Handbook of the Sociology of Education in the 21st Century

14  The Social Contexts of High Schools 325

• Network structure, or the basic topographical density, however, may subsume many students
features of a network, such as how densely whose personal networks vary in density. The
connected people are to each other. density of the network in a school or within cer-
tain segments of the school population is impor-
• Network composition, or the individual and tant in many ways because dense ties facilitate
collective attributes of the people in a net- the creation and enforcement of norms (both
work, such as how racially diverse it is. positive and negative) while also reducing access
to diverse resources and stifling innovation, cre-
• Network norms and values, or the prevailing ativity, and non-conformity. In one study, Falci
behavioral and attitudinal patterns in a net- and McNeely (2009) showed that, whereas boys
work that shape the status and integration of suffered more depressive symptomatology when
individual people, such as the emphasis placed they were embedded in personal networks in
on academic success. their schools that included a large numbers of
densely connected peers, girls suffered more
• Network influence, or the degree to which a when they were embedded in personal networks
network shapes the behaviors and attitudes of in their schools that included large numbers of
individual people, such as the higher odds of relatively disconnected peers. Boys seemed to be
drinking when surrounded by drinking peers. reacting to a sense of being over-controlled,
whereas girls were reacting to a reduced sense of
As we discuss these four aspects of networks, belongingness.
we will occasionally cross “levels.” In the broadest
sense, peer networks capture the entire socioe- Second, the compositional characteristics of
metric “map” of a school—all possible social ties networks offer insight into how schools organize
and the aggregate characteristics and patterns diverse populations. They also illuminate how
that encompass the entire school. In the interme- schools reinforce or break down sharp divisions
diate sense, peer networks capture specific group- in the larger society among sociodemographic
ings within the student body, or smaller groups, defined by social class, immigration, or
collections of people who share many ties among race/ethnicity. Segregation—how the school net-
themselves. In the narrowest sense, peer networks work is divided into distinct sub-networks
capture all of the social ties of a single person according to sociodemographic characteristics—
(i.e., the ego network). For example, if two people has long been a focus of school network research.
attend the same school, any school-level network Segregated networks represent inequality, and
characteristic will be the same for both of them. integrated ones represent more fluid social sys-
They might have different intermediate-l­evel net- tems that likely facilitate more equitable distribu-
works, however, because they are in different tions of opportunities. The level of segregation
social spaces of the school that they share. Even also signals whether the social context of a school
if they are in the same intermediate-level peer is characterized by trust and community rather
network in their school, they might have different than conflict and alienation. Along these lines,
individual-level peer networks because they each Moody (2001) showed that the racial segregation
have their own friends within the same general of school-level networks increased as school
social space. racial diversity increased but then eventually
declined at high levels of diversity. These find-
First, peer networks have basic structures that ings suggest that students took comfort in
are larger and more stable than any of the rela- homophily when confronted with difference but
tionships (or people) embedded in them, and only up to the point where doing so was feasible
those structures help to define what a school is and would not unduly constrain their social ties.
like. One structural feature is density, which This work also demonstrated that extracurricular
refers to how interconnected network ties are. In activities could be mechanisms of creating more
dense school networks, more and more people integrated social contexts in diverse schools.
are tied to the same other people, and the student
body is not divided into specific groups that are
disconnected from each other. School-level

326 R. Crosnoe et al.

Another study—this one focusing on college others. Yes, much of the reason that students’
students—offered insight into how peer networks behaviors and attitudes mirror those of the other
can become racially segregated. Partly, segrega- students in their networks is because they seek
tion occurs because of racial homophily, but it out friends and social opportunities that reflect
also occurs because of the tendency for people to who they are, what they do, and what they want.
reciprocate friendships with each other and to This strong selection effect, however, does not
become friends with the friends of their friends, totally explain that similarity. Peers socialize
regardless of homophily. In other words, network each other too through modeling, encourage-
segregation can become self-fulfilling (Wimmer ment, and coercion (Dishion et al. 2015; Osgood
and Lewis 2010). et al. 2013). As such, carefully studying the links
between network characteristics and student
Third, the prevailing norms and values within behavior can offer a window into how the strong
a school network can affect the degree to which peer influence of adolescence will vary across
any one student is socially integrated (i.e., is schools. In some schools, peer networks are char-
widely connected with others) and/or popular acterized by anti-social attitudes. Consequently,
(i.e., enjoys the esteem of others in the network transitioning into that school will expose students
and has a high status) in a school. Conversely, the to negative influences that, over time, could facil-
average network positions of students and stu- itate more problematic developmental trajecto-
dent groups with different behavioral and attitu- ries than if they had attended another school. In
dinal profiles and changes in network positions other schools, peer networks are characterized by
among students or student groups who change pro-social attitudes, and the peer influence that a
characteristics or behaviors speak to the prevail- new student will encounter upon entering that
ing norms and values in school social contexts school—relative to another school—will likely
(Ueno 2005; Gest et  al. 2001). If students who facilitate more positive trajectories over time.
look or act a certain way are overrepresented Consider the case of drinking. Although some
among social isolates or among those who are schools have networks in which drinking is wide-
well-connected, researchers can discern what spread, others have networks in which drinking is
tends to be valued or punished within a school. rare. In both cases, smaller and more specific
For example, Kreager and Staff (2009) used net- peer networks within the school-level network
work analyses to provide concrete evidence of might have drinking profiles that are discordant
the oft-discussed sexual double standard in some with those larger networks in which they are
high schools. Specifically, in those schools, the embedded. Not surprisingly, students tend to
more that girls added sexual partners, the less drink more when they attend schools in which the
likely other students in their schools named them overall level of drinking among schoolmates is
as friends. The opposite happened for boys. As high and when their own personal peer networks
another example, Martin-Storey et  al. (2015) are consistently high in drinking. This influence
exploited longitudinal network dynamics to show is not limited to the friends that students have
that LGBT students were seemingly more at-risk within their own networks. Also important, some-
for being isolated within peer networks in pre- times even more, are the acquaintances that a stu-
dominantly White schools than in racially diverse dent meets through their friends and romantic
schools. This pattern likely occurred because partners that characterize more intermediate-­
sexuality was one of the few ways to differentiate level networks. At the same time, drinking helps
and stigmatize in the former but one of many students meet new people and gains them entrée
potential dividing lines in the latter. Such research into parties and social activities. Thus, the social
demonstrates how broad school networks are contexts of schools are a major factor in adoles-
micro-cultures that amp up or downplay youth cent behavior, both because students’ susceptibil-
culture more generally. ity to peer influences and their more agentic
socializing are symptoms of their strong drive to
Fourth, peer networks are a primary channel
through which young people are influenced by

14  The Social Contexts of High Schools 327

become socially integrated during this stage of become larger, more impersonal, and more
life (Cheadle et al. 2013; Crosnoe 2011; Kreager diverse just as adolescents’ developing brains,
and Haynie 2011). pubertal development, and normative individua-
tion from parents increase their need to find sup-
Peer networks, therefore, are multidimen- portive and tight-knit niches that enhance their
sional systems of interpersonal relations and sense of belongingness. Students in the same
interaction that help to characterize the social crowd are viewed by others as a group, tend to
contexts of schools and differentiate one school interact more with each other than with students
from another. They also shed light on the poten- outside the crowd, share some common identity,
tial divergence in life course trajectories between and tend to have behavioral and attitudinal simi-
students in the same school and in different larities. Some crowds may be defined by particu-
schools. Students react to their network positions lar activity orientations (such as jocks), others by
(i.e., stress over being marginalized) and are common behaviors (such as partiers), and still
influenced by those in their networks (i.e., mod- others by demographic compositions (such as
eling the behaviors of others) in ways that shape ethnic groups, like Asian-Americans, within
their general socioemotional development, affect diverse schools). Even though not everyone in a
their academic progress, and moderate the links crowd knows each other, they are much more
between the two. likely to be friends than any two other students
randomly chosen from the student body (Brown
14.2.2 P eer Crowds and Larson 2009; Brown et al. 2008). Importantly,
students can and do change their crowds over
Peer networks capture concrete aspects of time, often through active strategic behaviors
school social contexts. They are defined by rela- during times of transition, as documented so well
tively tangible ties, such as self-identification as in Kinney’s (1999) ethnography titled From
friends or frequent contact. As such, network Headbangers to Hippies. Because crowds exist
positions and features can be quantifiably iden- somewhat independently of the people in them,
tified and linked to individual students, such as however, they are fairly stable over time and
assessing individual students’ popularity by the eventually replace all members over long periods
number of fellow students who claimed them as (Milner 2013; Brown and Larson 2009).
friends. Other research on school social con-
texts—as exemplified by Jocks and Burnouts— Earlier, we mentioned the 1980s movie, The
strives for a more general sense of the various Breakfast Club, which explored some basic high
groups of students who make up the student school social archetypes—the Jock, Princess,
body, the different venues for socialization that Brain, Basket Case, and Criminal—and how they
they offer, and how they reflect or undermine relate to each other. This movie perfectly illus-
the general norms and values of the student trates the idea of peer crowds, as individual peo-
body as a whole. They are defined less by con- ple are perceived and treated according to their
crete ties between specific students and more by group identities rather than their own selves.
shared identities among certain types of stu- Indeed, the movie is so closely related to this line
dents in a school, who may or may not be of research that a team of educational scholars
directly tied to each other. incorporated it into their large-scale high school
data collection. Adolescents were asked to self-­
Such groupings go by many names. We use a identify their school crowds and associated iden-
single term, crowds, here. This term refers to tities according to the five Breakfast Club
large groupings of students that cut across the archetypes, and they were then followed over
student body, loosely linking many smaller time. Not only were the adolescents in the vari-
cliques and friendships into a pool of potential ous crowds behaviorally more similar to their
friends and romantic partners. Crowds emerge as same-crowd peers than to other peers during the
secondary schooling progresses—as schools high school years, they remained more similar

328 R. Crosnoe et al.

well after high school was over (incidentally, the speaks to how important they are (Milner 2013;

Jocks and Brains turned out to be the best-­ Kinney 1999).

adjusted in the long run) (Barber et  al. 2001). Compared to the literature on peer networks,

This research echoed economic research on iden- the literature on peer crowds has paid less atten-

tity groupings in high schools, showing how peer tion to the ways in which the structure and orga-

crowds cluster liked-minded students together nization of a school influence the creation and

and make them more similar over time (Akerlof maintenance of crowds in the school. Recent

and Kranton 2002). Such studies also demon- developments in sociology of education, how-

strate how similar crowd structures are across ever, have sought to better situate peer crowds

schools. within particular school and curricular settings.

Sociologists have argued that the level of peer Specifically, instead of drawing on network data

crowds—not cliques or individual friends—is or self-reports of crowd membership, Frank and

where the pressures towards conformity and the colleagues analyzed thousands of academic tran-

bullying that high schools have become famous scripts and course schedules across a number of

for are most likely to occur. With longer shared schools to identify students who tended to move

personal histories and stronger affective bonds, through school together—sharing similar aca-

friends typically are accepting of each other and demic statuses (and all of the background charac-

are willing to tolerate differences and unconven- teristics associated with those statuses),

tionality. Absent those factors keeping them con- populating the same classes from year to year,

nected, students’ positions in the peer crowd are and participating in the same activities.

much more vulnerable, as unusual or undesirable Conceptualized much like peer crowds, these

behaviors, unacceptable attitudes, and stigma- local positions grouped together students who

tized traits could lead to marginalization and were having a similar experience of attending

exclusion from the crowd. As such, crowds have their high schools, regardless of whether they

strong influence over behavior (McFarland and were friends or not or saw themselves as a group

Pals 2005; Giordano 2003). Indeed, one of the or not. Some were defined by an orientation

major qualitative findings of Fitting In, Standing towards math and science, some were organized

Out was that students tended to view their own by specific activities like band, and some were

cliques in highly positive terms but the large peer defined by being on a clear path to dropout.

crowds that organized the school in highly nega- Unlike “identity” peer crowds, these curricular

tive terms. In their minds, the internal policing of peer crowds varied in both number and nature

crowds and clashes among crowds in the school from high school to high school. The local posi-

were what fulfilled all of the stereotypes of high tions that students were in defined which particu-

schools as miserable places to be. Moreover, stu- lar pocket of the social contexts of their high

dents embraced membership in specific friend- schools that they inhabited, and the configuration

ship groups in their schools but consistently of local positions within a school differentiated

refused to place themselves in a particular crowd. its social context from other schools. Given their

Instead, they saw themselves as bridging multi- curricular nature, this version of peer crowds

ple crowds or above the crowd structure alto- appeared to have particularly strong associations

gether, no doubt influenced by the pejorative with students’ academic progress (Frank et  al.

view of crowds as agents of conformity and 2008; Field et al. 2006).

social oppression in the school (Crosnoe 2011). In many ways, the concept of peer crowds bet-

Because crowds are typically defined by single ter captures how the public thinks about the social

identities related to specific characteristics (e.g., contexts of high schools. With recognizable iden-

academic achievement = Nerds), they may strike tities and names that divide the student body into

adolescents as too narrow and simplistic, even as a manageable number of smaller groups,

the everyday reality of high school social life peer crowds are straightforward, have face

14  The Social Contexts of High Schools 329

validity, and are related to student outcomes in One important theme of research on school
expected ways. climates (and related concepts like school cul-
ture) concerns the tendency for the climates of
14.2.3 School Climate schools to become racialized in often counter-­
productive ways. For example, one of the stron-
Even more abstract than the concept of peer gest school-level influences on the interpersonal
crowds is the notion that schools have a general climate of a school is the racial/ethnic composi-
climate of social relations among students—from tion of the student body. Schools with heteroge-
conflictual, oppressive, and toxic to harmonious, neous student bodies tend to have less positive
affirming, and supportive and everything else in interpersonal climates and are especially low on
between. When scholars, educators, and parents feelings of connectedness among students and
talk about the climate of a school, they are simply their perceptions of schools as fair and safe.
trying to get at whether that school is a good Students of all race/ethnicities tend to like school
place for students, both in terms of their aca- less when they are not in a clear majority. As
demic prospects but also their general socioemo- already mentioned, student bodies tend to divide
tional development (Crosnoe 2011; Blum et  al. down racial/ethnic lines, and, up to a point, the
2004). more diversity there is, the less students feel the
need to cross those lines. Diversity also provides
Beginning with the affective or interpersonal the opportunity for racial/ethnic discrimination
dimensions of school climate, Cohen et al. (2009) and segregation to become apparent to students,
saw the concept as connecting a widespread and students from racial/ethnic minority groups
sense of school belonging, perceptions of fair- may more fully grasp when they are being mis-
ness and safety, and feelings of interpersonal treated by the system if they are in a heteroge-
connectedness. Not surprisingly, students tend to neous setting in which they can make cross-group
do much better academically and otherwise when comparisons (Benner and Graham 2013; Wells
they attend schools high on these aspects of posi- et  al. 2009; Carter 2005; Johnson et  al. 2001;
tive climate. They feel comfortable and secure in Moody 2001). The fact that diverse schools may
such schools and encounter fewer stressors and, struggle building positive interpersonal climates
therefore, are better able to meet the challenges does not provide evidence against the value of
of schooling (Akiba 2010; Hallinan 2008). continuing school desegregation. Instead, it sug-
Contrary to popular opinion, smaller high schools gests that desegregation efforts need to attend to
do not necessarily foster more positive interper- the special climate-related challenges of diverse
sonal climates, and large high schools are not sig- student bodies in order to fully realize the bene-
nificantly more likely to have negative climates fits of desegregation (Crosnoe 2009; Wells et al.
(Gregory et al. 2011; Koth et al. 2008). Another 2009).
dimension of climate is the general academic cli-
mate of the school. Some schools are defined by The large literature concerning the much-­
a clear push for academic success, where achieve- debated oppositional culture thesis (see Ogbu
ment is valued quite broadly, expectations are 1997) delves deeply into the racialization of
consistently high, and support is plentiful. That school climate. One key feature of this multi-­
kind of climate scaffolds students’ navigation of faceted thesis is the argument that Black and
the increasingly differentiated curriculum of high Latino/a peers de-emphasize school achievement
school (including and especially in the face of and equate it with acting White, which is
academic challenges), opens up rather than fore- clearly relevant to the academic climate of
closes academic opportunities more equitably, predominantly racial/ethnic minority schools
and facilitates the flow of information and or schools with sizeable pockets of such students
resources more broadly (Smerdon 2002; Lee and within a diverse student body. Sociologists of
Smith 1999; Shouse 1996). education have been particularly active in
debunking this thesis (e.g., Harris 2011;

330 R. Crosnoe et al.

Ainsworth-Darnell and Downey 1998). theory. It is also important because the social
Ethnographers have also used it as a venue for contexts of schools are relevant to many pro-
making deeper arguments about school climate. grams aiming to improve the academic function-
For example, Carter (2006, 2005) has argued that ing of schools in an era of greater accountability.
the perceptions of an oppositional culture among Many of these programmatic efforts have failed
Black and Latino/a students in schools are mani- to change the formal processes of schooling in
festations of the historical ways that school per- the desired ways or, at least, underperformed as
sonnel have misunderstood minority group a result of not taking the informal processes of
culture, including their tendency to imbue non- schooling into account when trying to achieve
academic behaviors and attitudes (e.g., acting those goals. Moreover, many of the other pro-
tough, rejection of White hegemony) with aca- grammatic efforts aiming to improve the health
demic significance. As another example, Tyson and wellbeing of young people—rather than pro-
(2011) has argued that acting White is a race- moting academic performance—have also been
specific illustration of a general phenomenon that disappointing because they did not harness the
crosses racial/ethnic lines. Specifically, peers power of the informal processes of schooling.
from a broad array of backgrounds denigrate try- In this final section, therefore, we attempt to
ing too hard academically, if not academic suc- connect what sociologists of education and
cess itself, and contribute to what might seem other researchers have learned about the social
like academically apathetic or antagonistic school contexts of schools to “action” in the form of
climates no matter the racial/ethnic composition policies and interventions, educational and
of the schools. Her work demonstrates that aca- otherwise.
demic climates are likely more similar across
racial/ethnic groups (in separate and racially/ 14.3.1 C hanging Social Contexts
ethnically homogenous schools or among differ-
ent racial/ethnic groups in the same schools) than If the peer networks of a high school transmit
the oppositional culture thesis (and the scholarly anti-social values that deflate students’ academic
and public debates on it) imply. efforts, then reversing the informal processes
within that school would help it meet academic
Much like the treatment of social class in benchmarks. If the most influential peer crowds
Jocks and Burnouts and the exploration of gender of a high school are characterized by academic
and sexuality in newer school ethnographies apathy that undermines students’ course-taking
(e.g., Pascoe 2011; Fields 2008), research on the trajectories, then improving the informal pro-
racialization of school climate has tethered what cesses of that school should enable it to reach a
is going on culturally in schools among young higher level of academic performance. In both
people to what is going on culturally outside of cases, the solution seems obvious—create pro-
schools among adults, including their biases, grams to instill and spread more pro-social,
prejudices, and inequities. As such, they illustrate academically-­oriented attitudes and values
how broader social influences are reworked and among students, who would then influence each
reimagined by young people into their own other. The problem with this obvious solution, of
unique school climates. course, is that conceptualizing and executing
such programs is exceedingly difficult.
14.3 P olicy Challenges
and Responses The social contexts of high schools embody
the policy dilemma, which refers to the inherent
As we have already argued, understanding the challenges when the factors most associated with
social contexts of high schools is important in its desired outcomes are the most difficult to change.
own right because doing so sheds much needed Peers powerfully influence students’ behaviors,
light on the ways schools work that is vital to including academic progress, but manipulating

14  The Social Contexts of High Schools 331

peer dynamics from the outside is a tall order both as a way of broadening the academic scope
(Crosnoe 2011). Past research from sociology of of what students are taught but also as a means of
education, public health, and prevention science easing social divisions among students and in
offers possible ways to overcome this challenge. society at large. In both cases, schools have
rejected the policy dilemma and instead actively
Recall Coleman’s conclusion in The Adolescent tried to improve their social contexts for students
Society that schools use academic competitions to (Tintiangco-Cubales et  al. 2014; Poteat et  al.
harness the social contexts of schools for more 2013). Whether the observed benefits of such
academic endeavors. This conclusion speaks to programs are causal, generalize across contexts,
the possibility of changing peer dynamics and endure remains unclear, as does the degree to
through indirect means. Time and again, the which such programs can be implemented out-
extracurricula of schools have been shown to side the realm of social justice issues to improve
influence peer networks, crowds, and other social basic academic norms and attitudes.
relations in schools, including across racial/ethnic
and socioeconomic lines. Also, extracurricular Other examples of efforts to directly change
activities are much easier to manipulate school peer cultures include positive behavioral
through policy than interpersonal dynamics. interventions and supports (PBIS) and social and
Consequently, increasing extracurricular offer- emotional learning (or SEL) efforts. The former
ings, altering requirements for participation, set- is a school-wide approach to reduce the need for
ting standards for the composition of participants, disciplinary actions in schools by developing
and using adult coaches/leaders as well as stu- positive social skills among students (Bradshaw
dent leaders within activities to deliver specific et  al. 2008). The latter involves incorporating
messages to students are avenues for changing socioemotional skill-building exercises into
the social capital that is being traded within peer school curricula and activities as a means of
networks, the peer crowd structure, and the inter- achieving a healthy and supportive school envi-
play of the interpersonal and academic climates ronment for students (Durlak et al. 2001). Both
(Mahoney et al. 2005; Moody 2001). Along those programs are exemplars for altering the peer cul-
same lines, identifying structural and composi- tures of schools in positive ways.
tional factors of schools with functional social
contexts is important. After all, policy interven- 14.3.1.1 Other Avenues of Action
tions aiming to change school structure and com- Another way to address the policy dilemma
position are widely seen as appropriate and related to the social contexts of schools is to
doable. If we know what those factors are and move beyond attempts to change informal pro-
can change or implement them in a school, then cesses and instead concentrate on breaking the
we may indirectly alter the social context of that link between informal and formal processes. In
school over time. other words, schools with negative social con-
texts might not be able to improve those contexts
Lessons for more direct interventions into the but could develop strategies to protect students
social contexts of schools can be derived from from being hurt by them. Consider the ample
recent efforts to create and refine programs to research by sociologists of education on Catholic
increase tolerance among diverse groups of stu- schools. Efforts to explain why student perfor-
dents and to develop multicultural curricula. For mance was much better and socioeconomic dis-
example, Gay–Straight Alliances—which have parities in performance much weaker in such
the goal of fostering greater acceptance of LGBT schools relative to public schools and other kinds
youth in schools through social activities, aware- of private schools eventually centered on the
ness campaigns, and peer advocacy—have benefits of a constrained academic curriculum.
become more common in U.S. high schools in Because all students took the same classes and
recent years. As another example, ethnic studies enrolled in the same programs, they had no
programs and associated culturally aware peda- opportunity to make academic choices that could
gogical practices have received increased attention,

332 R. Crosnoe et al.

be undermined by social influences. All students exacerbated by the strong emphasis on fidelity in

also experienced much greater similarity in the intervention and prevention (i.e., ensuring that

academic norms and expectations to which they programs are implemented in exactly the same

were exposed (Hallinan and Kubitschek 2012; way across different sites), since a “one size fits

Coleman et al. 1982). This constrained curriculum all” mentality is incompatible with the basic

idea has been utilized more recently in policy themes of research on the social contexts of

efforts to equalize the math course-taking of pub- schools (Steiker 2008). The point is that the

lic school students in California and other states. translation of social context research into policy

Although the academic benefits of this transla- action is not just about what can be done about

tion of policy from the private to the public realm the social contexts of schools. The knowledge

have been disappointing (see Penner et al. 2015; derived from this research can improve policy

Attewell and Domina 2008), evaluations have not and intervention far more broadly.

looked at other unintended consequences, such as

whether the academic progress or general behav-

ior of students in the most negative school social 14.4 Conclusion

contexts are protected from further harm.

Finally, understanding the social contexts of When magazines and think tanks put out lists of

schools may help to support the effectiveness of “good” schools and “bad” schools, they are

interventions in changing students’ non-­academic focusing almost solely on the formal processes of

behaviors. Because they provide one-s­ top access education. A school is considered “good” if it

to large numbers of adolescents, high schools consistently meets certain academic benchmarks

have long been popular sites for behavioral inter- (e.g., high test scores) or consistently produces

ventions, even those that seem unrelated to aca- academically successful students (e.g., National

demic performance. Examples include efforts to Merit Scholars, enrollees at prestigious colleges).

curb drinking, improve sexual health, and These discussions rarely touch on what going to

decrease obesity. These programs are often such “good” schools is like. Schools that work

doomed to failure when they are implemented well as educational institutions often have posi-

with inadequate attention to the specific school tive and healthy social contexts, but this “all good

contexts in which they are situated. Messages things go together” pattern is not absolute. Some

about anti-social behavior could fall flat if they academically successful schools might have toxic

contradict what is valued among peers in a social contexts, some academically struggling

school, programs that group together many stu- schools might help students feel safe and develop

dents at the same time might double as social in healthy ways, and some schools could be doing

activities with diluted impact, and the efficacy of better or worse academically if not for the atti-

increasingly popular peer educator and peer men- tudes, norms, and behaviors prevalent in their

toring techniques in programmatic interventions social contexts.

depends on picking the right peers to lead With a significant assist from scholars in other

(Crosnoe and McNeely 2008; Bearman and fields and disciplines, sociologists of education

Brückner 2001; Dishion et  al. 2001). Indeed, have done a great deal to shed light on what a

research on the social contexts of schools points “good” school is and, perhaps more importantly,

to the value of enlisting high-status or well-­ what a “bad” school is. That research has involved

connected students as agents of norm change for inquiry into the social contexts of schools on

interventions to combat key social problems of their own as well as how the social contexts of

childhood and adolescence (e.g., bullying, sub- schools work at cross-purposes with or in support

stance use) in schools (Osgood et al. 2013; Paluck of the educational mission of schools. Without

and Shepherd 2012). the insights of this literature, our understanding

The difficulty of dealing with peer influences of schools would be incomplete and our

and the potential value of incorporating them are policies to reform schools would be misguided.

14  The Social Contexts of High Schools 333

By emphasizing the social contexts of schools, Bryk, A., & Schneider, B. (2003). Trust in schools: A core
therefore, all of those seemingly shallow movies, resource for improvement. New York: Russell Sage.
shows, and books were focusing attention where
it was needed. Buchmann, C., & Dalton, B. (2002). Interpersonal influ-
ences and educational aspirations in 12 countries:
References The importance of institutional context. Sociology of
Education, 75, 99–122.
Ainsworth-Darnell, J.  W., & Downey, D.  B. (1998).
Assessing the oppositional culture explanation for Carbonaro, W.  J. (1998). A little help from my friends’
racial/ethnic differences in school performance. parents: Intergenerational closure and educational out-
American Sociological Review, 63, 536–553. comes. Sociology of Education, 71, 295–313.

Akerlof, G.  A., & Kranton, R.  E. (2002). Identity Carter, P. L. (2005). Keepin’it real: School success beyond
and schooling: Some lessons for the economics Black and White. New York: Oxford University Press.
of education. Journal of Economic Literature, 40,
1167–1201. Carter, P. L. (2006). Straddling boundaries: Identity, cul-
ture, and school. Sociology of Education, 79, 304–328.
Akiba, M. (2010). What predicts fear of school violence
among U.S. adolescents? Teachers College Record, Cheadle, J. E., Stevens, M., Williams, D. T., & Goosby,
112, 68–102. B.  J. (2013). The differential contributions of teen
drinking homophily to new and existing friendships:
Arum, R. (2000). Schools and communities: Ecological An empirical assessment of assortative and proximity
and institutional dimensions. Annual Review of selection mechanisms. Social Science Research, 42,
Sociology, 26, 395–418. 1297–1310.

Attewell, P., & Domina, T. (2008). Raising the bar: Cohen, J., McCabe, L., Michelli, N.  M., & Pickeral, T.
Curricular intensity and academic performance. (2009). School climate: Research, policy, practice,
Educational Evaluation and Policy Analysis, 30, and teacher education. Teachers College Record, 111,
51–71. 180–213.

Barber, B.  L., Eccles, J.  S., & Stone, M.  R. (2001). Coleman, J.  (1961). The adolescent society: The social
Whatever happened to the jock, the brain, and the life of the teenager and its impact on education.
princess? Young adult pathways linked to adolescent New York: Free Press.
activity involvement and social identity. Journal of
Adolescent Research, 16, 429–455. Coleman, J.  S., Hoffer, T., & Kilgore, S. (1982). High

Bearman, P.  S., & Brückner, H. (2001). Promising school achievement: Public, Catholic, and private
the future: Virginity pledges and first intercourse. schools compared. New York: Basic Books.
American Journal of Sociology, 106, 859–912. Crosnoe, R. (2009). Low-income students and the
socioeconomic composition of public high schools.
Bearman, P., Moody, J., & Stovel, K. (1997). The Add American Sociological Review, 74, 709–730.
Health network variable codebook. Chapel Hill: Crosnoe, R. (2011). Fitting in, standing out: Navigating
University of North Carolina.
the social challenges of high school to get an educa-
Benner, A.  D., & Graham, S. (2013). The antecedents tion. New York: Cambridge University Press.
and consequences of discrimination during adoles- Crosnoe, R. (2012). Schools, peers, and the big picture
cence: Does the source of discrimination matter? of adolescent development. In E. Amsel & J. Smetana
Developmental Psychology, 49, 1602–1613. (Eds.), Adolescent vulnerabilities and opportuni-

Blum, R. W., Libbey, H. P., Bishop, J. H., & Bishop, M. ties: Developmental and constructivist perspectives
(2004). School connectedness—Strengthening health (pp.  182–204). New  York: Cambridge University
and education outcomes for teenagers. Journal of Press.
School Health, 74, 231–235. Crosnoe, R., & McNeely, C. (2008). Peer relations, ado-
lescent behavior, and public health research and prac-
Bradshaw, C. P., Koth, C. W., Bevans, K. B., Ialongo, tice. Family & Community Health, 31, S71–S80.
N., & Leaf, P. J. (2008). The impact of school-wide Dishion, T. J., Poulin, F., & Burraston, B. (2001). Peer
positive behavioral interventions and supports (PBIS) goup dynamics associated with iatrogenic effect
on the organizational health of elementary schools. in group interventions with high-risk young ado-
School Psychology Quarterly, 23, 462–473. lescents. New Directions for Child and Adolescent
Development, 2001(91), 79–92.
Brown, B.  B., & Larson, J.  (2009). Peer relationships Dishion, T. J., Kim, H., & Tein, J. Y. (2015). Friendship
in adolescence. In R.  Lerner & L.  Steinberg (Eds.), and adolescent problem behavior: Deviancy train-
Handbook of adolescent psychology (pp.  74–103). ing and coercive joining as dynamic mediators. In
New York: Wiley. T. P. Beauchaine & S. P. Hinshaw (Eds.), The Oxford
handbook of externalizing spectrum disorders (pp.
Brown, B. B., Herman, M., Hamm, J. V., & Heck, D. J. 303–311). New York: Oxford University.
(2008). Ethnicity and image: Correlates of crowd Dornbusch, S., Glasgow, K., & Lin, I. (1996). The social
affiliation among ethnic minority youth. Child structure of schooling. Annual Review of Psychology,
Development, 79, 529–546. 47, 401–429.
Durlak, J. A., Rubin, L. A., & Kahng, R. D. (2001).
Cognitive behavioral therapy for children and

334 R. Crosnoe et al.

adolescents with externalizing problems. Journal of Kreager, D. A., & Haynie, D. L. (2011). Dangerous liai-
Cognitive Psychotherapy, 15, 183. sons? Dating and drinking diffusion in adolescent
Eckert, P. (1989). Jocks and burnouts: Social categories peer networks. American Sociological Review, 76,
and identity in the high school. New  York: Teachers 737–763.
College Press.
Epstein, J.  L., & Karweit, N. (Eds.). (1983). Friends in Kreager, D.  A., & Staff, J.  (2009). The sexual double
standard and adolescent peer acceptance. Social
school: Patterns of selection and influence in second- Psychology Quarterly, 72, 143–164.
ary schools. New York: Elsevier.
Falci, C., & McNeely, C. (2009). Too many friends: Social Kubitschek, W. N., & Hallinan, M. T. (1998). Tracking and
integration, network cohesion and adolescent depres- students’ friendships. Social Psychology Quarterly,
sive symptoms. Social Forces, 87, 2031–2061. 61, 1–15.
Faris, R., & Felmlee, D. (2011). Status struggles network
centrality and gender segregation in same-and cross-­ Lawrence-Lightfoot, S. (1983). The good high school:
gender aggression. American Sociological Review, 76, Portraits of character and culture. New  York: Basic
48–73. Books.
Field, S., Frank, K. A., Schiller, K., Riegle-Crumb, C., &
Muller, C. (2006). Identifying positions from affilia- Lee, V., & Smith, J. B. (1999). Social support and achieve-
tion networks: Preserving the duality of people and ment for young adolescents in Chicago: The role of
events. Social Networks, 28, 97–123. school. American Educational Research Journal, 36,
Fields, J. (2008). Risky lessons: Sex education and social 907–945.
inequality. New Brunswick: Rutgers University Press.
Frank, K. A., Muller, C., Schiller, K. S., Riegle-Crumb, Lin, N., Cook, K., Burt, R., & Burt, S. (Eds.). (2001).
C., Mueller, A. S., Crosnoe, R., & Pearson, J. (2008). Social capital: Theory and research. New Brunswick:
The social dynamics of mathematics coursetaking Transaction Publishers.
in high school. American Journal of Sociology, 113,
1645–1696. Mahoney, J.  L., Larson, R.  W., & Eccles, J.  S. (Eds.).
Gest, S.  D., Graham-Bermann, S.  A., & Hartup, W.  W. (2005). Organized activities as contexts of develop-
(2001). Peer experience: Common and unique fea-
tures of number of friendships, social network central- ment: Extracurricular activities, after school and
ity, and sociometric status. Social Development, 10, community programs. New York: Psychology Press.
23–40. Martin-Storey, A., Cheadle, J.  E., Skalamera, J., &
Giordano, P.  C. (2003). Relationships in adolescence. Crosnoe, R. (2015). Exploring the social integration
Annual Review of Sociology, 29, 257–281. of sexual minority youth across high school contexts.
Gregory, A., Cornell, D., & Fan, X. (2011). The rela- Child Development, 86, 965–975.
tionship of school structures and support to suspen- McFarland, D. A. (2001). Student resistance: How the for-
sion rates for Black and White high school students. mal and informal organization of classrooms facilitate
American Educational Research Journal, 48, 904–934. everyday forms of student defiance. American Journal
Hallinan, M.  T. (2008). Teacher influences on students’ of Sociology, 107, 612–678.
attachment to school. Sociology of Education, 81, McFarland, D. A., Moody, J., Diehl, D., Smith, J. A., &
271–283. Thomas, R. J. (2014). Network ecology and adoles-
Hallinan, M., & Kubitschek, W.  N. (2012). A compari- cent social structure. American Sociological Review,
son of academic achievement and adherence to the 2, 1088–1121.
common school ideal in public and Catholic schools. McFarland, D., & Pals, H. (2005). Motives and contexts
Sociology of Education, 85, 1–22. of identity change: A case for network effects. Social
Harris, A. L. (2011). Kids don’t want to fail. Cambridge, Psychology Quarterly, 68, 289–315.
MA: Harvard University Press. McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001).
Johnson, M.  K., Crosnoe, R., & Elder, G.  H. (2001). Birds of a feather: Homophily in social networks.
Students’ attachment and academic engagement: The Annual Review of Sociology, 27, 415–444.
role of race and ethnicity. Sociology of Education, 74, Milner, M. (2013). Freaks, geeks, and cool kids.
318–340. New York: Routledge.
Kinney, D. A. (1999). From “headbangers” to “hippies”: Moody, J.  (2001). Race, school integration, and friend-
Delineating adolescents’ active attempts to form an ship segregation in America. American Journal of
alternative peer culture. New Directions for Child and Sociology, 107, 679–716.
Adolescent Development, 84, 21–35. Ogbu, J.  U. (1997). African American education: A
Koth, C. W., Bradshaw, C. P., & Leaf, P. J. (2008). A mul- cultural-­ecological perspective. In H.  P. McAdoo
tilevel study of predictors of student perceptions of (Ed.), Black families (pp. 234–250). Thousand Oaks:
school climate: The effect of classroom-level factors. SAGE Publications.
Journal of Educational Psychology, 100, 96–104. Osgood, D.  W., Ragan, D.  T., Wallace, L., Gest, S.  D.,
Feinberg, M.  E., & Moody, J.  (2013). Peers and the
emergence of alcohol use: Influence and selection pro-
cesses in adolescent friendship networks. Journal of
Research on Adolescence, 23, 500–512.
Paluck, E.  L., & Shepherd, H. (2012). The salience of
social referents: A field experiment on collective

14  The Social Contexts of High Schools 335

norms and harassment behavior in a school social net- Smerdon, B. A. (2002). Students’ perceptions of member-
work. Journal of Personality and Social Psychology, ship in their high schools. Sociology of Education, 75,
103, 899. 287–305.
Pascoe, C.  J. (2011). Dude, you’re a fag: Masculinity
and sexuality in high school. Berkeley: University of Steiker, L.  K. H. (2008). Making drug and alcohol pre-
California Press. vention relevant: Adapting evidence-based curricula
Penner, A.  M., Domina, T., Penner, E.  K., & Conley, to unique adolescent cultures. Family & Community
A. (2015). Curricular policy as a collective effects Health, 31, S52–S60.
problem: A distributional approach. Social Science
Research, 52, 627–641. Tintiangco-Cubales, A., Kohli, R., Sacramento, J.,
Poteat, V. P., Sinclair, K. O., DiGiovanni, C. D., Koenig, Henning, N., Agarwal-Rangnath, R., & Sleeter,
B.  W., & Russell, S.  T. (2013). Gay–Straight C. (2014). Toward an ethnic studies pedagogy:
Alliances are associated with student health: A mul- Implications for K–12 schools from the research.
tischool comparison of LGBTQ and heterosexual Urban Review, 47, 104–125.
youth. Journal of Research on Adolescence, 23,
319–330. Tyson, K. (2011). Integration interrupted: Tracking, Black
Rohlen, T.  P. (1983). Japan’s high schools. Berkeley: students, and acting White after Brown. New  York:
University of California Press. Oxford University Press.
Rutter, M., Maughan, B., Mortimer, P., & Ouston,
J. (1979). Fifteen thousand hours: Secondary schools Ueno, K. (2005). The effects of friendship networks on
and their effects on children. Cambridge, MA: Harvard adolescent depressive symptoms. Social Science
University Press. Research, 34, 484–510.
Shouse, R. C. (1996). Academic press and sense of com-
munity: Conflict, congruence, and implications for Wells, A.  S., Jellison Holme, J., Tijerina Revilla, A., &
student achievement. Social Psychology of Education, Korantemaa Atanda, A. (2009). Both sides now: The
1, 47–68. story of school desegregation’s graduates. Berkeley:
University of California Press.

Wimmer, A., & Lewis, K. (2010). Beyond and below
racial homophily: ERG models of a friendship net-
work documented on Facebook. American Journal of
Sociology, 116, 583–642.

Work Intensity and Academic 15
Success

Jeremy Staff, Jeylan T. Mortimer,
and Monica Kirkpatrick Johnson

Abstract 15.1 Work Intensity and Academic
In this chapter, we review prior research exam- Success
ining how teenage work intensity, indicated by
the average hours of paid work, its quality, and Compared to just 20  years ago, teenagers today
duration, relates to both short- and longer-term are much less likely to be holding a part-time job
success in school. We examine the evidence for during the school year. In Table  15.1, we high-
three plausible propositions: (1) that work light this recent shift in school-year youth
intensity in adolescence has a causal effect on employment and average weekly work hours
school achievement and educational attain- based upon nationally representative data from
ment; (2) that these effects are moderated by two cohorts of teenagers in the Monitoring the
gender, race/ethnicity, and family socioeco- Future (MTF) study (n = 88,195 students). In
nomic background; and (3) that the relationship 1994, only about one quarter of high school
between work intensity and academic success is seniors did not work at any point during the
spuriously related to preexisting differences school year, and only about one half of 8th and
between students. We also highlight shifts in the 10th graders avoided employment. By 2014, the
employment experiences of teenagers over the percentage of non-working youth had climbed
past 20 years based on cross-sectional data from dramatically to 41% of 12th graders, 76% of 10th
the Monitoring the Future study, we offer four graders, and 80% of 8th graders. It is also clear
suggestions for future study, and we discuss from Table  15.1 that employed youth nowadays
implications for policy based upon what we are less likely to work intensively (i.e., average
know now about the intensity of teenage work. more than 20 hours per week working during the
school year) compared to teenagers in 1994. For
J. Staff (*) instance, the percentage of 12th, 10th, and 8th
Department of Sociology, Pennsylvania State graders working intensive hours dropped by 36%,
University, University Park, PA, USA 67%, and 75%, respectively, from 1994 to 2014.
e-mail: [email protected] The percentage of youth working low to moderate
hours (i.e., averaging 1–20 h per week) has also
J. T. Mortimer declined since 1994, especially among 8th and
Department of Sociology, University of Minnesota, 10th graders, though the drop was not as steep.
Minneapolis, MN, USA
e-mail: [email protected] This dramatic decline in youth employment
over the past 20 years, especially among teenag-
M. K. Johnson ers who once devoted substantial time to paid
Department of Sociology, Washington State work while attending secondary school, may
University, Pullman, WA, USA
e-mail: [email protected]

© Springer International Publishing AG, part of Springer Nature 2018 337
B. Schneider (ed.), Handbook of the Sociology of Education in the 21st Century, Handbooks
of Sociology and Social Research, https://doi.org/10.1007/978-3-319-76694-2_15

338 J. Staff et al.

Table 15.1  Work intensity during the school year among sity” of the work. While work intensity is usually
teenagers in 1994 and 2014 operationalized by long average weekly hours of
work, in this chapter we consider “intense work”
12th graders Not working 1994 2014 as paid employment that is highly engaging tem-
10th graders 24% 41% porally or subjects the novice worker to stressful
8th graders Moderate hours (1–20) 46% 40% or other low-quality experiences. We also exam-
30% 19% ine risk and protective factors that precede entry
Intensive hours (21 or into the world of work and that can moderate the
more) 54% 76% impacts of employment. In addition, we review
37% 21% research that has shown that certain workplace
Not working 9% 3% experiences can enhance rather than detract from
academics and ultimately benefit long-term edu-
Moderate hours (1–20) 57% 80% cational and occupational attainment (Mortimer
39% 19% 2003).
Intensive hours (21 or 4% 1%
more) In the first part of the chapter, we highlight
some key studies that have documented benefits
Not working as well as drawbacks of working while attending
secondary school for academic achievement,
Moderate hours (1–20) school engagement, problem behaviors, high
school dropout, and long-term educational and
Intensive hours (21 or occupational attainment. We consider several
more) aspects of these work experiences that can make
them more or less intense, such as the average
Source: Monitoring the Future study (sample size = hours spent working each week, the length of the
88,195 students) employment, the type of job and the quality of
the work, and whether it occurs during the school
come as welcome news to scholars who have year or the summer months. We organize this sec-
long encouraged adolescents to wait to seek a job tion by examining the evidence for three plausi-
until after they finish school (Greenberger and ble propositions: that employment intensity
Steinberg 1986; Marsh 1991). The central con- influences school achievement and attainment;
cern surrounding youth employment is that that these effects are moderated by gender, race/
spending time at work detracts from time and ethnicity, and family socioeconomic background;
effort that teenagers could devote to academic and, finally, that employment intensity has no
pursuits and school-related activities. effects on school achievement and attainment
Accordingly, the recent drop in employment and once selection factors and other sources of spuri-
work hours should give today’s students more ousness are accounted for. In the second section
time to prepare for exams, complete school of the chapter, we propose four directions
assignments, meet with teachers or guidance researchers might want to explore in the future,
counselors, spend time in supplemental educa- and discuss policy implications of what we know
tional academies (e.g., college test preparation, now about the intensity of teenage work.
math enrichment courses, etc.), or participate in a
variety of extracurricular activities, such as 15.2 P rior Research on Teenage
sports, intramural clubs, or other school-related Employment
organizations. In addition, by limiting time teens
spend at work, it is sometimes presumed that not Assessing the impact of early employment on
only will youth today be better students, they will school outcomes is not a frequently studied topic
also minimize the risk of the long list of problem in the sociological study of education. More
behaviors that have been associated with teenage
employment, such as delinquency, substance use,
poor mental health, physical injury, school mis-
conduct, negative attitudes toward work, occupa-
tional deviance, and precocious family formation
behaviors.

Though there are some well-documented aca-
demic and developmental risks associated with
adolescent employment (Staff et al. 2015), stud-
ies have shown how these risks vary depending
on several key factors, most notably the “inten-

15  Work Intensity and Academic Success 339

often, variation in the intensity, duration, and 15.2.1 P roposition 1. Employment
quality of adult employment is considered as a Intensity Affects School
consequence of educational achievement and Success and Long-Term
attainment (Kerckhoff 2000). For instance, in Socioeconomic Attainment
longitudinal models illustrating the process of
occupational attainment (Sewell and Hauser In their now classic studies, D’Amico (1984) and
1975), the young person’s pay and job status is Greenberger and Steinberg (1986) hypothesized
only measured after the completion of schooling, that youth employment may have both beneficial
as educational attainment is considered to be a and detrimental effects on academic achievement
key mediator linking parental and offspring occu- and long-term attainment. On the beneficial side,
pational status. Yet, research shows that these work in high school may help youth better pre-
“first” jobs following the completion of school- pare for adult work roles. Teenagers can learn the
ing often occur long after youth enter the labor skills and effort necessary to obtain a job, such as
market (Entwisle et al. 2000; Staff and Mortimer drafting a résumé, locating a job, and interview-
2007), and ignoring employment experience ing with potential employers. These early
gained while young people are still attending employment experiences can help youth learn
school can lead to overestimates of education’s what it takes to keep a job (e.g., punctuality,
positive impact on young adult occupational sta- appropriate workplace conduct), teach them spe-
tus and wage attainments (Light 2001). cific and more general vocational skills, and push
them to think more about the type of work or job
Among social scientists who do study youth conditions (e.g., working with people, high
employment, a debate in the literature surrounds autonomy) they would like after they finish
whether or not early experiences in the work- school. Such progress in vocational development
place, especially after school or during the week- may focus teens’ attention on the kinds of educa-
ends while school is in session, impact success in tional experiences and credentials necessary to
secondary school and long-term educational obtain the kinds of jobs they are looking for,
attainment. Whereas influential committee heightening academic effort and achievement.
reports have summarized research documenting Working may also contribute to the development
the benefits of paid work (National Reseach of interpersonal skills, as young workers learn
Council 1998), and encouraged teenagers to how to relate to co-workers and supervisors and
obtain jobs (Coleman et  al. 1974), research has come in contact with the public (for example, in
also shown how early employment can carry commonly held fast food jobs and retail trade).
some risk to education as well as health and well- Furthermore, the adult supervisors and cowork-
being. As we review below, a sizeable body of ers teens meet at work may help them connect
evidence suggests that employment during the with other employers and serve as valuable non-­
teenage years does in fact affect educational out- familial references as they try to establish them-
comes, but the direction and strength of the effect selves in the adult world of work. In short, by
depends on the intensity of this experience. providing training, skill development, contacts,
Further complicating work–school associations and experience about what it takes to be an
in adolescence are studies that suggest that pur- employee, high school students who have work
ported work effects may vary by a diverse set of experience may gain an advantage in the labor
moderating or spurious influences, such as prior market over their peers who have not yet been
school engagement and academic success, behav- employed.
ioral adjustment problems that emerge long
before entering the world of work, the young per- On the detrimental side, teenage employment
son’s motivation to work and their occupational may interfere with educational progress and
aspirations, and employment opportunities in the encourage problem behaviors, ultimately com-
local labor market. We begin with the proposition promising socioeconomic attainment. As men-
that work intensity affects educational outcomes. tioned previously, time spent at work may limit

340 J. Staff et al.

time that could be devoted to homework, study- (2003) detailed longitudinal analyses of time use
ing, and extracurricular activities. If working patterns of youth followed through the high
teenagers spend late nights at work they may be school years reveal that teens who worked mod-
exhausted the next morning and unprepared for a erate hours during high school spent similar
day of learning. Work may also increase the risk amounts of time doing homework, participating
of problem behaviors through a variety of mecha- in extracurricular activities, volunteering, and
nisms: (1) Exposure to older teens and adults in spending time with friends as those teenagers
the workplace may encourage substance use and who did not work. Other researchers have simi-
family formation behaviors; (2) Coworkers and larly found that employed teenagers do not sacri-
supervisors may emphasize the immediate fice time for leisure reading, studying, doing
rewards of work (money, autonomy) at the homework, or hanging out with their families
expense of more long-term rewards that come when they spend a moderate amount of time at
from education, undermining bonds to school; work (D’Amico 1984; Mihalic and Elliott 1997;
(3) Parents may grant more freedom and auton- Schoenhals et al. 1998; Kalenkoski and Pabilonia
omy to their daughters and sons when they work, 2012). Moderate workers do show an increase in
which in turn increases the time working youth school absences compared to their non-working
spend in unstructured activities with their peers; counterparts (Bachman et  al. 2003; Schoenhals
and (4) Certain jobs and coworkers may enable et al. 1998), but they also spend less time watch-
workplace misconduct, such as theft or providing ing television or in front of a computer screen
friends and family free services, food, and prod- (Kalenkoski and Pabilonia 2012), and show
ucts. Work may also increase the risk of adjust- increases in both school activities (Mihalic and
ment problems by exposing teenagers to undue Elliott 1997) and grade point averages (Mortimer
stress and demands, noxious or unsafe work envi- and Johnson 1998). Overall, teenagers who work
ronments, and workplace sexual harassment. limited hours appear to have sufficient time to
Mental distress and poor behavioral adjustment pursue a wide range of school, work, family, and
could, in turn, lead to immediate declines in leisure activities.
school performance and increase the risk of high
school dropout and diminished long-term educa- Risks to school progress and positive adjust-
tional attainment. ment do emerge when students average long
hours on the job during the school year. For
The majority of research on how youth instance, longitudinal studies have documented
employment impacts educational outcomes has declines in academic achievement and school
focused on one dimension of work intensity: the performance when teenagers average more than
average hours of work youth spend per week in 20 h per week during the school year (Marsh and
their jobs during the school year. We should note Kleitman 2005; Marsh 1991; Monahan et  al.
that this focus on work hours is primarily due to 2011; Staff et al. 2010b; Tyler 2003). For some
survey data limitations, as teenagers rarely are teenagers, the decline in school performance that
asked about other dimensions of their employ- comes with spending long hours on the job may
ment, such as its duration and quality, or their be due to time tradeoff between work and school,
subjective responses to their work. Nonetheless, as intensive work is associated with a greater
if work and school have a zero-sum association in ­frequency of incomplete assignments and tru-
adolescence, then school performance should ancy as well as low school effort and participa-
monotonically decline as the hours of work inten- tion in school activities. In addition, the risks of
sify. However, high school students do not neces- delinquency, substance use, school misbehavior,
sarily lose 2  h of homework time when they truancy, and school suspensions are elevated
spend 2 h at work, as they could spend that free when youth spend long hours on the job (Johnson
time doing a variety of more or less academically 2004; Mortimer et  al. 1996a; Staff and Uggen
or developmentally beneficial activities. 2003; Staff et al. 2010a), compared to occasions
Shanahan and Flaherty (2001) and Mortimer’s when they do not. These findings suggest that the

15  Work Intensity and Academic Success 341

problem behaviors that coincide with intensive the dropouts were the least likely to have previ-
hours of work in adolescence may make it even ously held a job (33%) compared to 47% of the
harder for youth to finish school. stopouts and 49% of the continuous students.
However, when they did work, dropouts were
Several studies have shown that youth who most likely of the three groups to have worked
average more than 20 h per week increase their intensively. For instance, among working youth,
risk of leaving high school without a degree 10% of the dropouts had averaged more than 4 h
(Apel et al. 2008; D’Amico 1984; Lee and Staff of work on a typical school day, compared to 7%
2007; McNeal 1997; Warren and Lee 2003). On of stopouts and 5% of continuous students. It is
average, youth increase their risk of dropout by also worth noting that the continuous students
approximately 40–64% when they work more had the highest percentage who had previously
than 20  h per week during the school year. averaged less than 1  h of work per school day
Importantly, moderate work hours do not carry (73%), compared to 63% and 60% of the drop-
this same risk. In fact, D’Amico’s (1984) seminal outs and stopouts, respectively. These findings
study found that moderate work was associated suggest that work hours continue to have a curvi-
with a reduced risk of high school dropout. linear relationship with school success, despite a
Intensive hours of work during adolescence have substantial drop in the number of youth who are
also been linked to reductions in long-term edu- employed and who work intensive hours.
cational attainment, as intensive workers are less
likely to complete four-year college degrees Whereas the majority of studies have focused
compared to their non-working and moderately on work hours, relatively little research has
employed counterparts (Bachman et  al. 2011b; assessed whether the quality of early employ-
Carr et  al. 1996; Mortimer 2003; Staff and ment experiences matters for short and long-term
Mortimer 2007). Overall, the weight of the evi- achievement. In James Coleman’s influential
dence indicates a non-linear relationship between report (Coleman et  al. 1974), teenagers in the
work hours and academic success: Intensive 1970s were encouraged to find jobs so they could
hours of work increase the risk of school failure develop vocational skills and gain real world
and low educational attainment, whereas moder- experiences in the workplace that would supple-
ate hours range from offering some educational ment what they were learning in school.
benefits to carrying no risk at all. Furthermore, Coleman and his colleagues
stressed that employment would force youth to
Given the dramatic decline in teenage employ- spend valuable time with adult supervisors and
ment, and especially the decline in young work- coworkers instead of idling about with other
ers who devote long hours to their jobs, it is teens. In the 1980s, Greenberger and Steinberg
important to assess whether high-intensity work (1986) argued, in contrast, that the employment
still carries a risk. A recent report from the opportunities available to teens had changed over
U.S. Department of Education (2015) examined the years as the vast majority of youth were toil-
the risk factors for high school dropout among ing in a low-quality “adolescent workplace” with
teenagers in the High School Longitudinal Study few opportunities for skill development, low
of 2009 (the HSLS:09). This study included career relevance, and a predominance of
approximately 24,000 ninth graders (ages 14–15) ­supervisors and coworkers who were also teenag-
who were first surveyed in the fall of 2009, and ers. Studies since then have shown that teenagers
then again in the spring of 2012 (ages 16–17). By work in a diversity of jobs (Hirschman and
the spring of the 11th grade, approximately 3% Voloshin 2007; National Reseach Council 1998)
were dropouts (i.e., not enrolled in school or had with a wide range of opportunities for learning,
not earned a high school diploma or an alterna- skill utilization, career potential, social support,
tive credential), 7% were stopouts (had experi- and interaction with adults (Call and Mortimer
enced at least one 4-week spell out of high school 2001; Finch et al. 1991; Mortimer 2003; Rauscher
but were currently enrolled), and the remaining et al. 2013; Shanahan et al. 1991).
were continuous students. Of these three groups,

342 J. Staff et al.

Yet, as mentioned previously, teenagers today to positive adult role models and controls and less
are much less likely to work compared to cohorts likely to be subject to the stresses of more formal
of youth from just 20  years ago, which reflects employment (Hansen and Jarvis 2000; Hansen
increasing job competition from older workers et  al. 2001). Rural youth still work on family
for entry-level employment (Smith 2011; Staff farms (though youth farm workers have relatively
et al. 2014), the popularity and perceived need for high rates of physical injury [National Research
quality internships, and perhaps increasing pres- Council 1998]).
sure to do well in school to gain access to selec-
tive postsecondary schools (Alon 2009). Has this Regarding the interpersonal contacts in
broad shift in youth employment limited the youth’s jobs, there is little evidence that employed
types of jobs youth could once obtain? Will teenagers mostly work with other teenagers. In
today’s working teenagers be overrepresented in fact, less than 4% of MTF high school seniors in
low-quality jobs, with the potential for intense, 2014 reported working in a job where almost all
stressful work experiences, or will they be spared of their coworkers are the same age and their
these conditions of work if confined to informal supervisor is age 25 or younger. In addition,
work experiences? about 42% of the 2014 seniors who were
employed also reported being “quite” or “com-
To answer these questions, we turn again to pletely” satisfied with their jobs. Despite these
recent survey data from the Monitoring the positive reports of work for students today (i.e.,
Future study. Regarding job type, about 64% of lack of age-segregation and high levels of job sat-
8th graders and 50% of 10th graders in 2014 isfaction), it is clear that their employers continue
worked in informal jobs, either doing lawn work to have very little contact with teachers and coun-
or babysitting. By the 12th grade, employed teen- selors. Among 2014 seniors, 89% of employed
agers transition from informal-type work (held youth reported that the teacher did not “at all”
by only about 14% of employed seniors) into a help them obtain their job, and even among youth
wider range of jobs. Most working seniors are who reported their job was part of a work-study
currently employed in either restaurants (19%) or program (only 7% of employed youth), more
in fast food (17%), and about 16% work in sales than half reported that their teacher did not help
positions. The percentage of working youth in them acquire their job.
office jobs also climbs from 1% in 8th grade to
4% in 12th grade. Not surprisingly, less than 5% Of course, there are other dimensions of
of teenagers nowadays work on farms, and only employment that could make it more or less con-
0.5% deliver newspapers. These statistics suggest ducive to positive youth adjustment. For instance,
that youth today still work in a broad range of research shows that teens report higher levels of
jobs, despite the overall drop in school-year school misconduct, class cutting, substance use,
employment compared to years past. Furthermore, poor self-esteem, and depressed mood when they
the high number of youth today employed in feel that their jobs are incompatible with school,
lawn work and babysitting may be a welcome a potentially stressful experience (Barling et  al.
trend given research suggesting benefits of 1995; Mortimer et  al. 2002; Staff and Uggen
informal-t­ype work on school success. For 2003). Excessive demands and stressors at work
instance, McNeal (1997), in his longitudinal have also been linked to poor mental health and
analyses of teenagers in the 1980s (i.e., High problem behaviors in adolescence (Bachman and
School and Beyond study), found that students Schulenberg 1993; Finch et  al. 1991; Mortimer
employed in farming, lawn work, and babysitting and Staff 2004; Shanahan et  al. 1991). On the
were less likely to drop out of high school com- positive side, research shows that jobs that pro-
pared to non-working youth. Informal work is vide learning opportunities and skill develop-
often performed in family and neighborhood set- ment are highly prevalent among teen workers
tings (e.g., lawn mowing and snow shoveling, (Mortimer 2003) and that these attributes of work
babysitting), where youth are likely to be exposed provide both short- and longer-term benefits. In
adolescence, these dimensions of youth work

15  Work Intensity and Academic Success 343

have been linked to heightened intrinsic and In short, these studies suggest that the type of
extrinsic work values (Mortimer et  al. 1996b), job and the subjective quality of the work (i.e.,
improved familial relationships (Mortimer and the degree of learning opportunities, skill utiliza-
Shanahan 1994), positive mental health tion, career relevance, and work stress) are
(Shanahan et al. 1991), and reduced alcohol and important for determining the “intensity” of
illicit drug use (Staff and Uggen 2003). These work, above and beyond the average hours of
work dimensions are also positively associated work. Despite the wealth of studies assessing
with success in the labor market (i.e., career rel- how these work dimensions among teens in the
evance, pay) in the years following high school 1980s and 1990s relate to mental health, problem
(Mortimer 2003; Stern and Nakata 1989). In behaviors, and later success in the labor market,
addition, Mortimer and Staff (2004) show that we know little about whether these work dimen-
work stressors in adolescence can buffer the neg- sions matter for academic outcomes and longer-­
ative effects of adult work stressors on adult term educational attainment. Furthermore, we
health, suggesting an “inoculation effect” from know little about the quality of work among teen-
these early work experiences. agers today, and whether these work dimensions
matter for achievement and social development
Do teenage workers still find educational rel- in adolescence and during the transition to
evance and opportunities for vocational develop- adulthood.
ment in their work? Unfortunately, little research
has assessed work quality among contemporary A final component of youth employment that
cohorts. Rauscher and colleagues (2013), using a garners little attention is the duration of the
2004 sample of students in Massachusetts, found work. A long-standing critique is that teenagers
that working teenagers reported high levels of often work in jobs that are temporary or sporadic
helping opportunities and support from supervi- in duration, which in turn limits potential oppor-
sors and coworkers, as well as moderate amounts tunities for learning, skill development, adult
of learning, skill use, autonomy, and work stress. mentorship, and coworker support. Furthermore,
Similarly, in the 2004 MTF study (the last year short or sporadic work spells might be especially
these work quality questions were asked), about disruptive to academic pursuits, as teenagers
30% of employed seniors believed to a “consid- will have little time to develop effective strate-
erable” or “great extent” that: (1) their job gies to balance work and school activities. In
allowed them to use their skills and abilities; (2) contrast, those who work over long uninter-
they would learn new skills that would be useful rupted periods of time during high school may
in future work; and (3) it was an interesting job to develop effective time management strategies
do (Bachman et  al. 2005). The vast majority of that foster academic success and, since most col-
teenagers in the MTF also noted that through lege students work while attending school, carry
their jobs they got to know people from different over to postsecondary educational pursuits (Staff
social backgrounds and adults over the age of 30, and Mortimer 2007). Furthermore, youth who
experiences that would enhance work-related hold jobs over a longer duration may be more
interpersonal skills. However, the teens were likely to experience learning opportunities and
clear that these jobs did not match their long-term career potential in their early work (Mortimer
career goals. In the MTF, for instance, over 70% 2003). They may be more likely to list these jobs
of employed 12th graders indicated this job was on employment and college applications, or call
not at all what they expected to be doing for most on former employers for references. Youth with
of their lives. Similarly, Schneider and Stevenson steady work experience might signal to future
(1999), in their longitudinal study of teenagers in employers a capacity to be dependable and trust-
the 1990s, found variation in the quality of paid worthy, compared to youth with a more transient
work during adolescence, though for most youth employment history. Whereas most studies of
these early experiences in the labor market were youth work focus only on “snapshots” of their
disconnected from long-term career goals. work experience, such as the average hours of

344 J. Staff et al.

work at the time of survey administration, the work, without progressing toward self-identified
duration of these experiences can shed light on careers. These findings highlight the importance
the overall intensity of work. of examining the duration of employment along
with hours of work. Employment at lower inten-
Using longitudinal data from the Youth sity work hours may only be beneficial when
Development Study (YDS), Mortimer (2003) jobs are held for some duration.
identified important patterns in youth work
based upon the average hours of work (cumula- In summary, there is support for the proposi-
tive hours of work divided by the total weeks tion that employment can be risky for secondary
work during the school year) and the duration students. However, for the most part this risk is
(total months of work during the school year). confined to students who average long work
The most invested workers, who were about hours during the school year. Research also
25% of the sample, averaged more than 20 h per shows that work in moderation can benefit aca-
week over almost all of the sophomore, junior, demic outcomes, especially if teens follow a
and senior years of high school. Sporadic work- steady pattern of work through high school.
ers (19% of the sample) also averaged more than Furthermore, there is little support for the notion
20 h per week in their jobs but were employed that teenage work experiences are mostly of low
only half of the months. Steady and occasional quality. Teenagers work in a variety of jobs with
workers, encompassing 25% and 18% of the some offering ample opportunities for vocational
sample, respectively, averaged 20 or fewer hours development, work–school balance, and adult
per week. However, the steady workers were mentorship and support. Low-quality work, when
employed nearly all of the months of high it occurs, appears to increase the risk of problem
school, whereas the occasional workers only behaviors much like working an intense number
worked half of the months. Consistent with the of hours, though we know little about whether the
relatively higher rates of employment for this quality of work leads to school failure. However,
cohort of teens (attending high school during the as we review in the next section, early work expe-
late 1980s and early 1990s), only 7% of youth in riences, both good and bad, are shaped by
this sample reported no work experience. ascribed characteristics such as gender, race/eth-
Mortimer found that teens who averaged moder- nicity, and family socioeconomic background.
ate work hours over most of the 3 years of high Sociodemographic background factors not only
school (i.e., were “steady workers” who were predict selection into work, but also influence
employed on average 22 of the 24  months of whether work has beneficial or detrimental
observation) were the most likely to receive a effects on school outcomes.
BA/BS degree in young adulthood, even after
controlling for a variety of background factors 15.2.2 P roposition 2. Employment
(e.g., prior achievements, aspirations, family Intensity Affects School
background). Consistent with proposition 2 (see Success and Long-Term
below), this steady pattern of work was espe- Socioeconomic Attainment,
cially advantageous among youth who initially But Its Effects Are Moderated
had low educational promise, substantially by Gender, Race/Ethnicity,
increasing their likelihood of receiving a 4-year and Family Socioeconomic
college degree (Mortimer 2003; Staff and Background
Mortimer 2007). The occasional work pattern,
which shared low average weekly hours with the Stratification researchers have well documented
steady worker group but with low duration, did how ascribed characteristics such as gender,
not confer this advantage. Vuolo et  al. (2014) race/ethnicity, and family background impact
also report that those who engaged in steady academic achievement and adult occupational
work during high school were less likely to attainment. Research has also shown how these
“flounder” during the transition from school to

15  Work Intensity and Academic Success 345

early life factors shape the timing, intensity, backgrounds were more likely to work in restau-
duration, and quality of youths’ early experi- rants and especially in fast-food.
ences in work (National Reseach Council 1998).
For instance, girls tend to enter the world of work Given these sociodemographic differences in
at slightly younger ages than boys. In these first the overall “intensity” of early work experiences,
jobs, girls are more likely to be working in infor- an important question is whether work effects on
mal or “freelance” jobs, such as babysitting, school achievement and dropout are conditioned
compared to boys who are more likely to work in by gender, race/ethnicity, and socioeconomic
“employee” jobs with a more established sched- background. Such moderation is plausible for
ule and pay (U.S.  Department of Labor 2000; several reasons. It could be that intensive work
Apel et al. 2006). In addition, employed girls are experience exacerbates disadvantages associated
less likely than working boys to spend long hours with low socioeconomic background and minor-
on the job, though gender differences in average ity status. For example, if parents with lower lev-
work hours tend to disappear by the end of els of education do not socialize their children as
high school. well for school achievement as those with higher
levels of education (Lareau 2003), such youth
Research also shows that White youth are may experience more difficulty in school and
more likely than Black and Hispanic youth to thus be attracted to the diversions of work; they
hold jobs at younger ages, as well as to hold jobs may also be more susceptible to negative influ-
during the school year. In the 2014 MTF senior ences in the workplace than more advantaged
year cohort, for instance, about 65% of White teenagers. Alternatively, if working offers chal-
youth had worked at some point during the past lenges and an arena for the development of time
school year, compared to about half of Black and management and related skills, employment may
Hispanic youth. When employed, Black and provide greater benefit to disadvantaged young-
Hispanic youth were more likely to work inten- sters but make less difference for more advan-
sively during the school year (National Reseach taged teens whose prior experiences have better
Council 1998; U.S. Department of Labor 2000), prepared them to be successful in school. Heller
though these differences tend to disappear by the (2014), in a study of over 1600 disadvantaged
senior year of high school (approximately 19% high school youth in Chicago, found that random
of Black, Hispanic and White youth worked over assignment to a summer jobs program reduced
20  h per week in the 2014 MTF senior year the likelihood of violence by 43%. We speculate
cohort). Parental education and income are posi- on the bases of subgroup moderation after exam-
tively linked to the likelihood that teenagers will ining some pertinent evidence.
hold a job during the school year. Youth from
families with higher levels of education and Regarding average work hours, there is evi-
income are more likely to hold jobs at younger dence that the effect on dropout varies by popula-
ages and work limited hours compared to youth tion subgroup. D’Amico (1984) found no effect
from more disadvantaged socioeconomic of high average work hours on dropout for minor-
backgrounds. ity females and males. Among White youth, the
effect of high work hours was inconsistent: Long
Gender, race/ethnicity, and family background hours of work in the 10th grade increased the risk
also impact the types of jobs teenagers hold. In of dropout for boys and in the 11th grade for
the 2014 senior year MTF cohort, for instance, girls. Lee and Staff (2007), using propensity
girls were more likely than boys to babysit, wait score methods to control for observable differ-
tables, and work in an office. Black and Hispanic ences between students in the National Education
youth were more likely than White youth to work Longitudinal Study of 1988, similarly found that
in fast food jobs. High school seniors whose the negative impact of long hours on the job on
mothers had the highest levels of education were high school dropout was not consistent among all
most likely to work in office jobs as well as baby- students. In particular, averaging over 20  h per
sitting, whereas those from more disadvantaged week during the school year was not associated

346 J. Staff et al.

with high school dropout among students who cles, in turn, may explain why the effect of work
had an especially high probability of long work intensity is moderated by race/ethnicity and fam-
hours, such as boys, Black and Hispanic students, ily socioeconomic background. In the face of
and youth from disadvantaged backgrounds. such obstacles, those who manage to get jobs
Research by Bachman et al. (2013) and Johnson may be a more select group, better poised to
(2004) has also shown null effects of long work make the most of the developmental opportuni-
hours on educational attainment and substance ties of early work experience and less vulnerable
use among minority students and youth from low to its risks. In addition, the very difficulty of
socioeconomic backgrounds. securing employment may shape adolescents’
attitudes and behaviors. Whereas White youth
Why are long work hours not as harmful for and teenagers from high-SES family back-
minority youth or students who come from more grounds may find it easy to lose and then regain
disadvantaged backgrounds? Compared to their work, disadvantaged youth, especially if they
more advantaged counterparts, minority and low- reside in poor neighborhoods, may find that these
SES youth may be working for different reasons jobs constitute a rare opportunity to contribute to
(i.e., for family expenses or college) or be more family expenses and save for college. If jobs are
likely to need to work more hours (Newman plentiful and easy to obtain, youth may have little
1999). In 2013, the MTF study asked high school stake in their work, fostering nonchalant attitudes
seniors about how they used their earnings, and and encouraging workplace misbehaviors, such
about half reported saving at least some of their as tardiness, absenteeism, and giving away goods
earnings for future education, and about 43% and services for free.
reported using their earnings to help pay family
living expenses (groceries, housing, etc.). If jobs are harder to come by, these early expe-
However, the use of earnings varied by race and riences may instill more positive work orienta-
ethnicity, as approximately 55% of Black and tions and help foster the development of soft
Hispanic youth used their earnings for family skills, increasing the chances of later employ-
expenses, compared to 37% of White youth. It ment and promoting positive behavioral change.
also varied by parental education: Among teen- Entwisle et al. (2000), in their analyses of mostly
agers whose parents did not finish high school, poor youth residing in Baltimore, found that
approximately 59% used at least some of their early work involvement increased the skill level
earnings for family expenses. Among teenagers of the job held in later adolescence. The authors
whose parents had a college degree, only 37% speculated that early employment (e.g., during
contributed to family living expenses. Use of middle school) provided poor youth an alterna-
earnings in “non-leisure” ways (saving for future tive arena to develop their skills and increase the
education, giving money to parents, or paying for chances of future employment, especially if they
school expenses), instead of just using them for had little interest in school. In fact, these more
discretionary spending, has been linked to adult-like experiences reduced the likelihood of
improved relationships with parents (Shanahan high school dropout (Entwisle 2005) and may
et  al. 1996). Furthermore, saving at least some reduce other problem behaviors. Thus, this
earnings for college has been positively linked to greater selectivity into employment might trans-
participation in extracurricular activities, educa- late into a better job (e.g., more opportunities to
tional aspirations, grades, and long-term educa- work with adults, skill utilization, or vocational
tional attainment (Marsh 1991; Marsh and development) or being a better worker (i.e., more
Kleitman 2005; Ruscoe et al. 1996). serious about keeping a job) for disadvantaged
students, leading to greater benefits in the long
Minority students and teenagers from disad- run.
vantaged backgrounds face more obstacles when
trying to get a job, such as discrimination, trans- Staff and Mortimer (2007) have argued that
portation difficulties, and limited jobs in the local early experiences in the labor market can help
labor market (Entwisle et al. 2000). These obsta- youth from disadvantaged backgrounds establish

15  Work Intensity and Academic Success 347

strategies of time management that persist in reflect preexisting differences between students
young adulthood and facilitate higher educa- in their motivations and resources. We discuss
tional attainment. Using data from the Youth this idea in the next section.
Development Study, the authors found that youth
from more advantaged backgrounds were likely 15.2.3 P roposition 3. Employment
to pursue a steady pattern of low-intensity and Intensity Does Not Affect
high-duration work during high school, followed School Success and Long-Term
by a similar pattern of part-time work combined Socioeconomic Attainment
with schooling in the years immediately follow-
ing high school graduation. The inclusion of Almost all of the research we reviewed up to this
accumulated months of postsecondary “school point is based on observational studies, so causal
and part-time work” mediated the benefits of claims that work “affects” academic achievement
steady high school work on subsequent receipt of and social development must be tempered by the
a bachelor’s degree. By contrast, youth from dis- lack of experimental evidence. To test whether
advantaged backgrounds were likely to pursue employment intensity affects school success, the
more intensive work (high average work hours ideal study would randomly assign jobs of vary-
and high employment duration), followed by ing intensities to a sample of teens, and then
full-time work immediately after the scheduled scholars would test for significant differences in
date of high school graduation. More intensive their academic achievement and adjustment at a
workers had little likelihood of acquiring 4-year later date. To thoroughly test the time trade-off
college degrees, and they were more likely to feel hypothesis, investigators would want the treat-
they were in “career” jobs during the years fol- ment (i.e., employment) to occur during the
lowing high school (Mortimer et  al. 2008). school year when work investments would be
However, when disadvantaged low-SES youth most likely to interfere with academic pursuits.
followed a steady work pattern during high Scholars would also want to ensure that the
school, their educational attainment and longer employment assignments vary randomly in the
term wages were especially enhanced (Staff and hours of work per week as well as other impor-
Mortimer 2008). tant dimensions, such as the type and quality of
the job (the degree of learning opportunities, skill
In summary, research shows that gender, race/ utilization, stressors, etc.). The investigators
ethnicity, and family socioeconomic background might want to place restrictions on how the earn-
influence the overall intensity of early experi- ings are used (i.e., for discretionary purposes, to
ences in the labor market, such as when youth help with family expenses, etc.), as well as dis-
first enter work, their average hours and duration, courage youth from finding jobs on their own.
the type and quality of the experience, and how Ideally, the sample size would be large and
the earnings from work are used. Some evidence diverse so comparisons of treatment effects could
indicates that the academic and developmental be made within population subgroups. Of course,
benefits, as well as drawbacks, of work intensity such designs are complicated by the fact that not
depend on these ascribed characteristics. The all teens want to work (though 88% of non-­
most consistent finding is that long work hours working seniors in the 2014 MTF study wished
are least detrimental for minority youth and teen- they could work during the school year), not all
agers from lower socioeconomic backgrounds. teens would agree to work in a high intensity job
Furthermore, research also shows that early (especially if they thought it might adversely
experiences in the workplace can facilitate both affect their school performance), not all teens
educational and occupational attainment for dis- who were assigned jobs would stick with them,
advantaged youth, especially when the hours of and some teens would seek jobs on their own
work are kept low. However, some scholars despite the random assignment.
would argue that these associations are not due to
working or the intensity of the job, but instead

348 J. Staff et al.

As noted earlier, race/ethnicity and family Teenagers who place a stronger emphasis on
socioeconomic status shape who works and at work than school tend to do poorly in school,
what intensities. And despite the lack of experi- even when they are not actually working
mental evidence (the exception is Heller’s (2014) (Bachman et al. 2003; Warren 2002). Using lon-
study showing negative effects on violent crime gitudinal data from three cohorts of 8th graders
of a summer employment program in Chicago), whose educational progress was tracked until the
research clearly shows that prior achievements 12th grade, and within-person analyses to control
and orientations also precede early work inten- for unobserved time-stable selection factors,
sity. Teenagers who have little interest or success Staff and colleagues (2010b) found that youth
in school gravitate earlier to the world of work performed more poorly in school (i.e., they had
compared to students who are better students or low GPA, limited extracurricular involvement,
more invested in school (National Research low educational expectations, limited school
Council 1998). Teenagers are more likely to effort, incomplete assignments, and engaged in
spend long hours on the job if they previously more school misbehavior, truancy, and suspen-
had low educational promise (low educational sion) when they worked more than 20 h per week
aspirations, poor school performance, little compared to when they worked fewer hours or
school engagement). Early substance use, school not at all. However, youth also had poorer school
misconduct, delinquency, and arrest also increase outcomes when they were not working but
the likelihood that teenagers will average inten- wished they could spend long hours on the job. A
sive hours of work during the school year strong desire to work (measured before youth
(Mortimer 2003; Staff and Uggen 2003), and obtained jobs) also predates both intensive work
accounting for these preexisting differences hours and problem behaviors in later adoles-
between students substantially diminishes the cence. Research shows a similar risk of juvenile
effects of intensive work hours on later problem delinquency when youth actually spend long
behaviors (Apel et  al. 2007; Paternoster et  al. hours on the job and when they merely wish they
2003). These findings suggest that problem could but are not employed (Staff et al. 2010a).
behaviors, school difficulty, and failure precede
rather than follow adolescent work intensity. Together, these studies suggest that school
disengagement precedes involvement in work
Studies using data from the National Education and any observed associations between paid
Longitudinal Study of 1988 report little evidence work and school success may be spuriously
of a relationship between paid work hours and related to preexisting differences between stu-
school performance once accounting for prior dents. A few studies suggest academic risks to
differences between individuals (Schoenhals working many hours per week, however, even
et al. 1998; Warren et al. 2000). Rothstein (2007), employing some of these more stringent tech-
using data from the National Longitudinal Survey niques. In contrast to the larger body of research
of Youth (1997 cohort), initially show small neg- that largely draws on a strategy of controlling for
ative impacts of high school work experiences preexisting differences among students, includ-
(measured as the total number of hours worked ing sociodemographic factors and prior achieve-
during the school year divided by the number of ment and adjustment, in regression-based models,
weeks in the school year) on grade point average. the research reviewed in this section is still small.
These effects became statistically non-significant It has not examined the range of behavioral, atti-
when instrumental variables are used to account tudinal and achievement outcomes that we ulti-
for both time-varying and time-stable unobserved mately need to assess to answer the question of
factors. However, research by Apel et al. (2008) how work intensity and academic success are
and Tyler (2003) still find negative impacts of related. It has also not considered the patterns of
work intensity on academic achievement and employment youth engage in over time (i.e.
high school completion even when instrumental d­ uration and intensity) nor variation in the qual-
variables are used. ity of jobs.

15  Work Intensity and Academic Success 349

15.2.4 T aking Stock of the Three least some time during the school year.
Propositions Deteriorating job markets for teenage workers
have increased the scarcity of teen work. Under
The extensive literature reviewed in this chapter these conditions, work may have assumed even
indicates that moderate hours of work involve- greater importance, accounting for the nearly uni-
ment during the school year will not hurt stu- versal desire of teenagers to have jobs. Moreover,
dents’ prospects in school, either in the short- or increasing difficulties for young people in acquir-
longer-term. For many youth, work intensity, ing employment after completing their educa-
gauged by long hours on the job, does increase tions provide an altered context of school-to-work
the risk of school failure, though research sug- transition that may make it all the more impera-
gests that these negative effects tend to be weaker tive for job seekers to have work experience to be
or non-existent among disadvantaged, minority, competitive in the labor market. Employment of
and lower socioeconomic status youth. Moderate moderate intensity in adolescence, which may
hours of employment may actually facilitate bet- facilitate educational pursuits, may then doubly
ter educational outcomes, especially when teens benefit young adults in a labor market increas-
work steadily. Most adolescent work is not con- ingly rewarding educational attainment by foster-
centrated in poor quality or stressful jobs, another ing degree attainment and building an employment
form of “intensity,” though studies indicate this history with which to compete well with other
can lead to problem behaviors when it does occur. degree holders. Though, as noted above, there is
While some studies suggest that relationships some evidence that those who have less salutary
between youth work hours and school perfor- employment prospects are less likely to suffer
mance may be spurious, more research is needed from intensive employment, it is an open ques-
to examine the dynamic processes through which tion as to whether the heightened precarity of
young people select themselves (and are selected work in general makes it more or less likely that
by employers) into work and respond to more or students will benefit from early jobs.
less intense work experiences.
In addition, we see at least four unanswered
questions that scholars should consider:

15.3 Unanswered Questions 15.3.1 Question 1. Are Teenagers
for Future Research Nowadays Seeking Unpaid
Work (i.e., Volunteering,
A theme throughout this chapter is that work Internships) as a Substitute
“intensity” is almost always focused on the aver- for Paid Work, and Does This
age hours of work, but we have stressed that work Shift in Type of Work Activity
can be intense in other ways. Research must con- Matter for School
tinue to address the quality of work as well as the Achievement and Dropout?
duration of these experiences over the high school
period. Because of the dramatic decline in teen- Given the difficulty of obtaining paid work, some
age employment, there is danger that sociology contemporary teenagers may seek unpaid work—
of education scholars will dismiss work experi- volunteering and internships—to obtain work
ence as no longer relevant to the academic experience. These activities may be seen as func-
achievement and educational persistence of high tional equivalents to paid work because they get
school students. Although contemporary teenag- teenagers into a workplace, where they gain expo-
ers are less likely to have paid jobs during the sure to working adults and obtain opportunities to
school year than prior cohorts, a majority of develop the same kinds of work-relevant knowl-
seniors (59% nationally, see Table  15.1) and a edge and skills that teens acquire in the paid
substantial proportion of 10th and 8th graders employment setting. Internships can also shape
(24% and 20%, respectively) continue to work at

350 J. Staff et al.

long-term educational and career goals (Schneider While employed students in the late 1980s and
and Stevenson 1999). Indeed, there is evidence early 1990s had similar patterns of time use as
from the now three-generational Youth those who did not work, especially if they moder-
Development Study that such work may have ated their work hours (Shanahan and Flaherty
become increasingly common. Comparing YDS 2001), this may no longer be the case in the sec-
11th graders with their 16- to 17-year-old children ond decade of the twenty-first century. Do those
showed that while only 12% of the 11th graders who are employed exhibit markedly different
performed any volunteer work (in 1990), 40% of constellations of time use than their non-­
their same age children did two decades later (in employed peers, especially those who are volun-
2009 and 2010). Only 14% of the parents had vol- teering or working in internships? How does this
unteer jobs when they were in the 9th grade, but a shift from paid to unpaid work in adolescence
third of their 14- and 15-year-old children held relate to academic success and longer-term socio-
such jobs. (In contrast, 55% of the YDS parent economic attainment? Understanding the place
sample held paid jobs back in the 11th grade in of employment in the changing historical context
1990, but only 25% of their 16 and 17  year-old of youth time use will shed light on whether work
children held such jobs in 2009–2010.) MTF (both paid and unpaid) has the same meaning and
cohort data also suggest a similar pattern of declin- consequences for teenagers today.
ing teenage employment (shown in Table  15.1)
coupled with an uptick in teenage volunteering 15.3.2 Q uestion 2. Does High
over the past 20  years. For instance, approxi- Intensity Work During the
mately 25% of 8th graders and 27% of 10th and Summer Months Also
12th graders volunteered at least monthly in 1994. Compromise School
By 2014, the percentage of youth who engaged in Achievement and Increase
at least monthly volunteering had increased to the Risk of Dropout?
approximately 28% of 8th graders, 33% of 10th
graders, and 38% of 12th graders. Just as scholars The controversy surrounding youth employment
turned to the near-­universal experience of adoles- has been clearly focused on work during the
cent paid work in the 1980s, more now needs to be school year, given the potential conflict between
learned about the consequences of unpaid work. working and academic engagement. Rates of
employment and the average hours devoted to it
Because volunteer work is more discretionary among workers are, not surprisingly, higher in
than paid employment, it is arguably less likely to the summer months than in the school year
involve long work hours, that is, more than 20 h (Perreira et  al. 2007). Though scholars have
per week, and less likely to interfere with school- linked summer employment to an increase in
work. However, precisely because of its more delinquency and substance use (Apel et al. 2006),
discretionary and sporadic character, such work to our knowledge, no one has seriously ques-
may not be as conducive to the acquisition of tioned the academic value of working during the
time management skills and vocational develop- summer months, when school is not in session,
ment. Furthermore, while youth of higher socio- nor examined the impacts of such employment
economic status and White youth have long been on school-related outcomes. For many years,
more likely to acquire paid jobs, unpaid work American parents have considered the employ-
may be even more subject to selection processes. ment of their children beneficial, heightening
Volunteering and internship opportunities may be independence, teaching children the value of
less visible to disadvantaged teens and access to money, and developing work-related skills
such jobs may be more dependent on parental (Aronson et  al. 1996; Phillips and Sandstrom
and other connections. Selection of volunteers by 1990), and scholars have not questioned these
organizations may also involve less formal proce- premises. It would be useful to examine the
dures than for employment, making it more sub-
ject to unconscious bias.

15  Work Intensity and Academic Success 351

effects of work of different levels of intensity market faced by young adults who start, but do
during the summer on academic outcomes, not earn any kind of college degree (Vuolo et al.
including consideration of whether summer work 2016). In fact, on several indicators of quality of
is performed in isolation (and not during the work, associate degree recipients fare similarly to
school year) or whether it represents a continua- 4-year college graduates (but not with respect to
tion of school-year employment. In this era of income), doing better than college dropouts.
diminished employment opportunities, is work-
ing during the summer more contingent than in Public sector austerity since the 2008 Great
prior periods on already having a job during the Recession has increased college costs, especially
school year? What does this mean for the pros- at public colleges where most students attend,
pects of disadvantaged youth? Are the different and pressures on students to work more hours to
kinds of jobs youth hold during the summer and finance their educations (Presley 2013). Many
the school year (with summer jobs offering college students, in fact, work full-time while
opportunities as life guards, camp counselors, taking college coursework, a feat that severely
etc.) more or less conducive to vocational devel- compromises their ability to stay engaged in
opment and academic engagement? school and to graduate on time. Zapata-Gietl and
­colleagues (2016) document the many role con-
15.3.3 Question 3. Does High flicts and challenges experienced especially by
Intensity Work in College so-­called “non-traditional” students (older, mar-
Negatively Impact ried, parents, first-generation, etc.), who balance
Achievement and Lead families, work, and other obligations as they
to Dropout? attend 2-year colleges. Many students at 4-year
colleges and universities face similar pressures.
Whereas most scholarly and public concern
focuses on dropout from high school, college However, college students, like high school
dropout, often called attrition, has crucial conse- students, work for a variety of reasons. Some
quences for students as well as for colleges and view work as more of a social experience than a
the public bodies that often support them. means of financing their educations, blurring the
Because of the increasing income payoff for col- boundaries between work and leisure. Besen-­
lege degrees, both students and their parents rec- Cassino’s study of White college students (2014),
ognize the value of graduating from college. mostly from higher socioeconomic status fami-
Given the increasing skill demands in the labor lies, found that these students worked not for eco-
market, it is widely recognized that a high school nomic reasons, but as a way to meet people,
diploma is not adequate to secure a well-paying, socialize with friends, and express their identi-
stable job that provides the wherewithal for eco- ties, all things they found difficult to accomplish
nomic independence and a satisfactory life style. at school. And while they highly valued the aca-
demic credentials they were pursuing at univer-
In fact, more than 90% of high school seniors sity, they had little confidence they would develop
plan to go to college (Reynolds et al. 2006), and relevant job skills through schooling. Indeed the
more than 50% plan to obtain even more educa- strong social ties developed and maintained at
tion (Bachman et al. 2011a). However, just over work motivated decisions to skip class and other-
half (57%) of students entering 4-year degree wise compromise academic work so as to not let
programs actually finish their degrees within friends and co-workers down.
6  years (Knapp et  al. 2010). Students attending
2-year college programs have even lower rates of Paralleling findings for high school students,
completion (Zapata-Gietl et  al. 2016). Recent moderate work hours during the school year in col-
research, using the Youth Development Study lege does not increase the risk of drop-out, and
panel, documents the difficulties in the labor may even help students to succeed (Presley 2013).
As discussed earlier, Staff and Mortimer (2007),
analyzing Youth Development Study data, found
that students’ work patterns in high school and col-

352 J. Staff et al.

lege were similar. That is, students who had devel- ment, and long-term educational attainment.
oped a pattern of high duration-­low intensity However, little research has examined whether
“steady” work during high school pursued a simi- nativity differences within population groups
lar pattern in college, and this work pattern contrib- also influence these short- and longer-term
uted to their acquisition of 4-year college degrees. associations.
However, working more than 20 h per week during
the first year of college increases the likelihood that Nearly two decades ago the National Research
students will leave early (Bozick 2007). Still, even Council ( 1998) concluded in their report on pro-
intensive work during postsecondary education tecting youth at work that “very little data on
may be conducive to vocational development and work among immigrants, in general, and immi-
success if it is related to fields of study or serves as grant youth, in particular, have been collected”
a steppingstone to occupational goals. Full-time (p.  52). Since this influential report, some
employment does not lower the likelihood of BA research has shown how nativity shapes early
completion, though it is associated with lower participation in work. For instance, 73% of
grades (Hamilton 2013). Some early evidence Hmong parents of refugee children in the Youth
(Tinto 1987; Ehrenberg and Sherman 1987) sug- Development Study did not want their offspring
gested that students who held jobs on campus were to do paid work (McNall et al. 1994), while the
less likely to drop out than those whose jobs were parents of other YDS children were near unani-
found off campus, and more recent research mous in their approval of their children’s employ-
(Presley 2013) confirms this observation. ment (Phillips and Sandstrom 1990). Kofman
On-campus employment, even if not educationally and Bianchi (2012), using data from the 2003 to
related (e.g., working in a school cafeteria) ties the 2010 American Time Use Survey, found that
student to the college in a way that off-c­ ampus jobs teenagers ages 15–17  in immigrant households
do not. Especially valuable in integrating the stu- (i.e., who were foreign born or who were residing
dent into academic life is employment that involves with at least one parent who was foreign born)
interactions between students and faculty. from Latin America or Asia spent less time in
Educationally and vocationally-r­elevant college paid work compared to native-born youths.
student employment is more likely in majors or Oropesa and Landale (2009) similarly showed
course sequences that prepare students for a par- nativity differences in employment among 16-
ticular type of work, as in community college pro- and 17-year-old Mexican youth in the 2000 cen-
grams that lead to a vocationally relevant associate sus. They found that native-born Mexican
degree or occupational certification. More research teenagers were more likely to be working while
is needed on how work intensity during college, in school compared to foreign-born Mexican
broadly construed as we have done here, influences youth. Perreira et  al. (2007), using longitudinal
college persistence and degree attainment, and the data from the Add Health study, found that ado-
extent to which such effects vary by the type of lescents of first- or second-generation immigrant
college or curricular program. parents spent less time employed while attending
high school and during the summer months com-
15.3.4 Question 4. How Does Work pared to native born youth.
Intensity Relate to Academic
Success Among Immigrant Why are youth in immigrant households less
Youth in the United States? likely to be employed compared to native-born
teenagers? One reason is that informal forms of
As reviewed previously, research has documented employment, such as playing music at parties,
how gender, socioeconomic background, and working as a D.J., or participating in other jobs
race/ethnicity moderate the effect of high work that are “off the books” may be more common
intensity on school success, positive youth adjust- among immigrant youth compared to youth with
native-born parents (Kasinitz et  al. 2008).
Moreover, Kasinitz and colleagues reported that
children of immigrants in New  York City often

15  Work Intensity and Academic Success 353

viewed work in businesses owned by their family nationally in 2014, see Table  15.1) could lead
or relatives as something they did to “help out” those who truly need to work, to support them-
rather than as more formal paid jobs. Additionally, selves or their families, to drop out of school in
analyses of time use data from the American order to maintain their hours of work. Hours
Time Use Survey reveals that immigrant youth restrictions based on age (for example, applying
spend more time studying than their native-born to all students under 19), meant to encourage stu-
counterparts (50  min per day versus 38  min, dents to stay in school and away from employ-
respectively), so children in immigrant house- ment, could cause widespread hardship for
holds may be discouraged from holding a job financially independent young people.
during the school year out of concern that it could
compromise academic pursuits (Kofman and Nonetheless, a number of steps could be taken,
Bianchi 2012). On average, Hmong youth in the short of prohibition, which might diminish the
Youth Development Study spent 19–26  h per likelihood that high school students work inten-
week on homework during the 4  years of high sively and thus jeopardize their academic attain-
school; other youth spent between 7 and 8  h ments. First of all, attempts could be made to
(McNall et al. 1994). Finally, it is plausible that spread the word—to teachers, counselors, par-
some youth from immigrant households, espe- ents, and employers—about the potential dangers
cially if any of the family members are undocu- of high intensity work. Rather than deflecting
mented immigrants, may also think their own students from all employment (a message likely
work is illegal and may be reluctant to report to be widely rejected given the demonstrated,
their previous experiences in the labor market to near universal, preference of students to work),
researchers, thus leading to an undercount of those who guide high school students should
actual employment experiences among immi- steer them toward work that will be conducive to
grant youth. However, we are not aware of stud- their educational and vocational development.
ies that have considered whether nativity is a That is, moderate work that provides learning
moderator of work intensity and academic suc- opportunities and whose hours, conditions, and
cess in adolescence. Echoing the National task requirements do not interfere with school. To
Research Council report, we encourage research- enhance the compatibility of paid work and
ers to pursue this line of inquiry in future work. schoolwork, educators could encourage students
to discuss their experiences on the job, the knowl-
15.4 P olicy Implications edge and skills that enable them to be effective in
their jobs, and how curricular offerings in the
In the absence of causal evidence, recommenda- school might enhance such human capital.
tions for shifts in policy are hazardous. While Sharing information about work would also serve
mounting evidence suggests that employed teen- to spread information to students about opportu-
agers are subject to higher risks of school diffi- nities for employment, as well as the most desir-
culty and dropout when they work more than able employers in their communities.
20 h per week, it is a considerable leap to con-
clude that school-age youth should be discour- Another potentially useful strategy to engage
aged or even prevented from working more than students in the educational enterprise and to
20 h per week. Uncertain, rapidly changing, and reduce dropout is to build connections between
increasingly diverse contexts make such conclu- school and work. Among modern countries, the
sions especially tenuous, and shifts in the regula- United States is distinctive in its lack of institu-
tory environment may have unforeseen and tional bridges from school to jobs. Unlike
deleterious consequences. For example, prohibit- Germany, Austria, and the Netherlands, with
ing high school students from working more than their apprenticeship systems, and Japan, with
20  h a week (approximately 20% of seniors linkages between high schools and employers,
school-leavers in the United States must rely on
their own contacts and, if they are among the for-
tunate who graduate from college, college career

354 J. Staff et al.

services and placement offices. While almost all sity, particularly the number of hours students
high school students hope and plan to enter a 2- spend working for pay each week. To a lesser
or 4-year college, approximately a third of con- extent, research has also considered whether
temporary cohorts do not do so. If little is offered alternative ways in which work is “intense,” such
with vocational relevance in high schools, the as working in lower-quality or stressful jobs,
non-college bound may see little reason to invest undermines educational success. The larger pic-
in, or even stay in, school whether they are ture that comes out of this research is that
employed intensively during high school or not. employment may have detrimental academic
consequences, including lowered performance
In addition to building more and better voca- and higher dropout, but only when students work
tional curricula, high schools might provide other an excessive number of hours per week or in jobs
opportunities and services that would enhance they report as of problematic quality. Neither of
students’ vocational development. They might those situations has characterized the majority of
offer occupationally-relevant after-school pro- teens’ work experiences, even at the height of
grams, such as the highly successful After School trends in teen employment in the 1980s.
Matters program in Chicago (Hirsch 2015), Moreover, at least with respect to work hours, a
which supports sustained contact and joint proj- growing body of evidence documents that minor-
ects involving students and professionals in the ity and lower socioeconomic status students do
community; develop connections between not experience the detrimental academic conse-
employers and teachers (Rosenbaum et al. 1999), quences of working intensely. And finally, recent
who may be able to guide youth toward good jobs studies using the most stringent models available
that promote work readiness; encourage students cast some doubt on whether there are any conse-
to enroll in dual-enrollment courses, which pro- quences to employment and work intensity at all.
vide opportunities to gain work experience while Whether and how much work teenagers engage
earning course credits (Schneider et  al. 2015); in, along with their academic engagement, per-
arrange job shadowing experiences for high formance, achievement, and persistence, may
school students; and promote internships, includ- both be driven by preexisting differences among
ing monetary supplement for those who would students.
otherwise not be able to participate in these, often
unpaid, work experiences. Such innovative meth- As we look to the future, there are excellent
ods of steering youth toward high-quality work opportunities for scholars of education to
experiences are becoming ever more popular, engage these issues further and deepen our
perhaps in response to the decline in labor market knowledge about the nexus of work and school-
demand for youth. Work that is connected to, and ing. Recent efforts to address causality are lim-
preferably monitored by, the schools may enable ited in scope to date and have been primarily
teenagers to have work experiences that are more directed toward examining the impacts of work
beneficial to them in the long run than previously hours. In addition, assessments of patterns of
typical youth jobs. Such work experiences may employment that examine investments in work
convince contemporary teens that staying in high over time are rare. Strategies of employment
school until completion and pursuing higher edu- over the adolescent years, as well as the balance
cation will have substantial payoffs in the labor during school and summer months of the year,
market. need further investigation. In addition, research
on employment during tertiary education,
15.5 Concluding Comments including that which attends to continuity and
discontinuity compared to the high school years,
The possibility that employment during the would be desirable. Finally, we need to under-
school year poses academic risks to secondary stand whether work is related to academic suc-
students has fueled much research on work inten- cess similarly or differently among immigrant
and native born youth.

15  Work Intensity and Academic Success 355

The fact that participation in paid work during Bachman, J. G., Johnston, L. D., & O’Malley, P. M. (2005).
adolescence has declined over the past several
decades does not make it any less important to Monitoring the future: Questionnaire responses from
study. Employment remains a key domain of ado- the nation’s high school seniors, 2004. Ann Arbor:
lescence and critical questions about how it Institute for Social Research.
relates to education remain, as noted above. In Bachman, J.  G., Johnston, L.  D., & O’Malley, P.  M.
addition, the nature and context of adolescent (2011a). Monitoring the future: Questionnaire
employment is changing, and as such, we need to responses from the nation’s high school seniors, 2010.
know whether its impact is changing as well. It is Ann Arbor: Institute for Social Research.
important that we learn how paid employment Bachman, J. G., Staff, J., O’Malley, P. M., Schulenberg,
compares to unpaid work experiences so that we J. E., & Freedman-Doan, P. (2011b). Twelfth-grade
can assess the educational implications of this student work intensity linked to later educational
historical shift. The extent to which time spent at attainment and substance use: New longitudi-
work is deemed a threat or complement to educa- nal evidence. Developmental Psychology, 47(2),
tional pursuits may depend on the impacts of 344–363.
alternative uses of time, including volunteering Bachman, J. G., Staff, J., O’Malley, P. M., & Freedman-­
and unpaid interning. Doan, P. (2013). Adolescent work intensity, school
performance, and substance use: Links vary by ­race/
References ethnicity and socioeconomic status. Developmental
Psychology, 49(11), 2125–2134.
Alon, S. (2009). The evolution of class inequality in Barling, J., Rogers, K.  A., & Kelloway, E.  K. (1995).
higher education: Competition, exclusion, and adapta- Some effects of teenagers’ part-time employment:
tion. American Sociological Review, 74, 731–755. The quantity and quality of work make the difference.
Journal of Organizational Behavior, 16, 143–154.
Apel, R., Paternoster, R., Bushway, S.  D., & Brame, R. Besen-Cassino, Y. (2014). Consuming work: Youth labor
(2006). A job isn’t just a job: The differential impact in America. Philadelphia: Temple University Press.
of formal versus informal work on adolescent problem Bozick, R. (2007). Making it through the first year of
behavior. Crime & Delinquency, 52, 333–369. college: The role of students’ economic resources,
employment, and living arrangements. Sociology of
Apel, R., Bushway, S.  D., Brame, R., Haviland, A.  M., Education, 80(3), 261–285.
Nagin, D.  S., & Paternoster, R. (2007). Unpacking Call, K. T., & Mortimer, J. T. (2001). Arenas of comfort
the relationship between adolescent employment and in adolescence: A study of adjustment in context.
antisocial behavior: A matched samples comparison. Mahwah: Erlbaum.
Criminology, 45(1), 67–97. Carr, R., Wright, J.  D., & Brody, C. (1996). Effects of
high school work experience a decade later: Evidence
Apel, R., Bushway, S. D., Paternoster, R., Brame, R., & from the National Longitudinal Survey. Sociology of
Sweeten, G. (2008). Using state child labor laws to Education, 69(1), 66–82.
identify the causal effect of youth employment on Coleman, J.  S., Bremner, R.  H., Clark, B.  R., Davis,
deviant behavior and academic achievement. Journal J. B., Eichorn, D. H., et al. (1974). Youth: Transition
of Quantitative Criminology, 24, 337–362. to adulthood. Report of the panel on youth of the
President’s science advisory committee. University of
Aronson, P.  J., Mortimer, J.  T., Zierman, C., & Hacker, Chicago Press.
M. (1996). Generational differences in early work D’Amico, R. (1984). Does employment during high
experiences and evaluations. In J.  T. Mortimer & school impair academic progress? Sociology of
M. D. Finch (Eds.), Adolescents, work, and family: An Education, 57(3), 152–164.
intergenerational developmental analysis (pp. 25–62). Ehrenberg, R., & Sherman, D. (1987). Employment while
Newbury Park: Sage. in college, academic achievement, and post college
outcomes: A survey of results. Journal of Human
Bachman, J. G., & Schulenberg, J. (1993). How part-time Resources, 22, 1–21.
work intensity relates to drug use, problem behav- Entwisle, D. R. (2005). Urban teenagers: Work and drop-
ior, time use, and satisfaction among high school out. Youth & Society, 37(1), 3–32.
seniors: Are these consequences or merely correlates? Entwisle, D. R., Alexander, K. L., & Olson, L. S. (2000).
Developmental Psychology, 29(2), 220–235. Early work histories of urban youth. American
Sociological Review, 65(2), 279.
Bachman, J.  G., Saffron, D.  J., Sy, S., & Schulenberg, Finch, M.  D., Shanahan, M.  J., Mortimer, J.  T., & Ryu,
J.  (2003). Wishing to work: New perspectives on S. (1991). Work experience and control orientation
how adolescents’ part-time work intensity is linked in adolescence. American Sociological Review, 56,
to educational disengagement, substance use, and 597–611.
other problem behaviors. International Journal of Greenberger, E., & Steinberg, L. D. (1986). When teenag-
Behavioral Development, 27(4), 301–315.
ers work: The psychological and social costs of teen-
age employment. New York: Basic Books.

356 J. Staff et al.

Hamilton, L. (2013). More is more or more is less: ing, subversion of academic goals, or a threshold?
Parental financial investments during college. American Educational Research Journal, 42(2),
American Sociological Review, 78, 70–95. 331–369.
McNall, M., Dunnigan, T., & Mortimer, J.  T. (1994).
Hansen, D.  M., & Jarvis, P.  A. (2000). Adolescent The educational achievement of the St. Paul Hmong.
employment and psychosocial outcomes: A compari- Anthropology and Education Quarterly, 25, 44–65.
son of two employment contexts. Youth & Society, 31, McNeal, R. B., Jr. (1997). Are students being pulled out of
417–436. high school? The effect of adolescent employment on
dropping out. Sociology of Education, 70(3), 206–220.
Hansen, D.  M., Mortimer, J.  T., & Kruger, H. (2001). Mihalic, S. W., & Elliott, D. (1997). Short- and long-term
Adolescent part-time employment in the U.S. and consequences of adolescent work. Youth & Society,
Germany: Diverse outcomes, contexts, and pathways. 28(4), 464–498.
In C.  Pole, P.  Mizen, & A.  Bolton (Eds.), Hidden Monahan, K.  C., Lee, J.  M., & Steinberg, L. (2011).
Revisiting the impact of part-time work on adolescent
hands: International perspectives on children’s work adjustment: Distinguishing between selection and
and labour (pp. 121–138). London: Routledge Falmer socialization using propensity score matching. Child
Press. Development, 82(1), 96–112.
Heller, S. B. (2014). Summer jobs reduce violence among Mortimer, J.  T. (2003). Working and growing up in
disadvantaged youth. Science, 346(6214), 1219–1223. America. Cambridge: Harvard University Press.
Hirsch, B. J. (2015). Job skills and minority youth: New Mortimer, J.  T., & Johnson, M.  K. (1998). Adolescent
program directions. New York: Cambridge University part-time work and educational achievement. In
Press. K.  Borman & B.  Schneider (Eds.), The Adolescent
Hirschman, C., & Voloshin, I. (2007). The structure of
teenage employment: Social background and the years: Social influences and educational challenges
jobs held by high school seniors. Research in Social (pp.  183–206). Chicago: National Society for the
Stratification and Mobility, 25(3), 189–203. Study of Education.
Johnson, M.  K. (2004). Further evidence on adolescent Mortimer, J.  T., & Shanahan, M.  J. (1994). Adolescent
employment and substance use: Differences by race work experience and family relationships. Work and
and ethnicity. Journal of Health and Social Behavior, Occupations, 21, 369–384.
45, 187–197. Mortimer, J. T., & Staff, J. (2004). Early work as a source
Kalenkoski, C.  M., & Pabilonia, S.  W. (2012). Time to of developmental discontinuity during the transition to
work or time to play: The effect of student employ- adulthood. Development and Psychopathology, 16(4),
ment on homework, sleep, and screen time. Labour 1047–1070.
Economics, 19, 211–221. Mortimer, J. T., Finch, M. D., Ryu, S., Shanahan, M. J.,
Kasinitz, P., Mollenkopf, J.  H., Waters, M.  C., & & Call, K. T. (1996a). The effect of work intensity on
Holdaway, J. (2008). Inheriting the city: The children adolescent mental health, achievement, and behavioral
of immigrants come of age. Harvard University Press. adjustment: New evidence from a prospective study.
Kerckhoff, A. (2000). Transition from school to work Child Development, 67(3), 1243–1261.
in comparative perspective. In M.  T. Hallinan (Ed.), Mortimer, J.  T., Pimentel, E.  E., Ryu, S., Nash, K., &
Handbook of the sociology of education (pp. 453–474). Lee, C. (1996b). Part-time work and occupational
New York: Kluwer Academic/Plenum Publishers. value formation in adolescence. Social Forces, 74(4),
Knapp, L. G., Kelly-Reid, J. E., & Ginder, S. A. (2010). 1404–1418.
Mortimer, J. T., Harley, C., & Staff, J. (2002). The qual-
Enrollment in postsecondary institutions, Fall 2008; ity of work and youth mental health. Work and
Occupations, 29(2), 166–197.
Graduation rates, 2002 & 2005 cohorts; and finan- Mortimer, J.  T., Vuolo, M., Staff, J., Wakefield, S., &
cial statistics, fiscal year 2008 (NCES 2010-152). Xie, W. (2008). Tracing the timing of “career” acqui-
U.S.  Department of Education. Washington, DC.: sition in a contemporary youth cohort. Work and
National Center for Education Statistics. Occupations, 35, 44–84.
Kofman, Y., & Bianchi, S. M. (2012). Time use of youths National Reseach Council. (1998). Protecting youth
by immigrant and native-born parents: ATUS results.
Monthly Labor Review, 135, 3–24. at work: Health, safety, and development of work-
Lareau, A. (2003). Unequal childhoods: Class, race, and ing children and adolescents in the United States.
family life. Berkley: University of California Press. Washington, DC: The National Academies Press.
Lee, J.  C., & Staff, J.  (2007). When work matters: The Newman, K. S. (1999). No shame in my game. New York:
varying impact of work intensity on high school drop- Knopf and Russell Sage Foundation.
out. Sociology of Education, 80(2), 158–178. Oropesa, R. S., & Landale, N. S. (2009). Why do immi-
Light, A. (2001). In school work experience and the grant youths who never enroll in U.S. schools matter?
returns to schooling. Journal of Labor Economics, School enrollment among Mexicans and non-Hispanic
19(1), 65–93. Whites. Sociology of Education, 82(3), 240–266.
Marsh, H.  W. (1991). Employment during high school: Paternoster, R., Bushway, S., Brame, R., & Apel, R.
Character building or a subversion of academic goals? (2003). The effect of teenage employment on delin-
Sociology of Education, 64, 172–189.
Marsh, H.  W., & Kleitman, S. (2005). Consequences of
employment during high school: Character build-

15  Work Intensity and Academic Success 357

quency and problem behaviors. Social Forces, 82(1), Staff, J., & Mortimer J.T. (2008). Social class background
297–335. and the “school to work” transition. New Directions
Perreira, K.  M., Harris, K.  M., & Lee, D. (2007). for Child and Adolescent Development 119, 55–69.
Immigrant youth in the labor market. Work and
Occupations, 34, 5–34. Staff, J., & Mortimer, J.  T. (2007). Educational and
Phillips, S., & Sandstrom, K. (1990). Parental attitudes work strategies from adolescence to early adulthood:
toward “youthwork”. Youth and Society, 22, 160–183. Consequences for educational attainment. Social
Presley, C. (2013). Advising and engaging the “working-­ Forces, 85(3), 1169–1194.
class” college student. The Mentor. An Academic
Advising Journal (available online only). Retrieved Staff, J., & Uggen, C. (2003). The fruits of good work:
January 9, 2016 at https://dus.psu.edu/mentor/2013/11/ Early work experiences and adolescent deviance.
advising-and-engaging-the-%E2%80%9Cworking- Journal of Research in Crime and Delinquency, 40(3),
class%E2%80%9D-college-student/ 263–290.
Rauscher, K. J., Wegman, D. H., Wooding, J., Davis, L.,
& Junkin, R. (2013). Adolescent work quality: A view Staff, J., Osgood, D.  W., Schulenberg, J.  E., Bachman,
from today’s youth. Journal of Adolescent Research, J.  G., & Messersmith, E.  E. (2010a). Explaining the
28(5), 557–590. relationship between employment and juvenile delin-
Reynolds, J., Stewart, M., Sischo, L., & MacDonald, R. quency. Criminology, 48(4), 1101–1131.
(2006). Have adolescents become too ambitious? High
school seniors’ educational and occupational plans, Staff, J., Schulenberg, J. E., & Bachman, J. G. (2010b).
1976 to 2000. Social Problems, 53, 186–2006. Adolescent work intensity, school performance, and
Rosenbaum, J.  E., DeLuca, S., Miller, S.  R., & Roy, K. academic engagement. Sociology of Education, 83(3),
(1999). Pathways into work: Short- and long-term 183–200.
effects of personal and institutional ties. Sociology of
Education, 72, 179–196. Staff, J., Johnson, M.  K., Patrick, M., & Schulenberg,
Rothstein, D.  S. (2007). High school employment and J. (2014). The Great Recession and recent employment
youths’ academic achievement. The Journal of Human trends among secondary students in the United States.
Resources, 42(1), 194–213. Longitudinal and Life Course Studies, 5(2), 173–188.
Ruscoe, G., Morgan, J. C., & Peebles, C. (1996). Students
who work. Adolescence, 31(123), 625–632. Staff, J., Mont’Alvao, A., & Mortimer, J.  T. (2015).
Schneider, B., & Stevenson, D. (1999). The ambitious Children at work. In M. H. Bornstein & T. Leventhal
(Eds.), Handbook of child psychology and develop-
generation: America’s teenagers, motivated but direc-
tionless. New Haven: Yale University Press. mental science: Ecological settings and processes
Schneider, B., Saw, G., & Broda, M. (2015). Work and (Vol. 4, pp. 345–374). Hoboken: Wiley.
work migration within and across countries in emerg- Stern, D., & Nakata, Y. F. (1989). Characteristics of high
ing and young adulthood. In L.  A. Jensen (Ed.), school students’ paid jobs, and employment experi-
ence after graduation. In D. Stern & D. Eichorn (Eds.),
Oxford handbook of human development and cul-
ture: An interdisciplinary perspective (pp.  554–569). Adolescence and work: Influences of social structure,
Oxford University Press: Oxford. labor markets, and culture (pp.  189–234). Hillsdale:
Schoenhals, M., Tienda, M., & Schneider, B. (1998). The Lawrence Erlbaum.
educational and personal consequences of adolescent Tinto, V. (1987). Leaving college: Rethinking the causes
employment. Social Forces, 77(2), 723–761. and cures of student attrition. Chicago: University of
Sewell, W. H., & Hauser, R. M. (1975). Education, occu- Chicago Press.
pation, and earnings. Achievement in the early career. Tyler, J. H. (2003). Using state child labor laws to identify
New York: Academic. the effect of school-year work on high school achieve-
Shanahan, M., & Flaherty, B. (2001). Dynamic patterns ment. Journal of Labor Economics, 21(2), 381–408.
of time use in adolescence. Child Development, 72(2), U.S.  Department of Education. (2015). High school
385–401. dropouts and stopouts: Demographic backgrounds,
Shanahan, M.  J., Finch, M.  D., Mortimer, J.  T., & Ryu, academic experiences, engagement, and school char-
S. (1991). Adolescent work experience and depressive acteristics. National Center for Education Statistics,
affect. Social Psychology Quarterly, 54, 299–317. 2015–2064.
Shanahan, M.  J., Elder, G.  H., Jr., Burchinal, M., & U.S.  Department of Labor. (2000). Report on the youth
Conger, R. D. (1996). Adolescent paid labor and rela- labor force. Washington, DC: U.S.  Government
tionships with parents: Early work–family linkages. Printing Office.
Child Development, 67, 2183–2200. Vuolo, M., Mortimer, J. T., & Staff, J. (2014). Adolescent
Smith, C. L. (2011). Polarization, immigration, education: precursors of pathways from school to work. Journal
of Research on Adolescence, 24, 145–162.
What’s behind the dramatic decline in youth employ- Vuolo, M., Mortimer, J. T., & Staff, J. (2016). The value
of educational degrees in turbulent economic times:
ment? Division of research and statistics and mon- Evidence from the Youth Development Study. Social
etary affairs (Finance and economics discussion series Science Research, 57, 233–252.
2011–2041). Washington, DC: Federal Reserve Board. Warren, J.  R. (2002). Reconsidering the relationship
between student employment and academic outcomes:
A new theory and better data. Youth & Society, 33(3),
366–393.

358 J. Staff et al.

Warren, J. R., & Lee, J. C. (2003). The impact of adoles- Zapata-Gietl, C., Rosenbaum, J., Ahearn, C., & Becker,
cent employment on high school dropout: Differences K.  I. (2016). College for all: New institutional
by individual and labor-market characteristics. Social conflicts in the transition to adulthood. In M.  J.
Science Research, 32, 98–128. Shanahan, J. T. Mortimer, & M. K. Johnson (Eds.),
Handbook of the life course, volume 2. New  York:
Warren, J.  R., LePore, P.  C., & Mare, R.  D. (2000). Springer.
Employment during high school: Consequences for
students’ grades in academic courses. American
Educational Research Journal, 37, 943–969.

Part IV
Educational Opportunities and the

Transition into Adulthood

Students’ Educational Pathways: 16
Aspirations, Decisions,
and Constrained Choices Along
the Education Lifecourse

Michal Kurlaender and Jacob Hibel

Abstract Educational pathways are marked by a series of
Educational pathways are marked by a series choices that individuals and their families make
of choices that individuals and their families that shape students’ development and educational
make that shape students’ development and destinations. Viewed from an individualistic per-
educational destinations. The education attain- spective, families invest time and resources in
ment model is defined by a notable tension children’s educational development early in
between individual choice and structural con- childhood. These investments are then comple-
straints that exist throughout the life course. mented and augmented by individuals’ own deci-
This chapter synthesizes research on the con- sions about how and where to invest their time
strained choices that typify educational path- and energies as they progress through school.
ways from early childhood to adulthood in the Although this individualistic view of education is
U.S. We focus on several areas in the literature represented throughout the sociology of educa-
in which the tension between individual tion, it is perhaps more closely associated with
choice and structural constraints plays out, cognate disciplines such as economics and psy-
specifically: educational aspirations, curricu- chology. Sociologists, rather, often take a more
lar differentiation, and informational barriers structural view of education, emphasizing the
and opportunities. Within each of these inter- ways choices are constrained by multiple forces
connected areas we describe the dominant and institutions sorting youth among unequal
theories that buttress the individual determi- pathways of educational opportunity, which
nants model, and the structural or institutional results in perpetuating social inequalities. The
forces that shape the educational attainment notion of constrained choice suggests an impor-
process. We also review policy trends that tant interplay between structural forces and indi-
have emerged over the past several decades vidual decision-making, which we argue
designed to attenuate structural inequalities in ultimately shapes students’ educational
students’ educational pathways. pathways.

We thank Thad Domina and Sherrie Reed for feedback on A “pathway” denotes a course individuals
an earlier draft of this paper, and Elizabeth Zeiger embark on; one in which social structures can
Friedmann and Jake Jackson for research assistance. constrain and define individual choice. Just as
pedestrians typically follow pre-defined paths
M. Kurlaender (*) · J. Hibel rather than blazing their own trails, students typi-
University of California, Davis, CA, USA cally move through pre-defined positions in edu-
e-mail: [email protected]; [email protected] cational institutions. However, students—like

© Springer International Publishing AG, part of Springer Nature 2018 361
B. Schneider (ed.), Handbook of the Sociology of Education in the 21st Century, Handbooks
of Sociology and Social Research, https://doi.org/10.1007/978-3-319-76694-2_16

362 M. Kurlaender and J. Hibel

pedestrians—can choose among multiple com- economically productive, creating a positive
peting paths and these choices have important association between educational attainment and
implications for their developmental, educa- earnings. In light of this well-documented corre-
tional, and socioeconomic destinations. lation, individuals seek to acquire as much edu-
Moreover, these choices are made within differ- cation as they can afford as a means of securing
ent types and “levels” of social structures—some higher earnings and status in adulthood.
more explicit or visible than others (Hays 1994;
Sewell 1992). Social scientists have produced multiple cri-
tiques of the rational choice explanation for the
This chapter synthesizes research on the con- link between educational attainment and earn-
strained choices that typify educational pathways ings, including credentialing theory (e.g., Collins
from early childhood to adulthood in the U.S. We 1979; Labaree 1997), screening or signaling the-
have organized the review by focusing on several ories (e.g., Rosenbaum and Binder 1997; Spence
areas in the literature in which the tension 1973; Stiglitz 1975), and conflict theories in the
between individual choice and structural con- Marxian (e.g., Bowles and Gintis 1976), Weberian
straints plays out, specifically: (1) educational (e.g., Collins 1971), and Bourdieuian (e.g.,
aspirations; (2) curricular differentiation; and (3) Bourdieu and Passeron 1977) traditions. Each of
informational barriers and opportunities. Within these perspectives suggests that the structure of
each of these interconnected areas we describe U.S. society and its central institutions leads indi-
the dominant theories that buttress the individual viduals onto educational pathways that are deter-
determinants model, and the structural or institu- mined by factors beyond straightforward
tional forces that shape the educational attain- cost-benefit analyses of potential educational
ment process. Finally, we review policy trends investments. While other chapters in this volume
that have emerged over the past several decades explore the implications of these theoretical per-
designed to attenuate structural inequalities in spectives in greater depth than the present chap-
students’ educational pathways. ter, we note that, regardless of the framework one
uses to understand the opportunities and con-
16.1 Educational Aspirations straints facing students as they navigate formal
in a College for All Era schooling transitions, a guiding principle of the
U.S. schooling structure’s design—both explicit
Educational pathways in the U.S. are now defined through compensatory policies and implicit
by a ubiquitous “college for all” ethos that domi- through the pervasive college for all ethos—is
nates individual students’ dialogues about their individual choice within open access pathways.
educational pathways and policy efforts aimed at Whether wholly realistic or not, this message has
reducing structural barriers to postsecondary clearly been communicated to young people in
schooling. This is most evident in discussions the U.S. Students perceive that they possess sub-
around a fundamental notion of choice—stu- stantial agency with respect to their educational
dents’ educational aspirations. futures, and their reported attainment expecta-
tions reveal that they generally intend to exercise
Rational choice or human capital perspectives this self-determination by obtaining degrees
suggest that an individual’s decision to invest in beyond the high school diploma.
education is based on an interaction of tastes,
abilities, and resources. With roots in neoclassi- Today’s youth have registered the college
cal economic theory, these perspectives rest on refrain. A majority of middle and high school stu-
the central assumption that individual actors dents, regardless of their academic performance,
seek, above all, to maximize their economic report that they will attend college (Jacob and
interests. According to this line of thought, the Wilder-Linkow 2011; Goyette 2008; Reynolds
knowledge and skills (i.e., human capital) and Pemberton 2001; Schneider and Stevenson
acquired through schooling make workers more 2000). The nearly universal orientation towards
college represents incredible growth in

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 363

­educational expectations (Jacob and Wilder- ful predictor of students’ ultimate educational
Linkow 2011; Goyette 2008; Reynolds and attainment (Alexander and Cook 1979; Kao and
Pemberton 2001). Over the past several decades Tienda 1998; Rosenbaum 2001).
the percentage of 10th graders with college
degree expectations has doubled, and has nearly In contrast to earlier periods, academic perfor-
doubled among 12th graders (Fig.  16.1). mance currently accounts for little of the variance
However, college degree attainment has not kept in students’ expected levels of educational attain-
up with the increased educational expectations ment. Reynolds et  al. (2006) find that between
present among today’s youth (Fig. 16.1). 1976 and 2000, the percentage of high school
seniors indicating that they probably or definitely
16.1.1 Aspirations and Expectations would complete at least a baccalaureate degree
as Determinants increased from 50% to 78%. At the same time,
of Educational Attainment the explanatory power of self-reported grades
and participation in a college preparatory pro-
The role of educational aspirations and expecta- gram for predicting high school students’ attain-
tions in the education and status attainment pro- ment expectations declined appreciably
cesses has been intensely debated in recent years. (Reynolds et al. 2006).
This topic captures one of sociology’s longstand-
ing debates over the role of educational aspira- Recent work, however, suggests that educa-
tions as a mediator of structural determinants of tional expectations remain a key determinant of
adult status (Sewell et  al. 1969; Sewell and later educational success, and of students’ atti-
Hauser 1975). As educational aspirations have tudes and behavior in high school (Domina et al.
become more uniform over time—a remarkable 2011). In their article linking educational expec-
93% of all seniors in the most recent large-scale tations to effort, Domina et al. (2011) test whether
national survey (ELS) report that they planned to students’ college expectations influence the
continue their education after high school—some importance they place on high school mathemat-
sociologists of education have raised questions ics. They find that “educational expectations have
about the relevance of aspirations as a meaning- a positive causal effect on student perceptions
regarding the importance of high school academ-
ics for their future success” (p. 101), and that this
relationship holds across the achievement

Percent 90
80
70 1970 1975 1980 1985 1990 1995 2000 2005 2010
60
50
40
30
20
10

0
1965

10th Grade BA Expectations (Goyette 2008; Reynolds & Pemberton 2001)
12th Grade Expectations (Jacob & Wilder-Linkow 2011)

BA Attainment, 25- to 29-year-olds (CPS)

Fig. 16.1  BA expectations and attainment, 1970–2008

364 M. Kurlaender and J. Hibel

­distribution, albeit attenuated for students at the ries (Cohen et  al. 2009). Research on

lower end. Their findings challenge Rosenbaum’s postsecondary STEM pathways illustrates the

longstanding critique of the false promise of fundamental importance of self-efficacy for edu-

expectations in the college for all ethos, namely, cational success, particularly for sub-groups his-

that students believe that college opportunities torically underrepresented in these fields. For

are available irrespective of their performance in example, researchers have established that the

high school, and as such their expectations are a under-participation of women in STEM majors is

weak predictor of their effort or attainment a function of disparities in interest in and affect

(Rosenbaum 2001, 2011). towards math/science, and not to disparities in

preparation or achievement (Mann and Diprete

16.1.2 S tudent Beliefs 2013; Morgan et  al. 2016; Riegle-Crumb et  al.
as Determinants 2012; Buchmann and DiPrete 2006; Xie and
of Educational Attainment Shaumann 2003). Given the importance of stu-
dents’ perceived sense of self-efficacy in their

choices and behaviors, researchers have explored

Social psychologists have long demonstrated that how to influence and strengthen this predictor of

students’ beliefs about their abilities to succeed educational attainment. Information and feed-

are related to their effort (Bandura 1982, 1997; back may play an important role for strengthen-

Dweck and Elliott 1983; Schunk 1991). These ing students’ sense of self-efficacy (a topic we

ideas are related to self-affirmation theory, which turn to in Sect. 16.3 of this review).

suggests that people are inherently motivated to

see themselves as competent and in control of
their futures and will work to restore their self-­ 16.1.3 H as Attainment Kept
Up with Aspirations?
worth when it is threatened (Steele 1988;

Sherman and Cohen 2006; Yeager and Walton

2011). Relatedly, the belief that people can Educational attainment has changed dramatically

achieve what they desire through their actions is over the past century in the U.S.  In particular,

the foundation of self-efficacy theory (Bandura high school completion rates have substantially

1993; Gecas and Schwalbe 1983, 1986; Gecas improved for all groups. Specifically, from 1990

and Seff 1989, 1990; Marcussen et  al. 2004; to 2014, the status dropout rate (representing the

Owens and Serpe 2003). Self-efficacy is a key percentage of the noninstitutionalized 16- to

component to how students may handle challeng- 24-year-old population who are not enrolled in

ing or unpredictable situations and, importantly, school and who have not completed a high school

how much effort they may decide to expend, or program) declined from 13.2% to 7.4% among

how long they persist in light of challenging or Blacks and from 32.4% to 10.6% among

unpredictable situations. Individuals’ sense of Hispanics. Both rates, however, remain higher

efficacy can influence actions indirectly by, for than the rate among non-Hispanic Whites (5.2%)1

example, impacting their goals and aspirations, (National Center for Education Statistics 2016b).

their effort and commitments to different pur- Although important disparities remain in high

suits, and how they cope with challenging situa- school completion, today race and income gaps

tions (Bandura 1981; Marsh et al. 1991; Murdock are notably wider in college degree enrollment

et al. 2000; Reyes and Jason 1993). and completion (Bailey and Dynarski 2011;

Experiments from social psychology demon- Black and Sufi 2002).

strate that accentuating positive growth rather

than shortfalls enhances self-efficacy, aspira-

tions, and performance (Bandura 1993). This is 1 There has been much discussion in the measurement of
critical because how students’ process early dif- high school completion/dropout status (see http://nces.
ficulties can influence their educational trajecto- ed.gov/pubs2016/2016007.pdf).

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 365

The number of students attending colleges and One of the most important determinants of
universities in the U.S. grew to 20.2  million in college entrance and completion is prior aca-
2015, an increase of nearly 33% since 2000 demic preparation. Given the push for college
(Kena et al. 2015). This increase was due in part participation, students’ pre-collegiate experi-
to growth in the size of the young adult popula- ences are a critical part of their educational path-
tion of the U.S. as well as increasing rates of ways, and where the notion of constrained
postsecondary participation. Approximately 40% choices—individual decision-making amidst
of 18- to 24-year-olds (i.e., the traditional college-­ forces of structural inequalities—play out
age population) were enrolled in a postsecondary through differentiation in schooling experiences
program in 2013, representing a 12.4% increase from early childhood to high school.
over 2000 enrollment levels. However, this
increase was not constant across all subgroups. 16.2 C urricular Differentiation
For example, while the percentage of Hispanic Along Students’ Educational
18- to 24-year-olds attending college grew by Pathways
56% between 2000 and 2013, the enrollment lev-
els of Black young adults demonstrated virtually Educational pathways are in large part a function
no change (NCES 2016a). In 1990, the White– of students’ schooling experiences, particularly
Black gap in college enrollment was 15 percent- their exposure to high-quality and rigorous cur-
age points, and the White–Hispanic gap was 12 ricula. Curricular differentiation, which refers to
percentage points. In recent years, the White– the process of sorting students into educational
Black gap has narrowed to about 7 percentage settings that differ according to substantive con-
points and the White–Hispanic gap to about 8 tent, pace of instruction, or pedagogical approach,
percentage points (NCES 2016a). College enroll- is a key feature of students’ educational path-
ment gaps by income have not narrowed nearly ways, starting with preschool environments that
as much as race gaps. Since the mid-1970s the promote school readiness, gatekeeping courses in
high–low income gap in college enrollment has the middle school years, and rigorous high school
stayed relatively constant at about 30 percentage curricula to facilitate successful transitions to
points (NCES 2016a). postsecondary schooling. Such curricular differ-
entiation is fraught with tensions of individual
Although more young people are choosing to choice and structurally constrained access to the
enroll in college than ever before, the rate of opportunities necessary to realize those goals.
degree completion has not kept up with participa-
tion, and disparities in college degree receipt There are several plausible mechanisms by
remain pronounced and in some cases are actu- which we would expect high-quality and rigor-
ally growing. Forty-three percent of non-H­ ispanic ous curricula and instruction through the educa-
Whites aged 25–29 held a bachelor’s degree or tional life course to lead to increased educational
higher in 2015, compared to 21.3% of Blacks and attainment. First, a rigorous course of study often
16.4% of Hispanics (National Center for provides exposure to more advanced material,
Education Statistics 2016b). Gaps by income in introducing students to topics they may encoun-
degree completion are also pronounced (Bailey ter in subsequent years thereby improving their
and Dynarski 2011). In 2013, less than 10% of schooling transitions and supporting greater aca-
young adults from the lowest income quartile demic success and confidence (Lee and Ready
earned a college degree, compared to 77% of 2009; Long et  al. 2012). Second, high-quality
those from families in the top income quartile (an content is often correlated with high-quality
increase since 1970 of over 30 percentage points instruction. For example, more rigorous courses
among high-income families and only by about 3 of study in high school (such as honors and AP)
percentage points among those in the lowest are frequently taught by more skilled teachers
income bracket (Pell Institute for the Study of (often with additional credentials, more
Opportunity in Higher Education).

366 M. Kurlaender and J. Hibel

­experience, or specialized professional develop- result of much more dynamic interactions

ment), than less rigorous courses (Ingersoll 1999; between structural barriers and individual selec-

Kalogrides et al. 2013). Third, rigorous schooling tion (i.e. constrained choices).

environments (across or within schools) attract Nevertheless, students do not enroll in a

particular students (and families), often those course of study purely based on their own prefer-

most socially, financially, or academically able ences, nor strictly by chance. Schools serving

and/or those most motivated (Lareau and Goyette high concentrations of low-income students often

2014). As such, engagement with these higher-­ have fewer advanced curricular offerings than do

achieving peers (based on ability, social class, schools serving a more affluent student popula-

motivation, etc.) may positively influence student tion (Adelman 1999; Conger et  al. 2009).

outcomes.2 Fourth, enrollment in more intensely Moreover, canonical studies that account for

rigorous schooling environments can serve as an school differences suggest that, like racial/ethnic

important, positive signal in future schooling and socioeconomic disparities in achievement

destinations. For example, among kindergarten (Coleman 1966), disparities in course-taking are

teachers who often differentiate students based largely within-school phenomena rather than

on their pre-schooling environments; or, among between-school phenomena (Gamoran 1987).

college admissions officers who rank high This suggests that curricular disparities are

schools on their academic intensity. mainly due to tracking or to inequalities in access

Importantly, the relationship between rigorous to more demanding courses among students who

course of study and student educational destina- are enrolled in the same school (Attewell and

tions may not be causal at all, because, to a large Domina 2008). The implications of such inequal-

extent, students self-select into rigorous courses ity suggest that researchers and educators must

in secondary and postsecondary education. continue to investigate more closely the pro-

Students who take a more rigorous set of courses cesses that contribute to course sorting, particu-

in high school likely have a host of other attri- larly when it results in within-school racial/ethnic

butes that also lead to their success in college and or socioeconomic segregation (Deil-Amen and

later in life (Domina et  al. 2014). For example, DeLuca 2010; Kelly 2009; Riegle-Crumb and

such students may simply have better academic Grodsky 2010).

skills, more motivation, and a stronger work

ethic, or perhaps more academic support and
encouragement from their families or teachers. 16.2.1 A cademic Curriculum
in the Pre-schooling Years
Several studies have also documented the quali-

tatively different ways parents from different

income backgrounds intervene in their children’s For most children in the United States, the path-

schooling experiences (Hamilton 2016; Stevens way through formal schooling begins with par-

2007; Lareau 2011, 2000). It is therefore likely ticipation in center-based pre-kindergarten

that all of these attributes contribute to students’ programming. Most pre-kindergarten programs

enrollment in more rigorous courses of study in have the overarching goal of increasing students’

the first place, making it difficult to test whether “school readiness,” the set of intellectual, social,

particular courses or curricular tracks directly and emotional competencies that foster success

cause students to succeed in college or later in in kindergarten and beyond (Duncan et al. 2007).

life. Thus, students’ educational outcomes that A recent meta-analysis of pre-kindergarten pro-

may appear to vary as a result of differential gram evaluation studies concluded that, on aver-

access to rigorous schools and/or curricula (i.e., a age, pre-kindergarten participants gained the

set of structural constraints) are in fact likely the equivalent of an additional four months of learn-

ing compared to children who did not attend pre-

2 See Sacerdote (2001) and Zimmerman (2003) for evi- school, providing them with a stronger foundation
as they entered kindergarten (Camilli et al. 2010).
dence of peer effects in education.

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 367

Moreover, longitudinal studies of pre-­Heckman et  al. 2010). Beginning in the 1960s,

kindergarten education’s effects demonstrate that findings from a series of now famous experi-

short-term improvements in language and math- ments began to emerge that demonstrated pre-­

ematics ability are accompanied by positive out- kindergarten education’s ameliorative potential

comes in the longer-term as well, including for low-income children (Schweinhart et  al.

increased educational attainment, higher earn- 2005; Campbell and Ramey 2010).4

ings, and less criminal behavior in early adult- Despite the aforementioned success stories of

hood (Campbell et  al. 2002; Heckman et  al. efforts to improve disadvantaged students’ edu-

2010).3 cational outcomes through pre-kindergarten pro-

Researchers studying the effects of pre-­ gramming, the federal government’s preschool

kindergarten education on children’s academic programs for children in poverty—Head Start

and socio-emotional development have identified and Early Head Start—have generated a mixed

largely positive effects of participation in aca- pattern of results. The Head Start Impact Study

demically oriented programming (e.g., Gormley (Puma et  al. 2010) used a randomized control

et al. 2005). Moreover, these effects demonstrate design to estimate the effects of 3- and 4-year-­

a compensatory effect: Children from low-­SES olds’ Head Start participation on their cognitive

backgrounds experience greater increases in and social-emotional outcomes at the end of first

early achievement and development than do chil- grade. While the study’s results indicated that

dren from middle- or upper-class families. Head Start participants enjoyed benefits during

However, because access to academically ori- program participation, these advantages “faded

ented pre-kindergarten programs is stratified out” over a relatively short period of time. Recent

along social class lines, fewer low-income chil- findings from a randomized control trial in Head

dren participate in such programs than do higher-­ Start programs suggest that this fade-out is attrib-

income students. Thus, as the first form of formal utable to elementary school quality, as program

education encountered by many U.S. children, participants who subsequently enrolled in high-­

structural inequalities in pre-kindergarten pro- performing elementary schools demonstrated

gramming establish unequal academic pathways continued benefits, while those who attended

that extend into the elementary school years. lower-performing schools experienced fade-out

Early childhood remains a primary area of (Zhai et  al. 2012). This finding echoed earlier

compensatory social investments aimed at atten- work by Currie and Thomas (2000), who demon-

uating inequality prior to formal schooling. strated that elementary school quality differences

Researchers, educators, and policymakers have at explained differential Head Start fade-out

turns considered the potential for pre-k­ indergarten effects among White and Black students. Thus,

education to improve the education and life

course outcomes of children from socioeconomi- 4 The Perry project enrolled 58 low-income, Black 3-year-
cally disadvantaged and ethnic minority back- olds in 2.5-h classes that met 5  days per week for the

grounds (e.g., Currie 2001; Duncan et  al. 2007; 2 years preceding kindergarten. Members of the treatment

group demonstrated multiple advantages relative to the

control group in the near-term (e.g., higher IQ scores,

3 While these outcomes carry clear benefits for individual increased standardized test performance, better teacher-

students, researchers have also performed cost-benefit rated classroom behavior) and in the long-term (e.g.,

analyses at aggregate levels, finding that preschool pro- higher high school graduation rates, less involvement in

grams provide benefits to society as a whole through cost the criminal justice system as adolescents and adults,

savings (e.g., reduced spending on expensive special edu- higher earnings in adulthood) (Schweinhart et al. 2005).

cation or juvenile justice programs) and participants’ Similarly, Abecedarian Project participants, who received

increased economic productivity in adulthood. Estimates pre-kindergarten educational intervention from approxi-

of these societal benefits tend to outweigh preschool oper- mately four months of age until kindergarten entry, expe-

ating costs by considerable margins, often on the order of rienced improved achievement, attainment, and health

$5 or more of economic return for every $1 spent on pre- outcomes compared to control group members from

kindergarten programs (Duncan et  al. 2007; Heckman childhood through adulthood (Campbell and Ramey

et al. 2010; Reynolds et al. 2011; Yoshikawa et al. 2013). 2010).

368 M. Kurlaender and J. Hibel

early investments in children’s schooling for curriculum, most often in language arts or math-
improving educational and ­occupational attain- ematics. Classroom teachers make group assign-
ment are largely only realized through sustained ments based on their assessment of students’
quality experiences in schooling. current knowledge and cognitive ability, with the
goal of allowing the teacher to present students
16.2.2 Academic Curriculum with a curriculum that is neither too challenging
in the Elementary Schooling (which might place students at risk for frustration
Years and discouragement) or too easy (which might
lead to developmental stagnation or disruptive
School districts in 46 states are required to offer behavior).
publicly funded schooling beginning with kin-
dergarten, in which children are typically eligible As a potential solution to the pedagogical
to enroll at age 5.5,6 Students proceed through the challenge of teaching groups of young students
elementary years along pathways that are differ- with widely varying levels of preparedness and
entiated by curricular content, pace of instruc- performance, ability grouping offers a compel-
tion, and pedagogical approach. ling logic. Indeed, on the face of things it might
even seem irrational to argue that administering a
Ability grouping in elementary school class- one-size-fits-all curriculum could ever be prefer-
rooms has been a frequent subject of sociological able to presenting students with tailored instruc-
research since the 1980s. The term ability group- tion matched to their specific learning styles and
ing refers to the practice of organizing a class- needs. However, research findings from the soci-
room of students into small groups for the ology of education complicate this picture, call-
purpose of delivering to each group a modified ing into question ability grouping’s educational
efficacy, and bringing to light the structural forces
5 34 States require districts to offer half-day kindergarten that determine students’ groupings, which often
programs and 12 states require full-day kindergarten. result in inequities along racial/ethnic and socio-
Kindergarten attendance is compulsory in 16 of these economic lines (Gamoran et  al. 1995; Hallinan
states. The age at which children must legally begin 1994; Oakes 2005; Slavin 1987).
attending school varies across states, ranging from five to
eight  years old. (Source: https://nces.ed.gov/programs/ The academic pathways constructed through
statereform/tab5_3.asp) within-class ability grouping are often less visi-
ble than those created by, for example, curricu-
6 “Academic redshirting,” the practice of voluntarily lum track placement in high school. Unlike those
delaying children’s kindergarten enrollment by one year, formal curricular placements, which require par-
has received abundant scholarly and popular attention in ents’ and/or students’ consent, elementary school
recent years. While research evidence suggests that the ability group decisions generally fall under the
practice is most common among boys, non-Latino Whites, classroom teacher’s sole purview. Moreover, stu-
children from high-SES families, and those whose birth- dents’ ability group placements are generally not
days fall close to kindergarten enrollment cutoff dates noted in their school records or transcripts.
(Bassok and Reardon 2013), estimates of academic red- Despite their comparative informality, however,
shirting’s prevalence indicate that it is not as widespread ability group placements have the potential to
as commonly believed, with between 3.5% and 5.6% of establish durable academic pathways for young
U.S. kindergarteners demonstrating delayed enrollment students, and these pathways feed directly into
(Bassok and Reardon 2013; Huang 2015; Snyder and the formally differentiated curricular pathways of
Dillow 2013). Increased age at kindergarten enrollment is middle and high school. The social-p­ sychological
associated with a host of short-term positive outcomes, consequences of such groupings on students’
including higher achievement (Datar 2006; Datar and subsequent choices about curricular tracks (when
Gottfied 2015), improved social-behavioral skills (Datar such choices are at the individual or parental
and Gottfied 2015), and dramatically reduced odds of level), however, are not well understood.
being diagnosed with attention deficit/hyperactivity disor-
der (Dee and Sievertsen 2015), yet evidence for positive
long-term effects is scant (Cascio and Schanzenbach
2016; Deming and Dynarski 2008; Lincove and Painter
2006).

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 369

Building on the concept of opportunity-to-­ inequalities along racial/ethnic, and socioeco-

learn (OTL), sociologists of education have made nomic lines (Gamoran et al. 1995; Hallinan 1994;

the straightforward argument that students are Oakes 2005; Slavin 1987).

more likely to learn material that is presented to

them in class than material they never encounter
(e.g., Porter 1989; Sørensen and Hallinan 1986). 16.2.3 Academic Curriculum
in the Middle School Years:
Extending this line of work to research on ability The Push for Universal

grouping, several researchers have found that the Algebra

amount of curricular material presented to stu-

dents in differentiated ability groups exhibits

considerable variation across learning groups in Following the publication of A Nation at Risk

the same classroom, with students in high-ability (National Commission on Excellence in

groups being exposed to a greater proportion of Education 1983), American public education

the intended curriculum than those in middle- or took a decided turn toward emphasizing aca-

low-ability groups (Eder 1981; Gamoran 1986; demic achievement, particularly in science and

Oakes 1985; Pallas et  al. 1994). Thus, ability mathematics. This sea change included an expan-

grouping potentially provides unequal OTL sion of rigorous curricula during the middle

according to teachers’ perception of students’ school years as a means of ensuring the United

ability, leading to further widening of initial States’ future economic competitiveness and

achievement gaps over time, a pattern some soci- national security (Schoenfeld 2004). The push

ologists refer to as “cumulative advantage” (e.g., for more and earlier student access to advanced

DiPrete and Eirich 2006) or “the Matthew effect” mathematics was promoted as a solution to A

(e.g., Kerckhoff and Glennie 1999). To the extent Nation at Risk’s prophesized “rising tide of medi-

that initially high achieving students cover more ocrity.” In response, a contingent of educators

curricular ground than initially lower achieving and civil rights leaders began to put forth an

students over the course of each school year, this equity-based argument for curricular reforms,

process tends to be self-reinforcing across the specifically in mathematics, targeting under-

elementary school grades (i.e., the students who served students and schools. Robert Moses, a

finish a given school year having learned the math educator and an influential activist in the

most material are the “high achievers” when the civil rights era of the 1960s, is most closely asso-

following school year begins, and are therefore ciated with this movement. Having founded the

placed in high-ability, high-OTL groups once Algebra Project in 1982 to improve mathematics

again). This curricular path dependence mani- education among low-income students and stu-

fests in the form of unequal educational pathways dents of color, Moses argued that access to

concealed within what, on the surface, appears to advanced mathematics is a requisite for full eco-

be a singular educational “mainstream.” nomic participation and citizenship in an increas-

Research findings suggest that more flexible ingly technological society, and one that is

(i.e., frequently adjusted) and appropriate (i.e., systematically denied to members of marginal-

accurate with respect to students’ learning abili- ized populations (Moses and Cobb 2002). These

ties) group placements lead to greater equality of distinct yet mutually reinforcing arguments—

academic outcomes experienced by students of excellence and equity—ushered in an era of

varying abilities (Sørensen 1970; Gamoran 1992; intense preoccupation with boosting algebra

Gamoran et al. 1995). In practice, however, abil- enrollments nationwide (Gamoran and Hannigan

ity grouping systems are highly imperfect along 2000).

these lines. Inappropriate and fairly static group The “algebra for everyone” perspective shaped

assignments tend to result in students being education reform in multiple ways, most notably

assigned to differentiated curricular pathways in in the form of heightened course-taking expecta-

ways that exacerbate pre-existing achievement tions that became part of the emerging standards

370 M. Kurlaender and J. Hibel

and accountability reform movement. Reports performance declines among initially high-skill
from the National Research Council (Everybody students (Nomi 2012; Nomi and Allensworth
Counts [NRC, 1989]) and National Council of 2013). Additional research will be necessary to
Teachers of Mathematics (Curriculum and understand the causes of these disappointing out-
Evaluation Standards [NCTM, 1989]) codified comes; initial results from quantitative work point
these new, intensified expectations, leading sev- toward “peer effects” and the complex set of
eral states and large school districts to respond social relations that result from heterogeneous
accordingly. grouping strategies like universal algebra as key
challenges (Hong and Nomi 2012; Domina et al.
These efforts reached a zenith in California, 2015). Other work, perhaps qualitative in nature,
where, in 1997, the state department of education is necessary for understanding why such reforms
revised its education standards to reflect an may not meet desired outcomes.
expectation that all students be enrolled in alge-
bra, a recommendation that became law with the 16.2.4 A cademic Curriculum
passage of the Public School Accountability Act and Rigor in High School
(PSAA) in 1999 (Domina et al. 2014). Response
to this legislative reform was swift: Over half of Not surprisingly, students with a more rigorous
California eighth graders were enrolled in alge- course of study in high school are more likely to
bra courses by 2008, up from only 16% at the apply and enroll in more selective campuses, and
time of the PSAA’s passage. For policymakers are less likely to require remediation when they
concerned with Americans’ declining technical enter college (Kurlaender and Howell 2012;
expertise, as well as education activists dedicated Long et al. 2012; Adelman 1999, 2006). Enrolling
to equalizing students’ pathways to college, this in a rigorous course of study in high school is not
“algebra for everyone” reform represented an only associated with higher educational attain-
encouraging step forward (similarly ambitious ment, but also with improved labor market out-
reforms in other states and large school districts, comes. Several studies find that enrolling in more
such as Chicago Public Schools were also under- advanced mathematics courses in high school
way (Allensworth et al. 2009; Nomi 2012; Nomi leads to higher wages once in the workforce
and Allensworth 2013)). (Altonji 1995; Levine and Zimmerman 1995;
Rose and Betts 2004).
While California made strides toward achiev-
ing the near-term goal of increasing access to Researchers have attempted to deal with the
algebra among middle school students, recent complexity of estimating the influence of curric-
assessments of the algebra for everyone move- ular intensity on future success by using a variety
ment’s longer-run impacts have been somewhat of approaches. When researchers control for as
disappointing. Despite the widely held under- many observable characteristics as are available,
standing that algebra operates as a “gatekeeper” they find a consistent positive association
for participation in future advanced mathematics between curricular intensity and the following:
coursework (Oakes 1990; Riley 1997; Smith student test scores (Attewell and Domina 2008),
1996), recent evaluations have revealed that man- high school graduation (Schneider et  al. 1997),
datory eighth grade algebra reforms do not lead to college entry (Long et al. 2012), type of college
increased advanced math course-taking (Liang entry (Attewell and Domina 2008), college
et al. 2012), nor have mandatory algebra reforms grades (Klopfenstein and Thomas 2009), college
led to increased average mathematics achieve- graduation, (Adelman 2006; Attewell and
ment in the high school years (Clotfelter et  al. Domina 2008), and wages (Altonji 1995; Rose
2015; Domina et al. 2014, 2015; Loveless 2008). and Betts 2004).
Similarly, evaluations of Chicago Public Schools’
mandatory ninth grade algebra reforms found that Using detailed information from students’
the program was associated with increased failure high school transcripts, Long et al. (2012) find
in subsequent mathematics coursework, as well as

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 371

a 7–11 percentage point increase in the likeli- are Black or Latino youth (Planty et  al. 2006).7

hood of high school graduation and 4-year col- Data on mathematics course-taking over time

lege entry between a student who takes no reveal that, although increasing numbers of stu-

rigorous high school courses and a student tak- dents have been completing precalculus or calcu-

ing just one rigorous course during high school. lus in high school in recent decades, the rates for

This study finds that the biggest differences in Black and Hispanic/Latino students clearly lag

student outcomes are based on math and behind the rates of White and Asian high school

English course levels, though enrollment in students, and these gaps have actually grown over

rigorous courses in other subjects also leads to time (see Table 16.1).

improved outcomes. Long and colleagues also Similarly, students from higher-income fami-

find that, although more rigorous courses are lies have higher levels of participation in more

associated with better student outcomes, the rigorous academic coursework than do their

differences were greatest between those taking lower-SES peers. This is consistent with research

no rigorous course and those taking only one. indicating that lower-SES students, in particular,

This result suggests “requiring or encouraging

students to enroll in even one rigorous course Table 16.1  Percentage of high school graduates who

in their first two years of high school can sub- completed precalculus or calculus, by race and socioeco-
stantially improve graduation and four-year nomic status: 1982, 1992, and 2004

college enrollment rates” (Long et  al. 2012, 1982 1992 2004

p. 315). Overall 10.7 21.7 33.0

Improving academic standards in secondary Race/ethnicity
White 12.2 23.0 36.7
schools has been at the heart of the Common Black 4.0 13.6 19.0
Core State Standards reform efforts, which has Hispanic/Latino 5.3 12.7 21.9
emphasized the need to better align K–12 edu- Asian/PI 30.1 41.6 56.8

cation systems with higher education to ensure Am Indian 2.3 3.1 12.9
a more seamless transition for young adults Socioeconomic status

between high school and college, and between 1st quartile (lowest) 2.7 8.0 17.7

high school and the labor market. The push for 2nd quartile 6.7 13.2 22.7

more academic rigor is evident in the course-­ 3rd quartile 11.3 21.9 34.0

taking trajectories of high school students. 4th quartile (highest) 2.05 38.5 52.4

Over the last three decades the percentage of Source: U.S.  Department of Education, National Center
students enrolled in precalculus or calculus in for Education Statistics, High School and Beyond
U.S. high schools has steadily grown. In 1982, Longitudinal Study of 1980 Sophomores
only slightly more than 10% of students gradu- (HS&B-So:80/82), “High School Transcript Study”;
National Education Longitudinal Study of 1988

ated high school with precalculus or calculus (NELS:88/92), “Second Follow-up, Transcript Survey,
coursework, by 1992 that figure more than dou- 1992”; and Education Longitudinal Study of 2002
bled to 21.7%, and in 2004, 33% of high school (ELS:2002), “First Follow-up, High School Transcript
students were enrolled in at least precalculus Study, 2004.” Available at: http://nces.ed.gov/
pubs2007/2007312.pdf

coursework.

Efforts to increase the academic intensity of

students’ high school curricula have also been 7 Source: U.S. Department of Education, National Center
spurred by an equity agenda that seeks to ensure for Education Statistics, Education Longitudinal Study of
access to rigorous courses for students from all 2002 (ELS:2002), “High School Transcript Study.”
demographic backgrounds. Data from a nation- Adapted from: Planty, M., Bozick, R., and Ingels, S.J.
(2006). Academic Pathways, Preparation, and

ally representative sample of high school stu- Performance — A Descriptive Overview of the Transcripts
dents’ course enrollment reveal that White and from the High School Graduating Class of 2003–04
Asian students are much more likely to be (NCES 2007–316). U.S.  Department of Education,
enrolled in a more rigorous set of courses than National Center for Education Statistics. Washington,
DC: U.S. Government Printing Office.

372 M. Kurlaender and J. Hibel

continue to be underrepresented at more selective In principle, any academically stimulating
postsecondary institutions because they have not environment may contribute to academic rigor. In
completed the appropriate coursework (Carnevale practice, however, evaluating the learning envi-
and Rose 2003). Importantly, Rose and Betts ronment for rigor can be difficult. It is common to
(2004) find that the type of math courses students use measures such as course titles and/or grades
take in high school explain 27% of the earnings as proxies for rigor. But even these do not mean
gap between students from the lowest-income the same thing everywhere. For example,
families and those from middle-­income families. Fig. 16.2 displays a scatterplot of the probability
Similarly, using data on students in Florida public of being college ready (Y-axis) against high
postsecondary institutions, Long and colleagues school grade point average (X-axis) among stu-
(2009) find that 28–35% of the gaps (and over dents attending one of the 23 campuses of the
three-quarters of the Asian advantage) in college California State University (CSU) system (the
readiness among college-­going Black, Hispanic, State’s primary public BA-granting higher edu-
and low-income students can be explained by the cation system and the nation’s largest public
highest math course taken in high school. 4-year postsecondary system). Beyond the obvi-
ous positive association between high school
There are a host of factors that contribute to GPA and college readiness is the stark difference
students’ sorting into various levels of courses in between School A and School B. In School A, a
high school: availability of courses, knowledge California public school that serves less than
of offerings at the school, academic ability, inter- 10% of students on free/reduced price lunch,
est, motivation, familial involvement, and the even a student with a 2.5 GPA enters the CSU
influences of teachers, counselors, and/or peers. system “college ready” (as measured by place-
As such, properly addressing racial/ethnic and ment tests); in contrast, in School B, which has
socioeconomic differences in analysis of course an over 90% free/reduced price meal eligibility
enrollment patterns requires further inquiry into rate among its students (a great majority of them
each of these (and other) possible sources of Latino), even the 4.0 student only has about a
existing disparities in curricular pathways. 40% likelihood of being “college ready” (i.e., not
needing any remediation when she enters
16.2.5 S tructural Differences college).
in Academic Preparation
for College In sum, students are not randomly placed into
their educational pathways, but rather their cur-
Studies that parse out the effects of academic ricular pathways are shaped by both the opportu-
rigor by race/ethnicity and SES find that the nities that they are exposed to (a structural
return to taking more advanced coursework could argument) and their choices (an individual agency
vary with the attributes of the school. For exam- argument). The result is that it is not only difficult
ple, Long et al. (2012) find that students attend- to separate out unobserved motivation, support,
ing high-poverty schools or those with lower or other characteristics that may be associated
average levels of student achievement experi- with both rigorous course-taking and better edu-
enced larger increases in their high school gradu- cational outcomes, but also the many structural
ation and college enrollment rates associated dimensions that constrain individual choice.
with taking more rigorous courses than students These competing forces often result in educa-
attending more affluent high schools. Efforts to tional pathways that self-perpetuate. That is,
ensure opportunities are more equally distributed quality early schooling experiences beget better
between schools have focused on addressing dis- placement into secondary schooling decisions,
parities in curricular offerings, particularly in and then more intense academic rigor in high
college gateway courses such as Advanced school that results in more selective college
Placement (College Board). admissions, and greater likelihood of degree
attainment and labor market success. Thus, stu-

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 373

Fig. 16.2  The association between college readiness and high school GPA in California Public Schools

dents do have many choices and self-select (often and the extent to which programs aimed at
with the aid of parents, teachers, or counselors) improving college information can attenuate
into a course of study; however, they do so within inequality in postsecondary pathways.
a set of structural constraints or opportunities. A
primary way in which educators and ­policymakers As a set of resources embedded in social rela-
hope to break the self-perpetuating nature of edu- tionships that facilitate certain actions (including
cational pathways and improve mobility between applying to or enrolling in a particular college),
educational destinations is through increased social capital plays a major role in shaping stu-
access to information about alternative pathways dents’ educational pathways. It is through their
and opportunities. social connections that students learn a normative
orientation toward higher education (i.e., the edu-
16.3 The Role of Information cational expectations to which they will be held
in Navigating Educational by others in their social networks, including their
Pathways: Barriers parents, teachers, and peers) and also acquire
and Opportunities valuable information from others about the col-
lege application and participation processes.
Despite a college for all culture, students often
have very limited and only vague information Research on social capital’s role in college
about what college will be like, which is particu- application and attendance has demonstrated the
larly true for students who are the first in their importance of students’ social ties to peers (Perez
family to attend college (Settersten and Ray and McDonough 2008; Perna 2000; Tierney and
2010). The research on inequality in educational Venegas 2006), institutional agents such as teach-
attainment, particularly examinations of path- ers and counselors (Perna and Titus 2005;
ways to college and college choice, is heavily Stanton-Salazar 1997, 2001), immediate and
framed by theories of social and cultural capital, extended family members (McDonough 1997;
Perna and Titus 2005), college outreach programs
(Gonzalez et  al. 2003), and the overall school
community (Sandefur et  al. 2006) for develop-

374 M. Kurlaender and J. Hibel

ment of college-going attitudes and behaviors. as suffering from “information poverty” with

Students who internalize socially constructed respect to the postsecondary school transition

norms of college attendance through their social and the pathway from college to work. These

interactions and those whose networks provide youths’ sense of urgency toward obtaining a solid

access to information about the multifarious job led them toward the ill-advised decision to

details associated with the college choice process enroll in for-profit trade programs rather than 2-

enjoy increased probabilities of college atten- or 4-year nonprofit institutions. As a consequence

dance and persistence relative to students with of their low levels of cultural capital, the students

fewer social capital resources. in Holland and DeLuca’s study ended up with

Like social capital, cultural capital’s role in fewer job prospects and more financial debt than

the college-going process has also received sub- they might have if their postsecondary choices

stantial attention from sociologists. In this con- had been better informed.

text, cultural capital refers to the status-linked Like Holland and DeLuca, Harding (2010)

sets of skills, knowledge, and preferences that are uses interviews with low-SES, ethnoracial minor-

rewarded by higher education institutions and are ity youth to examine the role cultural capital

transmitted from parents to their children (Lareau plays in shaping their educational pathways.

and Weininger 2003). In her work on the college While the youth in Harding’s study overwhelm-

choice process, McDonough (1997) describes ingly aspired to a college degree, the most effec-

how middle-class parents’ access to first-hand tive strategies for reaching this goal were

information about college admissions procedures obscured by their lack of cultural capital. Unlike

and strategies for maximizing their children’s the higher-SES youth who are surrounded by

odds of admission (e.g., through the use of pri- individuals who espouse a “mainstream” model

vate SAT tutors) represent a form of cultural cap- of desirable educational pathways, low-SES

ital—valuable information that is readily youth exist in a context of “cultural heterogene-

available only to children of high-status parents ity,” which produces multiple alternative logics

and is not transmitted through schooling. of educational success, including alternative cre-

Compared to students who lack dominant cul- dentials (e.g., the GED), attending trade schools,

tural capital (particularly in in the form of college or choosing job training programs over tradi-

admissions information), those who possess tional college. Whereas higher-SES youth are

institutionally valued cultural capital are more presented with a unified cultural front regarding

likely to hold high educational aspirations, enroll the desirability of a 4-year degree (and the cor-

in college, and reap positive returns to their post- responding undesirability of other pathways),

secondary education investments (Aschaffenburg lower-SES youths’ cultural repertoires include

and Maas 1997; Bourdieu and Passeron 1977; support for multiple educational and occupa-

DiMaggio and Mohr 1985; Dumais and Ward tional pathways, which weakens the relationship

2010; Lamont and Lareau 1988; Schneider and between their (almost universally high) postsec-

Stevenson, 1999). For high-achieving, low-SES ondary aspirations and their ultimate educational

youth, this lack of cultural capital also leads them attainment (Harding 2010, 2011). Along similar

to disproportionately apply to nonselective lines, Lee and Zhou (2015) attribute children of

schools that offer a poor match to their academic Asian immigrants’ “paradoxically” high levels of

skills and to be unsuccessful in navigating the educational attainment to a set of culturally

financial aid process (Goldrick-Rab 2006; Hoxby grounded “success frames” through which com-

and Avery 2013). munity members establish a narrow definition of

A recent example of sociological research on academic success as attaining an advanced degree

information qua cultural capital can be found in from an elite college or university.

the work of Holland and DeLuca (2016). Sociologists often view social and cultural

Analyzing data from interviews with 150 low-­ capital resources as intertwined and mutually

income youth, the authors describe these students dependent, with access to one set of resources

16  Students’ Educational Pathways: Aspirations, Decisions, and Constrained Choices Along… 375

potentially mitigating low levels of the other. For the individual and to the public), we also know

example, Grodsky and Riegle-Crumb (2010) that students who enter college in need of reme-

find that social capital may be especially impor- diation are less likely to persist and less likely to

tant to the college choice process for students complete a college degree than those who do not

who do not possess a “college-going habitus.” A require remedial coursework (Bettinger et  al.

concept originating in cultural capital theory 2013). Part of the explanation for the large share

(Bourdieu and Passeron 1977), habitus refers to of students requiring remediation once they

the attitudes and dispositions an individual arrive in college may be a result of the limited

unconsciously develops through repeated inter- information students possess regarding what they

actions with the social world. Grodsky and need to do to succeed in college.

Riegle-Crumb (2010) identify an individual’s An important effort to improve alignment

taken-for-granted belief that they will attend col- between K–12 and postsecondary systems is to

lege as the hallmark of a college-going habitus provide high school students with early informa-

that is disproportionately possessed by young tion about college expectations. High school stu-

members of the elite, for whom the postsecond- dents use information from many sources to

ary transition occurs almost as a natural matter make numerous decisions, such as whether and

of course. Unlike these privileged students, those how to complete high school, and whether and

who do not possess a college-going habitus must where to attend college. Early information may

develop the knowledge, skills, and attitudes con- help students realize that they need additional

sistent with college attendance via the social academic preparation, and motivate students to

capital resources they manage to access over the do well with their remaining time in high school.

course of their educational careers (Grodsky and Moreover, there is evidence that high school stu-

Riegle-­Crumb 2010). dents update their college-going trajectories

Because social and cultural capital, like other based on information that they receive during

valuable resources, are differentially distributed secondary school (Jacob and Wilder-Linkow

along typical axes of social stratification, inequal- 2011). In fact, students respond to labels assigned

ities in social and cultural capital tend to magnify to them by standardized tests. Papay et al. (2011)

existing gaps in college attendance and persis- show that the labels assigned to students through

tence. Programs designed to facilitate the transfer state standardized testing impact college-going

of college information to all students aim to inter- decisions. A “Needs Improvement” label causes

vene by interrupting the link between social sta- urban, low-income students to be more likely to

tus and social capital. Programs such as AVID, enroll in college than a “Warning” label.

Upward Bound and Summer Bridge have long Moreover, Papay et  al. (2011) show that urban,

focused on providing students (particularly first-­ low-income students were shown to update their

generation college students from underrepre- educational attainment expectations based on

sented backgrounds) with exposure to not just the standardized test result labels as early as eighth

academic, human capital skills, but also the social grade.

and behavior skills (i.e., cultural capital) required Early information from college assessments,

for college success. The evidence about the effec- which are intended to motivate students toward

tiveness of these programs, however, is mixed their postsecondary goals, could hypothetically

and limited (Domina 2009; Barnett et al. 2012). be discouraging to lower-performing students.

The most compelling evidence that educa- Students taking state assessments who are told

tional pathways are not seamless for many stu- that they may require remediation upon entering

dents is found in the high rates of college a particular college may feel that they do not fit

remediation present across broad access colleges well with that college, and decide to enroll else-

and universities throughout the U.S. (where the where or not at all. However, research on

majority of students go to college). Beyond the California’s effort to provide students with col-

great financial expense of college remediation (to lege readiness information in 11th grade found

376 M. Kurlaender and J. Hibel

that the early signal of “not ready” did not dis- income families are also likely to face higher

suade students from applying or enrolling in col- compliance costs because they most likely lack

lege, or push them into attending a less college-going peers and relatives to assist them

academically demanding college, and actually (Dynarski and Scott-Clayton 2006). If these bar-

improved overall remediation rates at California’s riers are larger for disadvantaged students, the

broad access 4-year institutions (Howell et  al. primary purpose of federal need-based financial

2010; Jackson 2015; Kurlaender et al. 2016). aid may be jeopardized (Scott-Clayton 2012).

The literature in education policy is also rich For students who do attempt the FAFSA, many

in studies focused on the role of information in have difficulty in answering questions, request-

college affordability. Despite being eligible, ing a high school diploma, or having a Social

many students do not apply for financial aid for Security number (Yonezawa 2013; McKinney

college (King 2004; College Board 2017; and Roberts 2012).

Yonezawa 2013). Information plays an important These information barriers could be especially

role in financial aid take-up because incomplete pronounced for non-traditional age students and

or insufficient information can lead students to students from low-income backgrounds attend-

underestimate benefits or overestimate costs of ing broad access institutions, such as community

college, and can preclude students from applying colleges (Bean and Metzner 1985; Taniguchi and

for financial aid (Perna 2007; Scott-Clayton Kaufman 2005). Compared to students at 4-year

2012). Household income and parent education institutions, community college students are

are positively correlated with knowledge of col- more likely to be first-generation college stu-

lege prices; minority and low-income parents are dents, to enroll part-time, have discontinuities in

less likely to provide an accurate estimate of col- terms enrolled, and switch between part-time and

lege costs when compared to more affluent or full-time enrollment (Crosta 2013; Provasnik and

White parents (Grodsky and Jones 2007; Horn Planty 2008; Bailey et al. 2005; Deil-Amen and

et al. 2003). Rosenbaum 2003; Dougherty 1994; Brint and

Financing college remains an important struc- Karabel 1989). The current financial aid system

tural constraint for many individuals. The pri- is largely designed to assist traditional under-

mary reason given by a representative sample of graduates enrolling right after high school (Long

youth that did not go to college is because they 2010). Community college students can also be

could not afford to attend (Bozick and DeLuca penalized if financial aid requires full-time atten-

2011). Need-based financial aid is designed to dance, a traditional high school diploma, or a

provide additional help for low-income students, specific goal for a credential or degree (Long

but complex aid formulas, poor marketing, and 2010; Terriquez and Gurantz 2014). Because

complex application procedures can create addi- need-based financial aid targets students at the

tional information barriers (Scott-Clayton 2012). margin of choosing whether or not to attend col-

Current financial aid barriers include lack of lege, the FAFSA’s complexity may lead to nega-

awareness about aid and the complexity of the tive decisions about college enrollment and/or

Free Application for Federal Student Aid persistence (Scott-Clayton 2012). In effect, the

(FAFSA) required for all federal and most state students least likely to be able to afford college

need-based aid programs (Long 2010). At five are the ones with the least amount of information

pages and 127 questions, the FAFSA is longer about college cost (Horn et al. 2003).

and more complicated than federal tax return In light of these information barriers, some

forms (Dynarski and Scott-Clayton 2006). This researchers have tested information-based inter-

complexity also has significant costs, including ventions in college financing. Most notably,

the time and resources it takes for individuals to Bettinger et  al. (2012) implemented a random-

read directions and requirements, collect all ized field experiment conducted with the tax

needed documents, and actually fill out the appli- preparation firm H&R Block to assist families

cation (Dynarski and Scott-Clayton 2006). Low-­ with FAFSA preparation. For dependent stu-