Use your sociological imaginations and findings of Renzulli et al., argument by Lee and Zhou, and two video clips that you viewed to respond to the following prompt:In your observations what types of

Adapting to Family Setbacks: Malleability of

Students’ and Parents’ Educational

Expectations

Linda Renzulli 1and Ashley B. Barr 2

1Purdue University, 2University at Buffalo, SUNY

ABSTRACT

Ambitions that students and their parents set during their adolescence have signiﬁcance

across the life course. It is yet unclear, however, how these expectations respond to chang-

ing family circumstances. In this work, we examine how negative family economic shocks

affect educational expectations for students and their parents by using two theoretical per-

spectives – status attainment and adopt-adapt models. Further, we move beyond these de-

bates about the malleability of expectations by considering how this malleability might differ

by your place in the social structure. We thus make hypotheses about how social class may

buffer or amplify the effect of negative family economic shocks on educational expectations.

Finally, and perhaps most importantly, we expand our conceptualization of educational

expectations beyond degree expectations to include educational institutional route expecta-

tions—the educational pathway that students plan to take to achieve their degree expecta-

tions. We ﬁnd that degree expectations are only somewhat malleable, but that route expec-

tations are malleable. Family economic shocks served to reduce students’ and parents’

expectations for beginning their post-secondary education at a B.A.-granting institution.

Further, we ﬁnd support for the ampliﬁcation hypothesis rather than the buffering hypothe-

sis; the expectations of middle-class students and parents prove more negatively responsive

to family economic shocks than do those of their lower SES counterparts. This work has

implications for examining the dynamic nature of the status attainment process and sug-

gests that expanding educational expectations to include institutional route may be vital for

understanding social mobility in the current educational climate.

KEYWORDS : education, degree expectations, college route, socioeconomic status, family

economics.

Access to higher education is the most important route to economic mobility. The initiation and

completion of higher education is now a central task of the transition to adulthood, one with

trajectory-modifying potential (Settersten and Ray 2010a,Furstenberg 2010). The transmission of in-

equality through access to higher education starts much earlier than this, however (Bozick et al.

2010). It is in adolescence when students start to establish expectations for their future (Steinberg

et al. 2009) and to prepare for the transition to adulthood (Settersten and Ray 2010a,2010b).

Authors would like to thank Elizabeth Stearns, Jeremy Reynolds, Maria Paino, and the anonymous reviewers for their thoughtful

comments and insights. Direct all correspondence to: Linda Renzulli, Department of Sociology, Purdue University, West Lafayette,

IN 47907. Email: [email protected].

VCThe Author 2017. Published by Oxford University Press on behalf of the Society for the Study of Social Problems.

351

Social Problems, 2017, 64, 351–372

doi: 10.1093/socpro/spw052

Advance Access Publication Date: 28 February 2017

Article Uncertainties or glitches in students’ expectations during adolescence, then, may yield substantial

implications for ambitions, paths, and opportunities for higher education. Further, when coupled

with the changing structure of higher education regarding access to loans, costs, and returns (Brand

and Xie 2010), even temporary setbacks in adolescence may have lasting effects on paths of social

mobility.

The current literature on adolescent educational expectations focuses primarily on students’ expec-

tations for degree attainment—that is, what is the highest degree students expect to achieve. With re-

gard to these expectations, the literature offers two seemingly contradictory findings. First, degree

expectations are indeed predictive of educational attainment and social mobility (Domina et al. 2011,

Schneider and Stevenson 1999,Wilkins 2014). Second, degree expectations in the US have risen dra-

matically over the past two decades. In fact, as many as 90 percent of students expect to go to college

and earn a Bachelor’s Degree (BA)(Reynolds et al. 2006,Vuolo et al. 2014). This paradox—that ado-

lescent degree expectations are predictive of how the life course unfolds but that the overwhelming

majority of adolescents now expect a BA—has led to debate concerning the meaning and malleability

of degree expectations. Are these expectations simply a mechanism in the transmission of inequality,

a representation of broader cultural schemas, or might they respond to changing circumstances?

In pursuing this line of inquiry, we build upon the existing literature in several ways. First, we do

not assume relative stability in either the conditions that foster degree expectations or in these expec-

tations themselves. Instead, we explore how degree expectations are marred by family economic

shocks over which adolescents have little or no control. Second, we examine how negative family eco-

nomic shocks affect not only youths’ degree expectations but also parents’ degree expectations for

their adolescents. Third, we consider how students and parents might respond differently to family

economic shocks by social class.

Finally, and perhaps most importantly, although key questions remain about the meaning and mal-

leability of degree expectations, the general inflation of these expectations over time suggests that we

must begin to expand our conceptualization of educational expectations beyond degree expectations.

In fact, Rosenbaum (2001) argued that the college-for-all model has essentially made degree expecta-

tions today meaningless and distinguished these expectations from those ofinstitutional route expecta-

tions—the educational pathway that students plan to take to achieve their degree expectations. These

institutional route expectations might include, for example, if a student plans to enroll in a BA-

granting institution first or if they plan to attend a two-year school and then transfer to a four-year

school. Conceptualizing educational expectations as not onlywhatstudents expect to achieve but also

howthey expect to achieve it is a new frontier for this literature and may provide insights to the way

mobility is facilitated or foiled in the contemporary educational landscape.

Thus, the story we tell about education and family setbacks is one of the interweaving relation-

ships between the role of social class, options within the higher education landscape, and changing ex-

pectations—both degree and route—for educational attainment.

THE MALLEABILITY OF STUDENTS’ DEGREE EXPECTATIONS

Three theoretical perspectives help us to understandthe potential malleability of students’ degree expec-

tations (i.e.the credentialstudents expect to earn) in response to family economic shocks. Two of these

perspectives —status attainment and college-for-all—suggest relative stabilityin degree expectations,

while the third perspective —adopt-adapt—suggests malleability given changing social contexts.

1 We conceptualize and operationalize routes in this work as educational routes. Because some students, for example, may take a

path through employment to higher education and their BA, our route conceptualization does not tap the multitude of pathways

students may pursue. In our work, however, we define route as theeducational institutionalroute (i.e. BA institution or not) rather

than life paths toward the BA.

352 Renzulli and Barr Status Attainment and College-For-All

The process of socioeconomic attainment is typically construed as one that flows from schooling ori-

entations and achievements to degree attainment to occupational status (Blau and Duncan 1967,

Warren et al. 2002). This process, however, starts before adolescence (Bozick 2009). The Wisconsin

model of educational attainment showed us that social origin matters for college going, in part be-

cause it shapes degree expectations (Haller and Portes 1973). These early degree expectations, then,

have a strong influence on future success in school (Spenner and Featherman 1978) and on attain-

ment (Schneider and Stevenson 1999,Shanahan 2000).

The Wisconsin model linking social origin to adolescent degree expectations and these expecta-

tions to attainment has been challenged recently by rapidly rising degree expectationsdespiteincreas-

ing income inequality and wide disparities in educational attainment. AsRosenbaum (2011)argues,

this seeming contradiction may be attributed to the college-for-all ethos. The college-for-all model of

degree expectations suggests that, rather than being rooted in social origins and a realistic assessment

of opportunities, the calculations students are making about their prospects may now be the result of

erroneous information and social pressure (Alexander and Cook 1979,Alexander et al. 1994,

Kerckhoff 1977,Reynolds et al. 2006,Rosenbaum 2011). In fact, asSettersten and Ray (2010)poeti-

cally describe: “...youths’ expectations...have risen like a helium balloon in thin air” (p. 5).

Reynolds et al. (2006)add that “adolescents’ expectations have risen so rapidly that they have be-

come increasingly unrealistic, compounding the need for further research on both the meaning of

and the trends in teenagers’ future plans” (p. 188).

Although offering different predictions regarding the role of degree expectations in actually pre-

dicting degree attainment (Domina et al. 2011), both the college-for-all argument and the Wisconsin

model of degree expectations are embedded in the same assumption about the rigidity of adolescents’

expectations. They both assume that degree expectations are a relatively static mental state, formed

either by social origin or by widespread cultural ideals (Haller and Portes 1973,Sewell et al. 1969).

Based on these models, one would expect family negative economic shocks, independent of family

SES, to have little or no impact on degree expectations.

Adopt-Adapt

The assumption that degree expectations are relatively static is challenged theoretically and empiri-

cally by the adopt-adapt framework (Andrew and Hauser 2011,Karlson 2015,Morgan 2005). This

framework suggests that students, in fact, modify their beliefs to changing circumstances (Bozick

et al. 2010,Breen 1999,Morgan 1998). That is, consonant with a social learning perspective, these

beliefs are “adoptedfrom significant social others” (Andrew and Hauser 2011:498) early in educa-

tional careers but may beadaptedas students move through their education and encounter some-

times changing social contexts. The adopt-adapt framework, then, gives us a motivation for

reexamining degree expectation formation and educational choices. The adaptation model suggests

that signaling devices such as high school achievement, counseling advice, and financesshouldbe

used to temper students’ expectations for degree attainment. Like the Wisconsin model, the adapta-

tion model views degree expectations as rooted in reality (Schneider and Stevenson 1999) but does

not assume stability of that reality or the expectations formed from it. Rather, it suggests that as sig-

nals change over an adolescent’s school career, his or her degree expectations change accordingly.

Our first theorizing about the adopt-adapt model may very well have been Burton’s (1960) thesis

on cooling out– the idea that students adjust their expectations for a BA once in a two-year institu-

tion. Empirical tests of the adaptation model, however, are quite new and mixed in their findings. For

instance, recent studies of community college women (Nielsen 2015) and of Katrina survivors

(Deterding 2015) show that students “hold steady” in their ambition even when the odds turn against

them. Researchers have also found, however, that some high schoolers responded to social influences

2 We note that our focus is on negative economic shocks- those that produce a loss rather than a windfall.

Adapting to Family Setbacks 353 from home and school (e.g. school tracking) and adjusted their expectations downward (Bozick et al.

2010,Karlson 2015). Typically these downward shifts are among low-resource students, whose expecta-

tions may rebound during the transition to adulthood (Alexander et al. 2008,Deterding 2015).

This mixed evidence for the adopt-adapt model suggests that adolescents subject to the college-

for-all ethos do not easily change their degree expectations based ongermaneevidence such as aca-

demic achievement (Andrew and Hauser 2011,Moller et al. 2011). Degree expectations may still be

adaptable with the occurrence of adireexogenous change, however. We propose that family negative

economic shocks might qualify as a dire exogenous change capable of altering students’ degree

expectations.

The financial burdens of higher education are increasingly falling on the shoulders of parents, as

aid programs are being shifted from federal grants to low-cost loans to private loans or to families’ di-

rect contributions to education (Hamilton 2013,Lucas 1996,Paulsen and St John 2002). To send

their children to college, parents often make difficult financial decisions, and it comes at a substantial

price tag, as they use savings or retirement accounts. This heavy reliance on parents to fund and sup-

port students throughout higher education means that sudden changes in a family’s economic status

might qualify as the “sharp and drastic exogenous changes, or shocks” thatAndrew and Hauser

(2011: 514) suggest might be necessary to cause substantial change in students’ degree expectations.

Hence, while the null hypothesis guided by the Wisconsin and college-for-all models suggests that

student degree expectations are not malleable, the adopt-adapt hypothesis predicts that when families

incur significant and negative economic shocks, students’ expectations for earning a BA will falter.

Hypothesis 1:Negative family economic shocks will decrease students’ expectations for a BA.

Given the growing role of parents in supporting their children in their educational pursuits, par-

ents’ degree expectations continue to play an increasingly important role in their children’s educa-

tional trajectories (Briley et al. 2014). We know that higher socioeconomic status parents are more

likely to assume an attitude that their children will attend college, come to expect it, and save the re-

sources to pay for it (Flint 1992,1997, Hamilton 2013,Olson and Rosenfeld 1984). Understanding

how not only students’ but also parents’ degree expectations change in response to family economic

shocks is a step toward understanding the relational underpinnings of status attainment in the context

of an extended transition to adulthood (Settersten and Ray 2010). Although parents’ degree expecta-

tions have largely been left out of debates on the meaning and malleability of educational expecta-

tions to date, one might presume that, if the adopt-adapt model is correct, parents’ degree

expectations for their adolescents might also be responsive to family setbacks.

Hypothesis 2:Negative family economic shocks will decrease parents’ expectations for their

high schooler to earn a BA.

IMPORTANCE AND MALLEABILITY OF INSTITUTIONAL ROUTE

EXPECTATIONS

Because the college-for-all ethos is so strong and growing among all social classes, and collegedegree

expectations have not yet proven highly adaptable among recent cohorts of students (Andrew and

Hauser 2011,Nielsen 2015), it may very well be that family economic shocks do not make students

or parents waver in their expectations for earning the BA credential. This would not necessarily mean

that family economic shocks are harmless, however. Although the Wisconsin model centers around

the role of degree expectations in predicting attainment,howfamilies expect to meet their goals is

also important (Karlson 2015,Kerckhoff 1976). Here we expand our understanding of educational

expectations to include institutional route expectations. Though students are increasingly expecting a

BA, they are also increasingly expecting to start their path to earning that BA at a two-year institution

354 Renzulli and Barr (Johnson and Reynolds 2013). In the 1970s and 1980s, less than 50 percent of students who went to

two-year institutions expected to earn a BA; they expected a two-year degree and an occupation com-

mensurate with that degree. As the costs associated with attending a four-year institution keep rising,

and families face new financial burdens in the aftermath of the Great Recession, community colleges

provide an increasingly attractive and affordable entry for many students who aspire and expect to

earn a BA (Wang 2013). A two-year path may help satisfy college-for-all demands for students. In

fact, states today are pushing the two-year route with elaborate articulation policies and transfer credit

guarantees, with some success.

Nonetheless, this path is much riskier for students: four-year institutions tend to maintain higher

degree attainments than two-year institutions despite students’ expectation to receive a BA. More

than two-thirds of students who start their postsecondary career in a community college identified

the BA as their aim (Hoachlander et al. 2003,Kojaku and Nunez 1998,Schneider and Stevenson

1999). The actual rate of transfer is much lower, with only 36% of all BA-seeking students at two-

year colleges transferring to four-year colleges (Bradburn et al. 2001,Dougherty and Kienzl 2006).

Even in good economic times, students use financial concerns to make the cost-benefit analysis of

going to a two-year or a four-year institution (Deil-Amen and Lopez Turley 2007). As two-year

schools have expanded in their organizational form and market-share (Goldrick-Rab 2010), along

with the increasing financial burdens on families, it is easy to predict that changing family circum-

stances would have consequences for the institutional routes that students plan to take to degree at-

tainment. That is, even if students’ and parents’ degree expectations are not substantially altered by

family economic shocks, they may now expect to take the less expensive route there – the community

college (Dougherty and Kienzl 2006). Hence, we extend the predictions of the adopt-adapt model to

educational route expectations.

Hypothesis 3:Negative family economic shocks will reduce the likelihood that students will ex-

pect to begin postsecondary education at a four-year institution.

Hypothesis 4:Negative family economic shocks will reduce the likelihood that parents will ex-

pect their high schoolers to begin postsecondary education at a four-year institution.

FAMILY SES AND THE EFFECT OF FAMILY ECONOMIC SHOCKS FOR STUDENT

AND PARENTS

The debates in the literature thus far concerning the malleability of educational expectations have

largely been concerned with the presence or absence ofdirect effects.For instance, the status attain-

ment and college-for-all models suggest no effect of economic shocks on expectations, while the

adopt-adapt model suggests a negative effect of shocks on expectations. This debate is somewhat lim-

ited, however, because there is strong reason to believe that any malleability of expectations, if found,

may be different for some families than for others. That is, the degree to which students and parents

respond to changing circumstances may be contingent upon their relative position in the social struc-

ture. For some, this position may limit opportunity in the face of financial struggle and for others this

position may open opportunity in the face of struggle (Johnson and Reynolds 2013,Turley et al.

2007). Hence, we explore the role of social class on conditioning the effect of family economic shocks

on degree and route expectations.

Family SES as a Potential Buffer

The cumulative advantage students and parents gain by their middle-class status may buffer them

from financial setbacks (Furstenberg Jr 2010,Merton 1988). Students from middle- and higher-SES

families have traditionally been more likely than their lower-SES peers to hold onto their expectations

to earn a BA (Bozick et al. 2010,Johnson and Reynolds 2013,Wang 2012). This has been argued to

Adapting to Family Setbacks 355 be the case because of the ingrained habitus among the middle class for college-going (Grodsky and

Riegle-Crumb 2010). Further, the schools of middle-class students may be equipped to offer assis-

tance in the event of family crises. Middle-class parents are prepared with the cultural and social capi-

tal necessary to find alternative means to pay for higher education should family resources suddenly

diminish (Charles et al. 2007,Lareau 1987,Paulsen and St John 2002) and likely have financial capi-

tal already saved (Hamilton 2013). This advantage in the transmission of status attainment predicts

that middle-class status buffers students and parents from the negative effects of family economic

shocks on expectations.

Intuitively one would surmise then that middle-class families might also be less likely to alter their

expected educational path from a four-year to a two-year college in the face of family economic

shocks. The benefits of middle-class neighborhoods, schools, parenting and lifestyles increase entitle-

ment to four-year schooling (Calarco 2014,Lareau 2002,Lareau and Goyette 2014), and years of

preparation before the setback might make path switching more unlikely and unnecessary (Grodsky

and Riegle-Crumb 2010).

Hypothesis 5a:The negative effect of family economic shocks on degree expectations will be

weaker among more middle-class students and parents than among lower SES counterparts.

Hypothesis 5b:The negative effect of family economic shocks on expected route to BA will be

weaker for more middle-class students and parents than for their lower SES peers.

Family SES as a Potential Amplifier

Though intuitively it may seem that those closer to the top of the socioeconomic ladder may be

more likely to maintain high educational expectations in the face of economic shocks, as predicted in

hypotheses 5a and 5b, there is new evidence that suggests the opposite: that being in the middle class

may in fact amplify the negative effects of a shock (Deterding 2015,Nielsen 2015). Those in the mid-

dle class have relied most heavily on savings, income, and access to private loans for college

(Hamilton 2013). Family economic shocks for these individuals might mean that they are suddenly

unable to save for college or are forced to use college savings for everyday living. On the other hand,

lower-SES students and parents may be more accustomed to economic shocks than their middle-

class peers (Lucas 2001). Lower-SES families often have little savings and instead rely on access to

public loans and low estimated family contributions to obtain college degrees; i.e. if you have nothing,

you have nothing to lose. Hence, it is possible that lower-SES families may not see what is likely to

be just another negative financial event as detrimental to their plans to obtain a four-year degree,

while more middle-class families may be so thrown off track that they reevaluate their degree expecta-

tions altogether (Karlson 2015).

Even if middle-class students hold steady in their expectation to earn a BA, however, they may

reevaluate their institutional route to achieve the BA. When middle-class families encounter eco-

nomic shocks they may see this as a blip in their financial stability and may feel able to wait it out in

a two-year college before transitioning to a four-year institution and earning the desired BA. And,

compared to their lower SES counterparts, the odds are in their favor that they will be able to obtain

that degree (Dougherty and Kienzl 2006,Wang 2013). On the other hand, lower SES students and

parents may be skeptical that a two-year school will lead them to a BA (Ovink 2014). Thus, during

times of even greater financial instability than is their norm, their expectations for a four-year institu-

tion may not falter (Deterding 2015). This counter-intuitive notion of institutional route expecta-

tions is, however, consistent with research showing that sanguine educational aspirations exist

despite unlikely probability for accomplishment (Frye 2012). Hence, here we offer competing pre-

dictions from hypotheses 5a-b.

356 Renzulli and Barr Hypothesis 6a:The negative effect of family economic shocks on degree expectations will be stron-

ger among more middle-class students and parents than among lower-SES students and parents.

Hypothesis 6b:The negative effect of family economic shocks on expected route to BA will be

stronger for more middle-class students and parents than for their lower-SES peers.

METHODS

Data and Sample

To test our hypotheses, we use data from the High School Longitudinal Study of 2009 (HSLS:09).

The HSLS:09 is a nationally representative cohort study of 9 thgraders across the United States. The

HSLS:09 is designed to capture students’ and families’ preparations for and transitions to postsecon-

dary education, as well as paths into and out of science, technology, engineering, and mathematics

(STEM) fields. The first wave of data collection took place during the fall of the 2009-2010 school

year. A follow-up was completed in 2012 when most students were in the spring term of 11

thgrade.

The sample design was a stratified, two-stage random sample with schools selected at the first

stage and students within those schools selected at the second stage. Hence, the sample is nationally

representative of 9

thgraders in 2009-2010 and of schools with 9 thand 11 thgraders in 2009. Sampling

weights are used to account for the complex sampling design.

In the absence of data that examine our key concepts immediately before and after the Great Recession,

the HSLS:09 is ideal for studying the impact of family economic shocks on educational expectations. First,

these data ask not only about educational attainment (i.e. degree) expectations but also about the type of in-

stitution that students expect to attend immediately following high school (i.e. institutional route expecta-

tions). Second, the HSLS:09 data contain informationon degree and institutionalroute expectations from

both students and parents. Third, degree and route expectations were assessed across two waves of data, en-

abling an assessment ofchangein expectations for both parents and students. Lastly, the timing of the

HSLS:09 is important. The first wave of data collection took place in 2009, just after the Great Recession

had officially ended. Although unable to assess the impact of the recession on families, the aftermath of this

recession provides a unique opportunity to offer insight into the malleability (or lack thereof) of educational

expectations today. The negative effects of the recession lingered long after the recession’s official end and

were not isolated to the poor (Grusky et al. 2011). Because many lower- and middle-SES families experi-

enced substantial setbacks in the aftermath of the recession, this period enables us to examine the degree to

which educational expectations might be responsive tochanging family circumstances, such as a foreclosure

on a home or a job loss, with more limited selection biasthan would be expected in other historical contexts.

To test our hypotheses regarding the effect of family economic shocks on educational expecta-

tions, we limit our analytic sample in several ways. Although over 23,000 students completed the

base year HSLS:09, only a random subsample of parents was targeted at each wave. Because all eco-

nomic shock data was gathered from the parent questionnaires, the analytic sample was first reduced

to those respondents with two waves of student and parent data (N¼8,650).

3We further limited

our analytic sample to students enrolled in public schools (N¼6,640). By focusing on public school

students, we maintain class variation but limit the range largely to middle- and working-class respon-

dents rather than include upper-class families with theoretically and empirically different experiences.

STUDY MEASURES

Dependent Variables

Student degree expectations.In grades 9 and 11, students were asked “As things stand now, how far in

school do you think you will get?” Responses ranged from “less than high school” to “complete a

3 Unweighted sample sizes have been rounded to the nearest ten as per NCES restrictions.

Adapting to Family Setbacks 357 Ph.D., M.D., law degree, or other high level professional degree.” Because response options varied

somewhat between waves, we used a simple yet comparable indicator of degree expectations at each

wave by dichotomizing the measure into 1¼expects BA or higher and 0¼does not expect BA.

Student institutional route expectations.In grades 9 and 11, students were asked about their educa-

tional plans immediately following high school. In grade 9, students were asked “What do you plan

to do during your first year after high school?” Students were coded 1 for expecting to enroll in a BA-

granting program if they answered affirmatively to the “enroll in a Bachelor’s degree program in a col-

lege or university” and 0 otherwise. In grade 11, students were asked: “In the fall of 2013, are you

most likely to attend a school that provides occupational training, a two-year college, a four-year col-

lege, or have you not thought about this yet?” Again, students were coded 1 for expecting to enroll in

a BA-granting program if they chose “four-year college” and 0 otherwise. Although our measure of

route expectations is limited to a simple dichotomy between BA institution and something else, this

measure is directly comparable to the measure of degree expectations. Without comparable measures,

it would be misguided to compare the effects of economic shocks on degree and route expectations.

Parent degree expectations.Parents’ degree expectations were assessed with questions similar to stu-

dents’ degree expectations. At each wave, the parent or guardian was asked “As things stand now, how

far in school do you think [student] will actually get?” As with students’ response options, parent re-

sponse options varied somewhat across waves but ranged from “less than high school” to “complete a

Ph.D., M.D., law degree, or other high level professional degree.” Responses were dichotomized into 1

¼expects BA degree or higher for student and 0¼doesnotexpectBAdegreeforstudent.

Parent institutional route expectations.At each wave, parents were asked where they expected their

high schooler to enroll after high school. More specifically, when students were in the 9

thgrade, par-

ents were asked “Do you think [your 9th grader] will start [his/her] college education at a technical

institute, a community college or other Associate’s granting school besides a technical institute, or a

Bachelor’s granting 4-year college?” Parents who responded affirmatively to the latter option were

coded 1 for expecting the student to enroll in a BA-granting program and 0 otherwise. Parents who

indicated that they “haven’t thought about this yet” were coded 0 for having no affirmative expecta-

tions for a BA-granting institution. Similarly, when students were in the 11

th grade, parents were

asked “If [teenager] attends school in the fall of 2013, will [he/she] be most likely to attend a school

that provides occupational training, a two-year college, a four-year college, high school, or have you

not thought about this yet?” Again, parents who chose “four-year college” were coded 1 for expecting

the student to enroll in a BA-granting program and 0 otherwise. As with the student expectation mea-

sures, our measures of parental degree and route expectations are simple yet directly comparable.

In all models, the 11

th-grade expectations serve as the dependent variables, while the 9 th-grade ex-

pectations serve as control variables.

Independent & Moderating Variables

Economic shocks.Family economic shocks were assessed in multiple ways. First, when students were in

11thgrade, parents were asked explicitly if, since the fall of 2009 (the time of the 9 thgrade survey), the

teenager’s parent or guardian has “lost a job” (lost job,1¼yes, 0¼no) or the “family’s home was

foreclosed” (foreclosed,1¼yes, 0¼no). Second, family income categories (ranging from “$15,000 or

below” to “more than $235,000” in $20,000 increments) were compared across waves. If the family

dropped in income categories from wave 1 to wave 2, they were coded 1 for having experienced a loss

of income (lost income,1¼yes, 0¼no), hence providing a conservative estimate of income loss.

4 Although this assessment does not capture the relative magnitude of income loss, it is largely unaffected by threshold effects and,

hence, not collinear with SES. A measure of the number of income categories a family dropped across waves or a measure of raw

income difference in dollars, which would assess magnitude of the income loss, would be constrained by SES given that only fam-

ilies with higher initial incomes can fall far. This problem is illustrated by the significant, positive correlation between SES and in-

come loss when measured as a continuous variable (p<.001) but the nonsignificant correlation between SES and income loss

when measured categorically (p¼.180).

358 Renzulli and Barr Finally, insecurity in the family’s housing situation across waves was assessed by comparing their hous-

ing status at wave 1 with their housing status at wave 2. At each wave, the parent respondent was

asked: “Do you (1) pay mortgage towards or own your home, (2) rent your home, or (3) have some

other arrangement?” Decreased housing security was indicated by a shift from category 1 to either cat-

egory 2 or 3, or a shift from category 2 to category 3. Respondents were coded 1 if the family experi-

enced a decrease in housing security and 0 otherwise (housing insecurity,1¼yes, 0¼no). A count

measure of economic shocks was constructed by summing these four indicators. The most common

economic shock was having a parent lose a job (24.8%), followed by a loss of income (24.6%), hous-

ing insecurity (7.9%), and home foreclosure (5.2%). Overall, 44% of public high schoolers experienced

at least 1 family economic shock between the fall of 2009 (9

thgrade) and the spring of 2011 (11 th

grade), supporting claims of the lingering effects of the recession long after its official end. Because

less than 1% of students experienced all 4 economic shocks, the final count variable was restricted to

four levels (0, 1, 2, 3 or more), although models using the full continuous measure were consistent

with those presented below.

Socioeconomic status.An index ofsocioeconomic status (SES) was computed by NCES. Included

in this index were five indicators of students’ home background (parent report used, if available), in-

cluding (1) the highest education among parents/guardians in the two-parent family of a responding

student, or the education of the sole parent/guardian, (2) the education level of the other parent/

guardian in the two-parent family, (3) the highest occupation prestige score among parents/guard-

ians in the two-parent family of a responding student, or the prestige score of the sole parent/guard-

ian, (4) the occupation prestige score of the other parent/guardian in the two-parent, (5) family

income. These items were standardized and averaged to form the SES index at wave 1 (9

thgrade).

This index is identical to SES measures in previous NCES cohort studies, like NELS:88 and

ELS:2002. For the sake of simplicity, we refer to lower-SES and more middle-SES students through-

out the results, but it is important to keep in mind the continuous nature of the SES index.

Control variables.We employ a series of individual, family, and school control variables to properly

specify the primary association of interest, that between economic shocks and educational expecta-

tions. All controls were assessed in 9

thgrade unless otherwise noted, and they include student gender

(female¼1), family structure (not living in married biological family¼1), race (non-white¼1),house-

hold size(number of people living in household), whether or not the studentrepeated a gradeby

grade 9 (1¼yes), whether or not the student transferred schools between grade 9 and 11 (transfer

¼1), and whether or not the parents or someone in the family has started aneducational savings ac-

count(1¼yes). Educational savings have been associated with higher college expectations, atten-

dance, and completion (Elliott and Beverly 2011,Zhan and Sherraden 2011) and may also serve to

cushion students and parents from economic shocks.

We also controlled for several factors indicating students’ preparation for and ability to complete a

BA. These include students’ academic track in 9

thgrade, as indicated by their enrolled math class.

Consonant with past work (Adelman 1999,Attewell and Domina 2008,Karlson 2015), students

were coded as being on aremedial trackif they were enrolled in basic math, pre-algebra, or no math

course.Nonselective college trackwas indicated by enrollment in algebra 1, andselective college track

was indicated by enrollment in an advanced math class, like geometry, algebra 2, or pre-calculus. We

also control for students’ math ability, as indicated by the students’ theta scores on the 9

th grade

mathematics assessment. These ability scores provide a summary measure of achievement and are

recommended by NCES to assess algebraic reasoning ability (NCES, 2013).

5 We also explored models with a dichotomous shock indicator (0¼no shocks, 1¼1 or more shocks). These models were largely

consistent with those presented, but the effects were generally weaker, indicating that the number of shocks, a measure that better

taps severity or intensity, and not simply the presence of 1 or more shocks, matters.

6 In the fall of 2009 (9

thgrade), students completed a mathematics assessment of algebraic reasoning designed to assess under-

standings of the major domains and key processes of algebra. Scores used to describe students’ performance on the mathematics

assessment are based on the Item Response Theory model (Hambleton & Swaminathan, 1985), which uses patterns of correct,

Adapting to Family Setbacks 359 In addition to these individual- and family-level controls, we constructed an index of the high

school’scollege preparatory resourcesby summing the number of programs (10 total) the school coun-

selor indicated were available to students. These programs included a counselor designated for col-

lege readiness/applications, college fairs, consultation with postsecondary representatives, organized

visits to colleges, college preparatory programs (e.g. Upward Bound), information sessions on the

transition to college, assistance with finding financial aid, counseling curriculum for positive academic

behaviors, and other steps to assist with the transition to college. The availability of these programs

are associated with students’ expectations and preparation for college and may also be associated

with students’ or parents’ experience of or response to economic shocks.

Lastly, as mentioned above, we control for 9

thgrade degree and route expectations in all models

predicting 11 thgrade expectations so that we account for a baseline assessment of expectations prior

to family economic shocks. It is important to note, however, that we have no assessment of economic

shocks prior to the 9

thgrade. Hence, it is possible that families could have experienced other negative

shocks of a similar or different nature prior to 9 thgrade and then additional shocks during the time

period captured by HSLS:09. This possibility, however, makes our estimates of the effect of economic

shocks on educational expectations rather conservative.

ANALYTIC STRATEGY

Given our dichotomous dependent variables, we utilized logistic regression models with robust

standard errors in MPlus Version 7. Logistic regression in MPlus has several advantages. First,

MPlus allows one to model the covariance between independent variables and to predict two out-

comes (e.g. degree expectations and route expectations) simultaneously. Doing so enables more

conservative estimates of the associations of interest. Second, because the HSLS:09 is not a simple

random sample of students, analysis of the data requires the use of sampling weights to account for

the complex sampling design and ensure representativeness. MPlus allows for the relatively straight-

forward use of sampling weights using Taylor Series Linearization. This approach takes into ac-

count the primary sampling unit (schools) and strata identifiers provided by NCES to calculate

appropriate standard errors. Third, MPlus allows one to retain cases with missing data via a full in-

formation maximum likelihood (FIML) estimator. FIML uses all available data yet is more efficient

and relies on far fewer decisions than multiple imputation (Allison 2012). Hence, we utilized the

imputed data for family SES provided by NCES (see HSLS:09 Data File Documentation at http://

nces.ed.gov/pubs2014361.pdf) and relied on FIML when estimating models with additional miss-

ing data.

For each outcome of interest, we ran two models: (1) a direct effects model in which the effects

of SES and family economic shocks, independent of controls, were estimated, and (2) an interactive

model examining the interaction between SES and family economic shocks. All continuous variables

were centered at their mean. In presenting each of the models below, we report the exponentiated co-

efficients, or odds ratios. Coefficients represent the factor change in the odds of expecting a BA de-

gree/institution given a 1-unit change in the independent variables. For instance, an odds ratio of 1.4

indicates that a one-unit increase in the independent variable increases the odds of affirmative expec-

tations by 1.4 times, or 40%.

RESULTS

Univariate and Bivariate Statistics

Before discussing the results of the logit models, several univariate and bivariate statistics are worth

noting. First, as shown inTable 1, BA degree expectations for both parents and students were

incorrect, and omitted responses to obtain ability estimates. Although NCES offers several types of scores based on the mathe-

matics assessment, ability (or theta) scores are used as a control here given that NCES notes they are best for assessing students’

absolute ability.

360 Renzulli and Barr relatively high in both 9 th(81% of students and 83% of parents) and 11 thgrade (70% of students and

73% of parents). Degree expectations were substantially lower for 11 thgraders than for 9 thgraders,

however. As for route expectations, about half of parents (46%) and students (53%) in the 9thgrade

expected to be enrolled in a four-year institution following high school. By 11 thgrade, both parents’

and students’ route expectations increased slightly, with 51% of parents and 59% of students expect-

ing a BA institution. Second, there is only a moderate correlation between degree and route expecta-

tions for both students (r

9th ¼.328,r 11th ¼.586) and parents at each wave (r 9th ¼.398,r 11th ¼

.576; correlation matrix available upon request). Further, there is only a moderate correlation be-

tween expectations across waves for both students (r

degree ¼.349,r route ¼.354) and parents (r degree ¼

.505,r

route ¼.502). Lastly and most importantly, there is only a small negative correlation between

SES and economic shocks (r¼ .070), supporting the contention that the residual effects of the

Table 1. Descriptive Statistics for Study Variables

Weighted Mean Std. Dev. Min Max

9th grade

Student expects BA degree 0.811 0.000 1.000

Student expects BA institution 0.530 0.000 1.000

Parent expects BA degree 0.825 0.000 1.000

Parent expects BA institution 0.461 0.000 1.000

Female student 0.489 0.000 1.000

Not living in married, biological family 0.416 0.000 1.000

Educational savings account 0.197 0.000 1.000

Student repeated a grade 0.113 0.000 1.000

Transferred schools between grade 9 and 11 0.113 0.000 1.000

Student is non-white 0.485 0.000 1.000

Household size 4.376 1.344 2.000 13.000

School resources 6.802 2.629 0.000 10.000

Remedial track 0.141 0.000 1.000

Nonselective track 0.512 0.000 1.000

Selective track 0.346 0.000 1.000

Math ability 0.048 0.930 2.570 3.028

Family SES 0.087 0.689 0.999 2.567

Between 9th and 11th grades

Number of economic shocks 0.589 0.736 0.000 3.000

Any economic shock 0.440 0.000 1.000

Housing insecurity 0.079 0.000 1.000

Foreclosure 0.052 0.000 1.000

Change in income category 0.246 0.000 1.000

Parent lost job 0.248 0.000 1.000

11th grade

Student expects BA degree 0.696 0.000 1.000

Student expects BA institution 0.587 0.000 1.000

Parent expects BA degree 0.730 0.000 1.000

Parent expects BA institution 0.514 0.000 1.000

Note:Sample is restricted to public school students. N¼6640; Unweighted sample sizes have been rounded to the nearest ten as per NCES

restrictions.

Adapting to Family Setbacks 361 Great Recession were not limited to the poor and working class but also affected more middle-class

families.

Degree Expectations

Table 2presents results predicting 11 th graders’ and their parents’ degree expectations—that is,

whether or not a BA is expected— from SES, economic shocks, and their interaction, independent of

all controls. As can be seen in this table, family SES was associated with higher degree expectations

among students. Further, as shown in model 1, economic shocks were only marginally associated

with students’ lower degree expectations. Independent of 9

thgrade expectations and all controls, each

family economic shock was associated with a 14% decrease in the odds of expecting a BA degree in

thgrade. As indicated in model 2 of this table, the negative association between economic shocks

and degree expectations was not moderated by SES. Such findings offer only moderate support for

the direct effect of economic shocks on students’ degree expectations articulated in hypothesis 1 and

no support for the conditional effects outlined in hypotheses 5a and 6a. These findings are shown in

Panel A ofFigure 1.

As with students’ degree expectations, family SES was associated with reduced degree expectations

for parents. Contrary to the findings for students, however, SES moderated the effect of family eco-

nomic shocks on parents’ degree expectations. As shown in model 3 ofTable 2, family economic

shocks were not directly predictive of degree expectations for parents (contrary to hypothesis 2), but

their effect was dependent upon SES (model 4). This interaction effect is illustrated in Panel B of

Figure 1. Here, we utilized coefficients from model 4 to plot the predicted probability of parents who

expected a BA degree in 9

thgradestill expectinga BA degree for their 11 thgrader by number of eco-

nomic shocks and SES. 7As Panel B ofFigure 1illustrates, there is a substantial decline in the proba-

bility of more middle-class parents still expecting a BA for their 11 thgrader with increasing economic

shocks (simple slope test: t¼ 13.377, p<.001). This was not the case for lower-SES parents.

Instead, the effect of economic shocks on lower-SES parents was significantly positive (simple slope

test: t¼5.689, p<.001). These findings offer support for the amplification hypothesis 6b.

Economic shocks were associated with a reduced probability of expecting a BA degree among more

middle-class parents but not lower SES parents, whose expectations actually rose with more eco-

nomic shocks.

Institutional Route Expectations

Our first set of models suggests that there is weak evidence that family economic shocks reduce stu-

dents’ degree expectations and stronger evidence suggesting that they reduce the degree expectations

of parents, but primarily middle-class parents. Despite these effects, degree expectations remained rel-

atively high. For instance, as can be seen inFigure 1, the majority of students and parents who ex-

pected to earn a BA degree in 9

thgrade and experienced 3 or more economic shocks still expected

that degree in 11 thgrade. Hence, we now turn to the models predicting institutional route expecta-

tions—that is,howstudents plan to achieve their degrees.

As shown in model 1 ofTable 3, economic shocks were only marginally significant in predicting

students’ institutional route expectations. As shown in model 2 of this table, however, the effect of

economic shocks on students’ route expectations was dependent upon family SES. In particular, the

negative effect of economic shocks on expecting to attend a BA-granting institution after high school

was significantly stronger for more middle-SES students. This interaction is graphed in Panel A of

Figure 2. As with the figure for parents’ degree expectations, the predicted probabilities represent

those for 9

thgraders who expected a BA. For instance, middle-class students, plotted at 1 standard

7 For illustration purposes only, we refer to 1 standard deviation about the SES mean as “middle SES” and 1 standard deviation be-

low the mean as “lower SES” (which we do in subsequent figures, as well). Coefficients from the continuous SES models were

used to calculate probabilities for these groups, however.

362 Renzulli and Barr Table 2. Logistic Regression of Students’ and Parents’ Degree Expectations in 11th Grade (Odds ratios presented)

Student Expects BA Degree Parent Expects BA Degree

9th grade controls Model 1 95%CI Model 2 95%CI Model 3 95%CI Model 4 95%CI

Student expects BA degree 2.582*** 2.054–3.247 2.585*** 2.058–3.246 — —

Student expects BA institution 2.367*** 1.919–2.919 2.367*** 1.919–2.920 — —

Parent expects BA degree — — 5.640*** 4.389–7.246 5.708*** 4.439–7.338

Parent expects BA institution — — 2.704*** 2.072–3.528 2.714*** 2.084–3.535

Female student 1.442** 1.190–1.747 1.441** 1.189–1.746 1.692*** 1.364–2.099 1.688*** 1.364–2.088

Not living in married, biological family 0.921 0.759–1.116 0.924 0.762–1.119 0.792

†

0.646–0.971 0.803

†

0.656–0.981

Educational savings account 1.290 0.990–1.681 1.291 0.990–1.682 1.275 0.964–1.687 1.280 0.969–1.691

Student repeated a grade 0.729

†

0.536–0.992 0.726

†

0.534–0.988 1.040 0.753–1.437 1.038 0.751–1.436

Transferred schools between grades 9 and 11 1.124 0.662–1.910 1.125 0.660–1.919 0.585

†

0.364–0.941 0.589

†

0.371–0.936

Student is non-white 1.078 0.888–1.309 1.082 0.891–1.313 1.838*** 1.475–2.291 1.859*** 1.491–2.318

Household size 1.027 0.951–1.108 1.026 0.951–1.107 1.037 0.955–1.125 1.032 0.952–1.118

School resources 1.008 0.961–1.057 1.008 0.961–1.057 0.998 0.953–1.046 0.997 0.952–1.045

Remedial track 0.700

†

0.514–0.952 0.701

†

0.516–0.952 0.462*** 0.332–0.644 0.452*** 0.325–0.631

Nonselective track 0.631** 0.470–0.847 0.631** 0.470–0.848 0.600*** 0.462–0.779 0.606** 0.466–0.789

Selective track ref ref ref ref

Math ability 1.570*** 1.391–1.772 1.567*** 1.388–1.769 1.740*** 1.532–1.977 1.732*** 1.524–1.969

Family SES 2.220*** 1.821–2.706 2.315*** 1.804–2.969 1.782*** 1.509–2.105 2.233*** 1.812–2.751

Number of economic shocks 0.858

†

0.746–0.986 0.848* 0.740–0.972 0.927 0.817–1.053 0.871

†

0.774–0.980

x Family SES 0.940 0.755–1.169 0.689** 0.555–0.856

N¼6640 public schoolers.†p<.10, *p<.05, **p<.01, ***p<.001; CI¼conﬁdence interval.

Adapting to Family Setbacks 363 deviation above the mean, who expected to attend a BA-granting institution in 9 thgrade and experi-

enced no economic shocks had a 70% probability of expecting that type of institution in 11 thgrade.

If they experienced 3 or more shocks, however, the predicted probability drops to less than 50%.

Slope tests revealed that the slope for more middle-class students was significant and negative (t¼

2.986, p<.01), whereas the slope for lower-SES students was not significantly different from zero

(t¼.817, p¼.414). Consonant with amplification hypothesis 6b, family SES amplified the negative

effect of economic shocks on students’ expectations to attend a four-year college.

As for parents’ institutional route expectations, family economic shocks directly reduced the odds

of parents’ expecting their high schooler to attend a four-year institution. More specifically, each fam-

ily economic shock was associated with a 16% decrease in the odds of parents expecting their high

schooler to start postsecondary education at a four-year institution (model 3). As indicated by a non-

significant interaction term in model 4 ofTable 3, the negative effect of economic shocks on parents’

institutional route expectations did not significantly differ by social class. This finding offers support

Figure 1.Degree Expectations for Students and Parents by Negative Economic Shocks

*Notes: Predicted probabilities assume that the student/parent had afﬁrmative expectations in 9th grade and so

can be interpreted as the probability that the student/parentstillexpects student to earn a BA. As shown in

Table 2, Model 2, the effect of economic shocks on students’ degree expectations did not vary by SES and

hence is not modeled as such in Panel A. Slopes for middle SES and low SES parents did differ, however, as in-

dicated inTable 2, Model 4. Hence, they are shown in Panel B, and both slopes are signiﬁcant at p<.05.

364 Renzulli and Barr Table 3. Logistic Regression of Students’ and Parents’ Institutional Route Expectations in 11th Grade (Odds ratios presented)

Student Expects BA Institution Parent Expects BA Institution

9th grade controls Model 1 95%CI Model 2 95%CI Model 3 95%CI Model 4 95%CI

Student expects BA degree 1.896*** 1.569–2.292 1.912*** 1.581–2.312 — —

Student expects BA institution 2.383*** 1.984–2.864 2.391*** 1.990–2.872 — —

Parent expects BA degree — — 2.913*** 2.161–3.927 2.912*** 2.159–3.927

Parent expects BA institution — — 4.558*** 3.546–5.859 4.561*** 3.547–5.866

Female student 1.766*** 1.476–2.114 1.763*** 1.475–2.108 1.580*** 1.265–1.973 1.579*** 1.265–1.970

Not living in married, biological family 0.929 0.753–1.146 0.938 0.758–1.160 0.899 0.714–1.131 0.900 0.714–1.133

Educational savings account 1.444** 1.162–1.794 1.448** 1.166–1.797 1.044 0.833–1.308 1.047 0.836–1.310

Student repeated a grade 0.691* 0.535–0.893 0.689* 0.533–0.890 0.667* 0.484–0.920 0.668* 0.484–0.921

Transferred schools between

grade 9 and 110.501*** 0.403–0.622 0.502*** 0.404–0.625 0.690* 0.506–0.941 0.692* 0.508–0.943

Student is non-white 1.255* 1.072–1.469 1.263* 1.079–1.478 1.205 0.991–1.467 1.206 0.991–1.468

Household size 1.044 0.973–1.120 1.042 0.971–1.118 1.053 0.993–1.117 1.053 0.992–1.116

School resources 1.009 0.970–1.049 1.008 0.969–1.048 1.028 0.989–1.069 1.028 0.989–1.068

Remedial track 0.779† 0.578–1.051 0.774 0.574–1.043 0.485*** 0.368–0.638 0.484*** 0.367–0.637

Nonselective track 0.647** 0.501–0.837 0.650** 0.502–0.842 0.638*** 0.514–0.791 0.640*** 0.516–0.795

Selective track ref ref ref ref

Math ability 1.576*** 1.412–1.759 1.569*** 1.405–1.752 1.763*** 1.588–1.958 1.763*** 1.588–1.957

Family SES 1.613*** 1.406–1.851 1.860*** 1.529–2.263 1.558*** 1.286–1.887 1.599*** 1.256–2.036

Number of economic shocks 0.901† 0.822–0.988 0.880* 0.802–0.965 0.836** 0.754–0.925 0.833** 0.753–0.922

x Family SES 0.786* 0.651–0.950 0.953 0.796–1.142 N¼6640 public schoolers.†p<.10, *p<.05, **p<.01, ***p<.001.

Adapting to Family Setbacks 365 for hypothesis 4 but not for the conditioning hypotheses specified in 5b and 6b. The direct effect of

shocks on parents’ route expectations is graphed in Panel B ofFigure 2.

Post-hoc Analyses: Does the Type of Economic Shock Matter?

Thus far, we have found that economic shocks are associated with parents’ and students’ educational

expectations, especially those of middle-class students and parents. Although not part of our hypothe-

ses, we went on to examine if the type of shock matters for these effects. That is, are the expectations

of middle-class students and parents more likely than those of low-SES students and parents to be af-

fected byallshocks or particular types of shocks? In supplemental models, we examined the poten-

tially unique effect of each type of shock, controlling for the presence of other shocks as well as the

extensive set of control variables used thus far. It is important to note that relatively small cell sizes

limit these models by class, particularly when the tendency to experience multiple types of shocks is

Figure 2.Institutional Route Expectations for Students and Parents by Negative Economic Shocks

*Notes:Predicted probabilities assume that the student/parent had afﬁrmative expectations in 9th grade and so

can be interpreted as the probability that the student/parentstillexpects student to attend a BA institution. As

shown inTable 3, Model 2, the effect of economic shocks on students’ degree expectations varied by SES, as

indicated in Panel A. In Panel A, the slope for middle SES is signiﬁcant at p<.05, while the slope for low SES

is not statistically signiﬁcant. As indicated in Panel B (and shown inTable 3, Model 4), the effect of negative

economic shocks on parents’ route expectations did not vary by SES and hence is not modeled as such.

366 Renzulli and Barr considered. Hence, these models serve simply to offer nuance to our findings. With this goal in mind,

we found that a loss of income and a foreclosure were more detrimental to the degree expectations

of middle-class parents than to those of lower SES parents. These differential effects by SES are

shown inFigure 3. For more middle-class parents who had expected their child to earn a BA,

experiencing a home foreclosure or a loss of income is associated with a lower probability of still ex-

pecting a BA degree. This was not the case for lower-SES parents. For students, a negative change in

the family’s housing situation had the most disparate effects by SES. As shown inFigure 4, more

middle-class students who experienced a housing decline (i.e. moved from owning to renting, or rent-

ing to living with relatives) were at an increased risk of dropping their once-high expectations. This

was not the case for lower-SES students.

CONCLUSION

The Great Recession and its lingering effects have not only caused concern for unemployment (Reid

et al. 2013), welfare (Salgado et al. 2014), and retirement crises (Szinovacz et al. 2014), but have also

provided a challenge to the notion that educational expectations are not very malleable. Given a

heavy reliance on parents and families to orient young people toward education and careers across

the transition to adulthood (Furstenberg Jr 2010), sudden changes in family circumstances are central

to understanding the potential plasticity of adolescents’ educational expectations.

In our work we examined how negative economic shocks in a family can, in fact, alter expectations,

particularly expected routes to degree attainment for both students and parents. This work allowed

us to examine preparation for one of the most important transitions for high-schoolers right after a

time of national financial crisis. We should note, however, that we were capturing the effect of nega-

tive events that altered expectations and routes. We recognize that positive events (e.g. a sudden

Figure 3.Parents’ Degree Expectations for 11th Graders by SES and Type of Economic Shock

Notes:Estimates based on model with full controls, including each type of family economic shock. Predicted

probabilities assume that the parent has afﬁrmative expectations in 9th grade and so can be interpreted as the

probability that the parentstillexpects student to earn a BA.

* Effects signiﬁcantly different by SES, p<.05.Adapting to Family Setbacks

367 increase in income) in a family could also positively alter expectations and routes. An important ave-

nue for future work on upward mobility (through positive economic shocks) would be a useful ave-

nue to see how expectations change. These data, however, may not be ideal for examining these

positive “shocks.” Nonetheless, positive shocks are important for future research, and especially so in

the context of economic recovery.

Family economic shocks were inconsistently predictive of decreased degree expectations. There

was only weak evidence that high schoolers lowered their degree expectations in the face of family

economic crises, while there was strong evidence that middle-class parents lowered their degree ex-

pectations upon experiencing economic shocks. It is important to note, however, that the majority of

students and parents who had high degree expectations in 9

th grade, even those who experienced

multiple economic shocks, held on to these high expectations. Hence, although degree expectations

were indeed somewhat malleable, particularly those of more middle-class parents, the strength of the

college-for-all ethos appears to be significant here as well. This was not the case for route expecta-

tions. Family economic shocks served to reduce students’ and parents’ expectations for beginning

their post-secondary education at a BA-granting institution. This effect was amplified for more

middle-class students.

These findings highlight the utility of examining the dynamic nature of the status attainment pro-

cess in the context of a changing economic and higher education landscape and in the context of a

changing transition to adulthood. Social learning theories and Bayesian models will be useful to un-

derstand more fully how students alter their expectations (Breen 1999,Breen and Jonsson 2005,

Morgan 1998). In addition, our findings also suggest that the extensive focus on degree expectations

in the literature thus far may be limited in its conceptualization of educational expectations. Although

a greater understanding of the meaning and malleability of degree expectations is an important under-

taking in the contemporary era of college-for-all, we must also begin to explore the meaning,

Figure 4.11th Graders’ Institutional Route Expectations by SES and Type of Economic Shock

Notes:Estimates based on model with full controls, including each type of family economic shock. Predicted

probabilities assume that the student has afﬁrmative expectations in 9th grade and so can be interpreted as the

probability that the studentstillexpects to attend a BA institution.

* Effects signiﬁcantly different by SES, p<.05.

368 Renzulli and Barr malleability, and implications of route expectations. Accumulating evidence suggests thathowstu-

dents expect to meet their degree expectations may be becoming an increasingly important fault line

of inequality.

The implications of these inequalities are beyond the scope of this work, yet we know that to-

day’s much-extended transition to adulthood serves as a “critical juncture” (Schulenberg et al. 2004)

in the life course, with the potential to “cast a long shadow over adult lives” (Shanahan 2000: 686).

Hence, we must attend to inequalities that emerge or deepen during the preparation for the transi-

tion to adulthood. Further, the relative lack of public support for students during this transition may

mean that family context, with respect to both changing family economic conditions and parents’ ex-

pectations for their children, may be increasingly important for understanding students’ educational

pursuits during the transition to adulthood and, hence, the unfolding of the remainder of the life

course.

Going to college and earning a BA is also important for society writ large. Students acquire values

consistent with their imminent standing in the social structure. The effect of economic shocks likely

means that students who traditionally did not start at two-year colleges might now begin their higher

education there. There is the possibility, then, that we will see higher transfer rates from these stu-

dents because they would have been more likely to start at four-year schools prior to their economic

shocks. Though beyond the scope of this paper, perhaps having greater rates of students starting at

two-year colleges will help bolster the completion and transfer rates for them and other students as

well.

This work’s advantage in better understanding expectations as a result of economic shocks is also

its limitation. In the future, we must better understand how family economic contexts and expecta-

tions together encourage or discourage behaviors – behaviors that actually help students reach college

classrooms and earn degrees. As we watch these young people think about and form expectations, we

must also watch their behaviors to understand the effects of economic shocks in the long-term. Do

they apply to higher education institutions, get financial aid, attend schools, move between them, and

earn degrees?

The route to a BA through a two-year college has been tenuous at best (Dougherty 1991,1992,

2002), despite state articulation of policies to encourage this transition (Dougherty 1994).

Nevertheless, this new generation of middle-class students who expect a more piecemeal path to ob-

taining a BA highlights the importance of learning more about the activities at two-year schools, their

course offerings, and the counseling opportunities. The structure of higher education is in flux and

the routes by which students earn their degrees may be changing. Given the move to community col-

leges as the path toward a BA, more research is warranted on this new population of recent high

schoolers (Wang 2013) and the challenge they may present for community colleges (Goldrick-Rab

2010).

REFERENCES

Adelman, Clifford. 1999. “Answers in the tool box. Academic intensity, attendance patterns, and Bachelor’s degree at-

tainment.” U.S. Department of Education.

Alexander, Karl L and Martha A Cook. 1979. “The motivational relevance of educational plans: Questioning the con-

ventional wisdom.”Social Psychology Quarterly:202-13.

Alexander, Karl L, Doris R Entwisle and Samuel D Bedinger. 1994. “When expectations work: Race and socioeconomic

differences in school performance.”Social Psychology Quarterly:283-99.

Alexander, Karl L, Robert Bozick and Doris Entwisle. 2008. “Warming up, cooling out, or holding steady? Persistence

and change in educational expectations after high school.”Sociology of Education81(4):371-96.

Allison, Paul D. 2012. “Handling missing data by maximum likelihood.” Pp. 1-21 inSAS global forum.

Andrew, Megan and Robert M Hauser. 2011. “Adoption? Adaptation? Evaluating the formation of educational expecta-

tions.”Social Forces90(2):497-520.

Attewell, Paul and Thurston Domina. 2008. “Raising the bar: Curricular intensity and academic performance.”

Educational Evaluation and Policy Analysis30(1):51-71.

Adapting to Family Setbacks 369 Blau, Peter and Otis Dudley Duncan. 1967.The American occupational structure. New York: John Wiley & Sons.

Bozick, Robert. 2009. “Job opportunities, economic resources, and the postsecondary destinations of American youth.”

Demography46(3):493-512.

Bozick, Robert, Karl Alexander, Doris Entwisle, Susan Dauber and Kerri Kerr. 2010. “Framing the future: Revisiting the

place of educational expectations in status attainment.”Social Forces88(5):2027-52.

Bradburn, Ellen M., David G. Hurst and Samuel Peng. 2001. “Community college transfer rates to 4-year institutions

using alternative deﬁnitions of transfer.” Vol.Research and Development Report.National Center for Education

Statistics.

Brand, Jennie E and Yu Xie. 2010. “Who beneﬁts most from college? Evidence for negative selection in heterogeneous

economic returns to higher education.”American Sociological Review75(2):273-302.

Breen, Richard. 1999. “Beliefs, rational choice and bayesian learning.”Rationality and Society11(4):463-79.

Breen, Richard and Jan O Jonsson. 2005. “Inequality of opportunity in comparative perspective: Recent research on ed-

ucational attainment and social mobility.”Annual Review of Sociology31:223-43.

Briley, Daniel A, K. Paige Harden and Elliot M Tucker-Drob. 2014. “Child characteristics and parental educational ex-

pectations: Evidence for transmission with transaction.”Developmental Psychology50(12):2614-32.

Calarco, Jessica McCrory. 2014. “Coached for the classroom parents’ cultural transmission and children’s reproduction

of educational inequalities.”American Sociological Review79(5):1015-37.

Charles, Camille Z, Vincent J. Roscigno and Kimberley C. Torres. 2007. “Racial inequality and college attendance: The

mediating role of parental investment.”Social Science Research36:329-52.

Clark, Burton R. 1960. “The “cooling-out” function in higher education.”American Journal of Sociology65(6):569-76.

Deil-Amen, Regina and Ruth Lopez Turley. 2007. “A review of the transition to college literature in sociology.”The

Teachers College Record109(10):2324-66.

Deterding, Nicole M. 2015. “Instrumental and expressive education college planning in the face of poverty.”Sociology of

Education88(4):284-301.

Domina, Thurston, AnneMarie Conley and George Farkas. 2011. “The link between educational expectations and ef-

fort in the college-for-all era.”Sociology of Education84(2):93-112.

Dougherty, Kevin J. 1991. “The community college at the crossroads: The need for structural reform.”Harvard

Educational Review61(3):311-37.

——. 1992. “Community colleges and baccalaureate attainment.”The Journal of Higher Education:188-214.

——. 1994.The contradictory college: The conﬂicting origins, impacts, and futures of the community college: SUNY Press.

——. 2002. “The evolving role of the community college: Policy issues and research questions.” Pp. 295-348 inHigher

education: Handbook of theory and research: Springer.

Dougherty, Kevin J and Gregory Kienzl. 2006. “It’s not enough to get through the open door: Inequalities by social

background in transfer from community colleges to four-year colleges.”Teachers College Record108(3):452-87.

Elliott, William and Sondra Beverly. 2011. “Staying on course: The effects of savings and assets on the college progress

of young adults.”American Journal of Education117(3):343-74.

Flint, Thomas. 1992. “Parental and planning inﬂuences on the formation of student college choice sets.”Research in

Higher Education33(6):689-708.

——. 1997. “Intergenerational effects of paying for college.”Research in Higher Education38(3):313-44.

Frye, Margaret. 2012. “Bright futures in malawi’s new dawn: Educational aspirations as assertions of identity.”American

Journal of Sociology117(6):1565.

Furstenberg, Jr, Frank F. 2010. “On a new schedule: Transitions to adulthood and family change.”The Future of

Children20(1):67-87.

Goldrick-Rab, Sara. 2010. “Challenges and opportunities for improving community college student success.”Review of

Educational Research80(3):437-69.

Grodsky, Eric and Catherine Riegle-Crumb. 2010. “Those who choose and those who don’t: Social background and

college orientation.”The ANNALS of the American Academy of Political and Social Science627(1):14-35.

Grusky, David B, Bruce Western and Christopher Christopher Wimer. 2011.The great recession: Russell Sage

Foundation.

Haller, A. O. and A. Portes. 1973. “Status attainment processes.”Sociology of Education46(1):51-91.

Hamilton, Laura T. 2013. “More is more or more is less? Parental ﬁnancial investments during college.”American

Sociological Review78(1):70-95.

Hoachlander, Gary, Anna C Sikora, Laura Horn and CD Carroll. 2003. “Community college students.”Education

Statistics Quarterly5(2):121-28.

Johnson, M. K. and J. R. Reynolds. 2013. “Educational expectation trajectories and attainment in the transition to adult-

hood.”Social Science Research42(3):818-35. 370

Renzulli and Barr Karlson, Kristian Bernt. 2015. “Expectations on track? High school tracking and adolescent educational expectations.”

Social Forces.

Kerckhoff, Alan C. 1976. “The status attainment process: Socialization or allocation?”.Social Forces55(2):368-81.

——. 1977. “The realism of educational ambitions in england and the United States.”American Sociological

Review:563-71.

Kojaku, Lawrence K and Anne-Marie Nunez. 1998.Descriptive summary of 1995-96 beginning postsecondary students,

with proﬁles of students entering 2-and 4-year institutions. National postsecondary student aid study: 1995-96. Statistical

analysis report: ERIC.

Lareau, Annette. 1987. “Social class differences in family-school relationships: The importance of cultural capital.”

Sociology of Education60(2):73-85.

——. 2002. “Invisible inequality: Social class and childrearing in black families and white families.”American

Sociological Review67(5):747-76.

Lareau, Annette and Kimberly Goyette. 2014.Choosing homes, choosing schools: Russell Sage Foundation.

Lucas, Samuel R. 2001. “Effectively maintained inequality: Education transitions, track mobility, and social background

effects.”American Journal of Sociology106(6):1642-90.

——. 1996. “Selective attrition in a newly hostile regime: The case of 1980 sophomores.”Social Forces75:511-33.

Merton, Robert K. 1988. “The matthew effect in science: Cumulative advantage and the symbolism of intellectual prop-

erty”.Isis79(299):606-23.

Moller, Stephanie, Elizabeth Stearns, Stephanie R Potochnick and Stephanie Southworth. 2011. “Student achievement

and college selectivity: How changes in achievement during high school affect the selectivity of college attended.”

Youth & Society43(2):656-80.

Morgan, Stephen L. 1998. “Adolescent educational expectations rationalized, fantasized, or both?”.Rationality and

Society10(2):131-62.

——. 2005.On the edge of commitment: Educational attainment and race in the United States: Stanford University Press.

National Center for Education Statistics. 2013. “High school longitudinal study of 2009 (hsls:09) base year to ﬁrst

follow-up data ﬁle documentation.” Vol. Washington, DC: U.S. Department of Education.

Nielsen, Kelly. 2015. “‘Fake it’til you make it’’why community college students’ aspirations ‘‘hold steady’.”Sociology of

Education88(4):265-83.

Olson, Lorayn and Rachel A Rosenfeld. 1984. “Parents and the process of gaining access to student ﬁnacial-aid.”

Journal of Higher Education55(4):455-80.

Osgood, D Wayne, Amy L Anderson and Jennifer N Shaffer. 2005. “Unstructured leisure in the after-school hours.”

Organized activities as contexts of development: Extracurricular activities, after-school and community programs:45-64.

Ovink, Sarah M. 2014. ““They always call me an investment” gendered familism and Latino/a college pathways.”

Gender & Society28(2):265-88.

Paulsen, Michael B and Edward P St John. 2002. “Social class and college costs: Examining the ﬁnancial nexus between

college choice and persistence.”The Journal of Higher Education73(2):189-236.

Reid, Neil, Michael C. Carroll and Xinyue Ye. 2013. “The great recession of 2007-2009.”Economic Development

Quarterly27(2):87-89.

Reynolds, John R, Michael Stewart, Ryan Macdonald and Lacey Sischo. 2006. “Have adolescents become too ambi-

tious? High school seniors’ educational and occupational plans, 1976 to 2000.”Social Problems53(2):186-206.

Rosenbaum, James E. 2011. “The complexities of college for all: Beyond fairy-tale dreams.”Sociology of Education

84(2):113-17.

Salgado, Mario F., Francesco Figari, Holly Sutherland and Alberto Tumino. 2014. “Welfare compensation for unem-

ployment in the great recession.”Review of Income and Wealth60:S177-S204.

Schneider, Barbara L. and David Stevenson. 1999.The ambitious generation : America’s teenagers, motivated but direction-

less, Edited by D. Stevenson. Yale University Press: New Haven.

Schulenberg, John E, Arnold J Sameroff and Dante Cicchetti. 2004. “The transition to adulthood as a critical juncture

in the course of psychopathology and mental health.”Development and psychopathology16(04):799-806.

Settersten, Jr, Richard A and Barbara Ray. 2010. “What’s going on with young people today? The long and twisting

path to adulthood.”The Future of Children20(1):19-41.

Sewell, Willam H, Acrchibald O Haller and Alejandro Portes. 1969. “Education and early occupational attainment pro-

cess.”American Sociological Review34(1):82-92.

Shanahan, Michael J. 2000. “Pathways to adulthood in changing societies: Variability and mechanisms in life course per-

spective.”Annual Review of Sociology:667-92.

Spenner, Kenneth I and David L Featherman. 1978. “Achievement ambitions.”Annual Review of Sociology:

373-420.

Adapting to Family Setbacks 371 Steinberg, Laurence, Sandra Graham, Lia O’Brien, Jennifer Woolard, Elizabeth Cauffman and Marie Banich. 2009.

“Age differences in future orientation and delay discounting.”Child Development80(1):28-44.

Szinovacz, Maximiliane E., Lauren Martin and Adam Davey. 2014. “Recession and expected retirement age: Another

look at the evidence.”Gerontologist54(2):245-57.

Turley, Ruth N Lopez, Martin Santos and Cecilia Ceja. 2007. “Social origin and college opportunity expectations across

cohorts.”Social Science Research36(3):1200-18.

Vuolo, Mike, Jeylan T. Mortimer and Jeremy Staff. 2014. “Adolescent precursors of pathways from school to work.”

Journal of Research on Adolescence24(1):145-62.

Wang, Xueli. 2012. “Stability of educational expectations among baccalaureate aspirants beginning at community colle-

ges.”Community College Review40(4):300-19.

——. 2013. “Baccalaureate expectations of community college students: Socio-demographic, motivational, and contex-

tual inﬂuences.”Teachers College Record115(4).

Warren, J. Robert, Robert M Hauser and Jennifer T Sheridan. 2002. “Occupational stratiﬁcation across the life course:

Evidence from the wisconsin longitudinal study.”American Sociological Review67(3):432-55.

Wilkins, A. C. 2014. “Race, age, and identity transformations in the transition from high school to college for black and

ﬁrst-generation white men.”Sociology of Education87(3):171-87.

Zhan, Min and Michael Sherraden. 2011. “Assets and liabilities, educational expectations, and children’s college degree

attainment.”Children and Youth Services Review33(6):846-54. 372

Renzulli and Barr