rich and the Poor: New Evidence
and Possible Explanations.” Pp. 91–116 in Whither Opportunity? Rising Inequality, Schools, and Children’s Life
Chances, edited by G. J. Duncan and R. J. Murnane. New York: Russell Sage.
Rothstein, Donna S. 2007. “High School Employment and Youths’ Academic Achievement.” Journal of Human
Resources 42(1):194–213.
Ryan, Camille L., and Kurt Bauman. 2016. “Educational Attainment in the United States: 2015: Current Population
Reports.” Washington, DC: US Census Bureau.
Scholl, Theresa O., and William G. Johnson. 2000. “Folic Acid: Influence on the Outcome of Pregnancy.” The American
Journal of Clinical Nutrition 71(5):1295S–1303S.
16 Heflin, Darolia, and Kukla-Acevedo
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
Schwartz, Seth J., M. Brent Donnellan, Russell D. Ravert, Koen Luyckx, and Byron L. Zamboanga. 2012. “Identity
Development, Personality, and Well-Being in Adolescence and Emerging Adulthood: Theory, Research, and
Recent Advances.” Handbook of Psychology. Vol. 6. 2nd ed. https://doi.org/10.1002/9781118133880.hop206014.
Stack, Carol B. 1975. All Our Kin: Strategies for Survival in a Black Community. New York: Harper & Row.
Stratton, Leslie S., Dennis M. O’Toole, and James N. Wetzel. 2008. “A Multinomial Logit Model of College Stopout
and Dropout Behavior.” Economics of Education Review 27(3):319–331.
Teachman, Jay D., Kathleen M. Paasch, Randal D. Day, and Karen P. Carver. 1997. “Poverty during Adolescence and
Subsequent Educational Attainment.” Pp. 382–418 in Consequences of Growing up Poor, edited by Greg J. Duncan
and Jeanne Brooks-Gunn. New York: Russell Sage Foundation.
U.S. Department of Agriculture. 2019. “Food Security in the U.S. Measurement.” https://Www.Ers.Usda.Gov/Topics/
Food-Nutrition-Assistance/Food-Security-In-The-Us/Measurement.Aspx.
Widome, Rachel, Dianne Neumark-Sztainer, Peter J. Hannan, Jess Haines, and Mary Story. 2009. “Eating When There
Is Not Enough to Eat: Eating Behaviors and Perceptions of Food Among Food-Insecure Youths.” American Journal
of Public Health 99(5):822–828.
Wood, J. Luke, and Frank Harris III. 2018. “Experiences with ‘Acute’ Food Insecurity among College Students.”
Educational Researcher 47(2):142–145.
Ziliak, James P., Margaret Haist, and Craig Gundersen. 2008. The Causes, Consequences, and Future of Senior Hunger in
America. National Foundation to End Senior Hunger. http://ukcpr.org/research/seniors/causes-consequences-
and-future-senior-hunger-america.
Food Insecurity 17
Downloaded from https://academic.oup.com/socpro/advance-article/doi/10.1093/socpro/spaa036/5936099 by Syracuse University Library user on 23 October 2020
