American Sociological Review 2013 Hamilton 70 95

American http://asr.sagepub.
com/
Sociological Review
More Is More or More Is Less? Parental Financial Investments during College

Laura T. Hamilton
American Sociological Review 2013 78: 70 originally published online 3 January 2013
DOI: 10.1177/0003122412472680
The online version of this article can be found at:
http://asr.sagepub.com/content/78/1/70
Published by:
http://www.sagepublications.com
On behalf of:
American Sociological Association
Additional services and information for American Sociological Review can be found at:
Email Alerts: http://asr.sagepub.com/cgi/alerts
Subscriptions: http://asr.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Jan 30, 2013

OnlineFirst Version of Record - Jan 3, 2013
What is This?
Downloaded from asr.sagepub.com by guest on September 1, 2014
472680
80American Sociological ReviewHamilton
2012
ASRXXX10.1177/00031224124726
More Is More or More Is Less?

Parental Financial Investments
during College
American Sociological Review

78(1) 7095
American Sociological
Association 2012
DOI: 10.1177/0003122412472680
http://asr.sagepub.com
Laura T. Hamiltona
Abstract
Evidence shows that parental financial investments increase college attendance, but we
know little about how these investments shape postsecondary achievement. Two theoretical
frameworks suggest diametric conclusions. Some studies operate from a more-is-more
perspective in which children use calculated parental allocations to make academic progress.
In contrast, a more-is-less perspective, rooted in a different model of rational behavior,
suggests that parental investments create a disincentive for student achievement. I adjudicate
between these frameworks, using data from nationally representative postsecondary datasets
to determine what effect financial parental investments have on student GPA and degree
completion. The findings suggest seemingly contradictory processes. Parental aid decreases
student GPA, but it increases the odds of graduatingnet of explanatory variables and
accounting for alternative funding. Rather than strategically using resources in accordance
with parental goals, or maximizing on their ability to avoid academic work, students are
satisficing: they meet the criteria for adequacy on multiple fronts, rather than optimizing their
chances for a particular outcome. As a result, students with parental funding often perform
well enough to stay in school but dial down their academic efforts. I conclude by highlighting
the importance of life stage and institutional context for parental investment.
Keywords
college completion, grade point average, higher education, parental investment, satisficing,
young adulthood
Higher education in the United States has

long been sponsored by parents funds. Over
time, however, the burden on parents has
grown heavier and more substantial. During
the past three decades, the federal government has increasingly transferred a greater
proportion of aid from grants to loans that are
often carried in part by parents (Baum and
Steele 2007; Paulsen and St. John 2002). At
the same time, the proportion of revenue
coming from state and local funding has
gradually declined (Fain 2009; McPherson
and Schapiro 1998).1 At most schools, tuition
has risen to make up the difference. In the
past decade alone, average tuition and fees at

private four-year colleges and universities
increased at an annual rate of 3 percent in
inflation-adjusted dollars; public schools saw
an even steeper annual increase of 5.6 percent
(Baum and Ma 2010).
University of California-Merced
Corresponding Author:
Laura T. Hamilton, School of Social Sciences,
Humanities and Arts, 5200 North Lake Road,
Merced, CA 95343
E-mail: [email protected]
Hamilton
71
These trends reflect a gradual shift in the

responsibility for U.S. higher education fundingfrom state and federal subsidies to individual families (Lucas 1996; Price 2004). As
public schools begin to operate more like
private schools and rely heavily on tuition
dollars, and federal funds for higher education fail to absorb the cost to students, the role
of parents becomes even more central. Parents often become the primary financiers of
higher education.
To send their children to college, parents
often make difficult financial decisions.
Financing childrens education comes at a
considerable cost for many, as they dip into
family savings or pull from retirement
accounts. Yet, little work examines whether
parental dollars translate into quantifiable
benefits for students. Simply put, can parents
purchase a better college outcome for their
children? Do parental dollars boost student
performance? Or is there a point of diminishing returns, even a negative influence? These
questions are important not only for individual parents as they make decisions about how
much to fund their children, but also for policymakers in understanding implications of a
funding structure that rests squarely on parents shoulders.
The central goal of this article is to determine what effect financial parental investments
have on two key postsecondary education outcomes: grade point average (GPA) and degree
completion. I rely on data from three nationally
representative datasets of postsecondary students collected by the National Center for Education Statistics (NCES) to address the
following empirical questions:
1) What is the direction and magnitude of parental aids influence on student GPA and
bachelors degree completion, net of student
sociodemographics, parental socioeconomic
status (SES), family structure, academic
ability, student characteristics, and institution characteristics?
2) What are the effects of parental aid when accounting for other sources of financial aid,
in both relative and absolute terms?
Provision of Parental Aid

Deil-Amen and Turley (2007) note that relatively little research within the sociology of
higher education focuses on college financing.
One exception is examination of factors shaping how much parental aid parents provide.
Higher socioeconomic status parents are more
likely to assume their children will attend college, have the resources to pay for it, and
make the necessary financial plans (Flint
1992; Olson and Rosenfeld 1984). Family
structure also plays a role. Married parents
contribute larger amounts of money to their
childrens college education (Turley and
Desmond 2011). When there are greater numbers of siblings, especially if they are closely
spaced, each child may receive less (Downey
1995; Steelman and Powell 1989). Parents
may invest more in children with higher academic aspirations and in children who demonstrate higher levels of achievement, although
evidence for the latter is mixed (Flint 1997;
Powell and Steelman 1995).
The cost of education also matters. Pricier
schools often require greater parental contributions, and students self-select based on
ability to pay (Paulsen and St. John 2002).
Institutions typically provide scholarships
based on prior achievement, and students can
also receive funds from programs such as
National Merit Scholarships. This type of aid
tends to disproportionately benefit students
from affluent families who arrive with
stronger records (Carnevale and Strohl 2010).
Need-based aid is usually calculated using the
expected family contribution (EFC) and does
not take into account the actual amount of
parental assistance students will receive.2 The
net cost to families includes the EFC plus any
additional costs not met by merit, need, or
other forms of aid (e.g., military or employer
benefits). Students and their parents must
decide how to cover these costsusing different types of loans, savings, or earnings.
Even if we match students on all the above
characteristics, we might expect considerable
heterogeneity in how much aid parents offer
their children. Indeed, attendance at a four-year
72
American Sociological Review 78(1)
college epitomizes a relatively new life

stage, young adulthooda period of the life
course devoted to academic, career, and selfdevelopment (Rosenfeld 2007; Settersten and
Ray 2010). In many ways, young adulthood
mirrors the development of adolescence around
the turn of the twentieth century, in that it is not
yet fully institutionalized or universal (Furstenberg, Rumbaut, and Settersten 2005; Osgood
et al. 2005; Zelizer 1985). Compared to parenting younger children, there is currently less
consensus about what amount of financial support constitutes good parenting for college studentseven within social class categories.
Past research on intergenerational effects
of parental funding for college underscores
this point. Accounting for income and education, parents who received financial support
for college from their own parents are significantly more likely to provide higher levels of
aid for their offspring (Flint 1997). Here we
see evidence that a cultural understanding
about young adulthoodtransmitted across
generationsshapes parental investment in
college. College funding thus cannot be
understood as entirely a matter of financial
calculus. Not all parents who can afford to do
so will cover the costs of college, and some
with limited resources will find ways to provide more than expected.
Effects of Parental Aid

Sociological knowledge of the effects of
parental support for older youth is limited,
although recent work notes links to college
attendance. For example, Charles, Roscigno,
and Torres (2007) demonstrate that parental
investmentsincluding financial support for
collegecan explain the BlackWhite gap in
college attendance. Kim and Schneider (2005)
show that parental social capital helps parents
channel outside information and financial
resources to increase the likelihood that students successfully transition into higher education. These findings are consistent with a
growing awareness that extending parental
investment later into the life course provides
youth with distinct advantages (Osgood et al.
2005; Settersten and Ray 2010). However, the

extent to which parental investments, specifically parental aid, shape specific academic
outcomes for college students remains largely
unexplored.
As Bowen and Shapiro (1998) note, this
issue plagues work on college funding more
generally. Most scholarship has focused on
the link between aid and access and has not
adequately addressed the effects of aid on college achievement and completion. Some
recent research has begun to move beyond
access. For example, scholars have shown
that in some cases merit-based aid can boost
achievement (Henry and Rubenstein 2002;
Stater 2009), and other researchers have
begun to trace the receipt of grants and scholarships to degree completion (DesJardins,
Ahlburg, and McCall 2002; Dynarski 2003).
Similarly, Paulsen and St. John (2002) indicate that unmet need is a barrier to persistence. Parental aid, however, has been largely
excluded from these efforts (but see Steelman
and Powell 1989).
Inattention to effects of parental aid during
college is surprising given the importance of
postsecondary achievement for students life
chances. For example, GPA can fundamentally shape students movement into advanced
degree programs and boost earning power
(Jones and Jackson 1990; Loury and Garman
1995; Mullen, Goyette, and Soares 2003).
Degree completion is associated with greater
access to higher paying and more privileged
occupations, a higher probability of marriage,
better health, greater civic participation, and
intellectual development (for a review, see
Stevens, Armstrong, and Arum 2008). Understanding what role parental investments play
in influencing these key educational outcomes is thus central to the goals of stratification research.
Two frameworks offer divergent views on
how parental investments shape student GPA
and degree completion. Studies from status
attainment, human capital, and cultural capital
traditions operate from a more-is-more perspective. This lens suggests that children use
parents calculated allocations in service of
Hamilton
73
higher education. According to this approach,

increasing parental investments improve, or at
the very least do not harm, student performance. In contrast, I examine a more-is-less
perspective that is rooted in a different rational
choice model of human behavior and is often
implied in recent concerns over young adults
financial dependence on their parents. The
theory of moral hazard suggests that parental
investments may create a disincentive for student achievement.
Traditional Approaches to Parental

Investment: More Is More
The first framework, more-is-more, draws
from sociological traditions that identify
parental investment in education as a key
mechanism driving the reproduction of advantage. For example, status attainment models
which dominate stratification research in
sociologypostulate that an individuals
socioeconomic destination is a function of
ability, background, aspirations, and parental
investments in education. Research has examined many forms of parental investments,
including social psychological aspects (e.g.,
educational aspirations), financial capital, cultural capital, and social capital (Blau and
Duncan 1967; Coleman 1988; Downey 1995;
Sewell and Hauser 1976). Central to this particular investigation, Steelman and Powell
(1989) identify financial investments in postsecondary education as a previously little recognized factor in status attainment processes.
These models often imply that parental
investments link class origins and destinations via a series of calculated exchanges,
characteristic of a rational choice perspective.
For example, in the human capital model
(Becker 1964) parents intentionally direct
financial resources to their childrens education, expecting children to optimize these
funds by accruing skills and credentials necessary for future socioeconomic success.
Scholarship from a cultural perspectiveperhaps best illustrated by Bourdieu (1984; see
also Lareau 2011)suggests that parents cultivate valuable social tastes, interests, skills,
and dispositions through exposure to certain

educational and extracurricular contexts. This
process can occur through calculated intent or
a less direct absorption of ones social and
cultural environs.
This research shares a basic tenet: parental
investments in education have a positive
effect on childrens academic, occupational,
and economic fates. There is little room in
status attainment, human capital, or cultural
approaches for ineffective or even problematic parental educational investments. These
core sociological traditions implicitly (and
sometimes explicitly) suggest that more is
better than less, and that parents cannot invest
too much in their children.
Young Adulthood and Moral Hazard:

More Is Less
A more-is-more perspective is deeply
ingrained in the sociology of education. In
recent years, however, concern about parental
investment has grown as the responsibilities
ascribed to parents continue to expand and
extend further into the life course (Acocella
2008). Intentionally crafting opportunities for
childrens intellectual, social, and emotional
growth has become an accepted norm of
middle-class parentingand one that may
generate its own set of problems (Hays 1996;
Lareau 2011; Warner 2006). For youth reaching the end of adolescence, this parenting
may be primarily financial in nature.
Scholars have begun to note the possible
high cost of creating a young-adult life stage
where parents free youth from the realities of
financial responsibility (Danziger and Rouse
2007). Some evidence indicates that allowing
youth to postpone adult statuses, like fulltime employment, may have unintended
effects on their independence. For example,
Newman and Aptekar (2007) document an
increasing delay in leaving the natal home in
Western Europea pattern also seen in the
United States. Similarly, Lareau (2011)
describes early educational benefits of an
intensive logic of childrearing, but her work
also hints at issues of entitlement among
74
young adults who are used to having their

needs met by parents. These emerging observations raise questions about how far into the
life course parental investments continue to
be beneficial.
Such concerns tap into an alternative theoretical framework for understanding parental
financial aid in college, where more may actually be less. Moral hazard theory, like prior
sociological research on parental investment, is
derived from a rational choice perspective on
human interactions (Heimer 1988). The assumptions, however, are much different. Moral hazard
arises in situations of information asymmetry:
individuals insulated from risk behave in ways
not compatible with their investors goals. This
may occur because they are not fully accountable for potential consequences and may pursue
different interests.3 When applied to this case,
moral hazard theory suggests that parental aid
can provide an educational disincentive for children. Children may direct more effort to school
when they personally feel the economic costs of
poor performance.
In earlier stages of education, children
spend the majority of time under the watchful
gaze of teachers or parents. During college,
particularly when students live on-campus and
away from home, parents are less able to
monitor their childrens academic behaviors to
ensure sanctioned use of educational resources.
Parental aid may also have unique characteristics distinguishing it from most other forms of
aid. For example, grants and scholarships are
often merit-based and come with strict performance standards that may keep students
focused. Similarly, work-study and veteran
benefits require students to have personal
responsibility in obtaining funds, which may
translate into greater investment in academic
performance. Loans are the only other source
of aid that is generally not tied to performance
and may divert financial responsibility from
students during collegeespecially if students are not responsible for payments.
Some studies suggest this more-is-less
framework may be promising. For example,
Davila and Mora (2004) invoke moral hazard
to explain why children of self-employed
entrepreneurial parents underperform in relation to their peers. They posit that these students are less motivated because of anticipated
economic security through the family business. Economists Bodvarsson and Walker
(2004) found that among a small sample of
students at two Midwestern schools, parental
aid for tuition and books significantly weakened self-reported academic achievement.
This study suggests the need for more rigorous testing. Finally, a large (although conflicted) literature on effects of student
employment on performance indicates the
importance of personal financial responsibility. Several studies have found no effects
(Curtis and Nummer 1991) or positive effects
of low to moderate levels of student employment on GPA (Kalenkoski and Pabilonia
2008; Pike, Kuh, and Massa-McKinley 2008).
This scholarship suggests that young adulthood, sponsored by parental funds, may provide a context in which more is less. Here,
parental funds set the stage for moral hazard by
offsetting costs associated with low academic
performance. From this perspective, parental
aid will not boost GPA or the likelihood of
graduating and may even prove detrimental.
Reverse Causality and Selectivity

Processes
These two perspectives point to parental aid
as a causal mechanism in college students
performance. However, the casual arrow may
be reversed. For example, more may look like
more if, as noted earlier, parents invest more
in better performing students (Powell and
Steelman 1995). In contrast, research from
earlier levels of schooling suggests parents
may invest more in students who are disadvantaged in some way, as a compensatory
mechanism (Hamilton, Cheng, and Powell
2007; Teachman, Paasch, and Carver 1997).
Selectivity processes may also create a situation where less motivated or less talented students are more likely to receive parental
investments. Because parental aid increases
access to college, students with parental assistance will likely display a wider range of ability
Hamilton
75
and motivation. In contrast, students who make

it to college with little to no parental help may
not only be exceptionally talented but also
uniquely motivatedfor which there is no
good empirical proxy (Gambetta 1987; Torche
2011). Unobserved heterogeneity among the
college student population may thus drive a
similar empirical finding as processes of moral
hazard, but for different reasons. More may
look like less if students receiving greater
amounts of support are qualitatively different
from those who do not receive similar help.
When relevant, I address these possibilities
with selectivity analyses using fixed-effects
models.
DATA, MEASURES, AND

METHODS
Primary analyses rely on two nationally representative postsecondary datasets collected
by the National Center for Education Statistics
(NCES): the Baccalaureate and Beyond Study
1993 (B&B93) and the Beginning
Postsecondary Students Study 1990 to 1994
(BPS90/94). These datasets are linked to the
National Postsecondary Student Aid Study
(NPSAS), which includes unparalleled data
on elements of college students funding situations, institution and student characteristics,
and academic records. Unlike more recent
NPSAS-linked data, these datasets offer continuous, rather than dichotomous or categorical, measures of parental aid.
Where GPA is the outcome of interest, I
focus on the B&B93. This dataset includes a
nationally representative sample of around
12,000 undergraduate students in their final
year at their U.S. bachelors granting institution during 1992 to 1993. The B&B93 is
preferred for its large and representative sample of students in four-year schools, but a
sample of students in their final year is not
appropriate for analyses in which bachelors
degree completion is the key outcome. For
these analyses, I turn to the BPS90/94, which
follows a nationally representative sample of
first-year students attending postsecondary
institutions of any type in the United States
during 1989 to 1990. This dataset includes

around 4,000 students who started at fouryear schools and were not lost before the final
follow-up.
For both datasets, I used multiple imputation techniques appropriate for data missing
completely at random (MCAR) or missing at
random (MAR). I generated m = 10 complete
datasets with multiple imputation by chained
equations, which uses the observed data to
simulate plausible missing values (Royston
2005). I analyzed each of the m datasets individually and combined results to produce
estimates that account for missing data uncertainty (Little and Rubin 2002). As indicated
by von Hippel (2007), I used the dependent
variable in imputation but excluded cases
missing on the dependent variable from analysis (although including them does not alter
the findings presented here).4 As a result, the
sample size for the B&B93 analyses is 10,870
students; for the BPS90/94 it is 3,810 students.5 Table 1 reports the percent of imputed
values in the analytic sample. The BPS90/94
shows fewer imputed values overall, given
that cases not present in the final wave were
excluded prior to imputation.6
For the GPA analyseswhich produce a
surprising findingI also included analyses
from the Beginning Postsecondary Students
Study 1996 to 2001 (BPS96/01) to confirm
the B&B93 findings and address selectivity
processes. As with the BPS90/94, I restricted
the sample to students who started their first
year in four-year schools. For this dataset, I
only included students who were age 30 years
or younger because older respondents were
not asked about parental assistance.
The BPS96/01 provides limited information on parental aid and student GPA at three
points throughout college: 1996, 1998, and
2001 (2001 data include final GPAs for
respondents who graduated). This allows me to
estimate fixed-effects models, which make
comparisons within students rather than
between them. These models estimate parameters for variables that experience any amount
of temporal variance, even if only within a
subset of cases, and control for effects of
76
Table 1. Descriptive Statistics for B&B93 and BPS90/94 Analytic Samples

B&B93 (N = 10,870)
Variables
Key Independent Variables
Parental aid
Parental aid/total funding
Dependent Variables
Student GPA (x100)
BA by 1994
Sociodemographics
Female
Age
Race
Black
White (reference)
Asian
Other race
Hispanic
Parental SES
Income
Independent status
Education
HS degree or less (reference)
Some college
College degree
Advanced degree
Family Structure
Parents married
Number of members in college
Academic Ability
SAT/ACT score
1989 to 1990 GPA
Student Characteristics
Major
Business and management
Computer science
Education
Engineering (reference)
Health
Humanities
Life sciences
Math
Physical sciences
Social/behavioral sciences
Vocational/technical
Other major
Undeclared
Enrolled full-year
Full-time intensity
Out-of-state student
% Imputed
Mean
20.92
27.50
3.44
.27
n/ab
BPS90/94 (N = 3,810)
% Imputeda
3.57
16.46
306.43
Mean
4.77
.37
n/ab
.59
2.39
2.29
.57
24.99
.00
.00
.52
18.798
10.38
10.38
10.38
10.38
10.28
.06
.86
.04
.03
.06
.03
.03
.03
.03
.21
.08
.86
.05
.01
.05
1.92
.20
45.77
.40
.00
.00
48.75
.08
7.23
7.23
7.23
7.23
.31
.19
.29
.25
1.97
1.97
1.97
1.97
.27
.23
.25
.25
13.31
27.17
.76
1.28
.00
3.65
.74
1.40
19.83
987.41
36.72
17.14
980.69
262.02
.45
.45
.45
.45
.45
.45
.45
.45
.45
.45
.45
.45
.15
.03
.15
.07
.07
.11
.08
.02
.02
.17
.03
.11
.00
.20
2.25
12.10
.65
.66
.29
.20
.08
.09
.06
.14
.06
.01
.02
.16
.02
.07
.10
.87
.85
.30
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
2.97
1.29
(continued)
Hamilton
77
Table 1. (continued)
B&B93 (N = 10,870)
Variables
Employment during school
Full-time
Part-time
Not employed (reference)
Institution Characteristics
Private institution
Selectivity
Institution cost
Alternative Funding
Grants and scholarships
Loans
Work-study
Other forms of aid
Student employment
BPS90/94 (N = 3,810)
% Imputed
Mean
% Imputeda
Mean
10.13
10.13
10.13
.16
.54
.29
.00
.00
.00
.23
.55
.23
.20
2.38
7.97
.35
3.44
4.64
.00
5.20
.18
.57
3.28
5.91
2.25
2.25
2.25
2.25
19.32
1.51
1.32
.14
.08
5.71
.00
.00
.00
.00
13.94
2.20
.95
.14
.30
2.93
Note: Means are from imputation dataset m = 1. Monetary variables are in thousands of dollars.
Imputations were based on students not lost to follow-up in 1994.
b
Analytic sample does not include imputed values for dependent variables.
a
variables that are time invariant. Fixed-effects

models are thus appropriate for addressing
issues of unobserved heterogeneity, because
effects of relatively stable factors that typically
vary only between individuals (e.g., gender,
parental background, and prior performance),
as well as intelligence, academic preparation,
and motivation, are canceled out. The fixedeffects models presented here rely on an unbalanced panel of 7,210 students, observed one to
three times throughout college (for a total of
18,050 student-year observations). Only 130
students are lost to missing data.
Key Dependent and Independent

Measures
Variables are coded in the same way across
the B&B93 and BPS90/94 datasets, except
where noted. See Table 1 for descriptive statistics from imputation dataset m = 1.7 I discuss the variables included in the
supplemental BPS96/01 analyses later.
Student GPA. The first dependent variable
is student GPA (multiplied by 100).8 Studies
often use student-reported GPA, but the

tendency to overreport is higher among
some sets of students than others, potentially
leading to spurious or inaccurate findings
(Kuncel, Crede, and Thomas 2005). Thus I
use institution-reported data.9
Degree completion. The second dependent variable is a dichotomous measure of
whether students received a bachelors degree
five years after they started college.
Parental financial aid. The key independent variable is parental financial aid for
college. The continuous measure reports a
dollar amount for all direct financial transfers
that both parents made on behalf of their
children during the academic year in
questioncovering tuition and fees as well as
money for books, rent, living expenses, food,
clothing, and transportation.10
Table 1 shows the mean amount of parental aid received by students in both datasets;
as we might expect, it is higher for the sample
of first-year students. This number obscures
class differences in the provision of parental
78
aid. Indeed, we see a correlation between

income and parental aid (r = .396, p < .001 for
B&B93; r = .461, p < .001 for BPS90/94). As
noted earlier, however, there are theoretical
reasons to expect heterogeneity in the provision of student aid for students who look
quite similar on objective indicators, like
parental income. Comparison of students on
the basis of their propensity for receiving any
parental aid (calculated using the explanatory
variables below) suggests overlap among the
type of students who receive no parental aid,
some aid, and very high levels of aid. For
instance, in the B&B93 imputed data, students who received no aid have an average
propensity score of .40this indicates a fair
number of students did not receive aid despite
profiles that would suggest otherwise. Similarly, respondents who received more than
$30,000 have an average propensity score of
.88this suggests that some students who
accepted high levels of aid were not expected
to receive any at all.
Values for parental aidlike all monetary
variables in the analysesare positively
skewed. For example, in m = 1 of the B&B93,
42.72 percent of students received no parental
aid, 31.98 percent received less than $5,000,
12.17 percent received between $5,000 and
$10,000, 9.38 percent received between
$10,000 and $20,000, and 3.75 percent
received $20,000 or more (with a maximum
of $40,000). I thus take the log of all monetary variables.11 Analyses restricted to students receiving far smaller amounts (e.g., less
than $5,000) produce consistent results and
suggest that outliers are not driving the findings presented below.
Explanatory Variables
The analyses also include a number of additional explanatory variables shown to shape
student GPA and BA completion, as well as
the provision of parental aid.12
Basic sociodemographics. Research shows
that women receive more parental aid for
higher education (Powell and Steelman 1995).
Women have higher college GPAs than men,

on average, and enjoy an advantage in college
completion (Buchmann and DiPrete 2006;
Sax and Harper 2007). Older students, who
generally do not attend college right after high
school, also have higher GPAs (Burton
et al. 2009). At the same time, nontraditional
students may find it more difficult to persist,
because they are more likely to have heavy
work and family demands (Brown 2002). I
therefore include controls for gender and age.
I include racial categories (with White as
the reference) and Hispanic as a separate
dichotomous variable. At the multivariate
level, Asian students receive the highest levels of parental aid, and Blacks receive the
lowest (Powell and Steelman 1995). Whites
and Asians have the highest GPAs; Hispanic
and Black students tend to have lower GPAs
(Fischer 2007). Black students are at the highest risk for educational stopout, although this
is primarily due to other factors such as lower
family income (DesJardins, Ahlburg, and
McCall 2006).
Parental SES. A vast body of literature
links parental income and education to positive educational outcomes (Blau and Duncan
1967; Jencks et al. 1972; Sewell and Hauser
1976). This relationship is thought to exist in
large part due to parental investments. Higher
levels of parental income and education boost
parental investments that are assumed to have
a positive effect on student outcomes. Wealthier and better educated parents, however, may
provide other advantagessuch as high educational aspirations for their childrenthat
lead to these benefits. This set of controls is
thus essential for determining if financial
investments themselves boost student performance or run counter to effects of parental
background. I treat education as a series of
dummies, with high school degree or less as
the reference category.
The NPSAS collects parental income only
for legally dependent college students
regardless of how much parental aid children
receive. The majority of students in both datasets are legal dependants of their parents. For
Hamilton
79
legally independent students, I use the students own income and include a variable
indicating independent status in the analyses.
Because there is a correlation between independent status and the log of parental aid (r =
.470, p < .001 for B&B93; r = .260, p < .001
for BPS90/94), including it provides a conservative estimate of effects of parental aid.
Family structure. Turley and Desmond
(2011) demonstrate that parents who are married to each other provide the greatest financial
contributions. Research also indicates that students from divorced families do not perform
as well in school as do those from nondivorced
families (Amato 2001). I include a dummy
variable indicating marriedas opposed to
single, remarried, or divorcedparents.
Strain placed on family resources by other
family members also shapes provision of
parental aid (Downey 1995; Powell and Steelman 1995). Therefore, I account for the number of family members in college. This
measure, like income, is reported on the basis
of dependency.
Academic ability. Although imperfect,
SAT scores are perhaps the most commonly
used predictor of academic ability and I thus
include them here (Camara and Echternacht
2000). In cases where SAT scores were not
available, I used the SAT-converted ACT
score. Because research has identified college
GPA as a factor shaping persistence, degree
completion analyses also include first-year
GPA as an independent variable (Cabrera,
Nora, and Castaneda 1993). The B&B93 and
BPS90/94 do not include high school GPA,
but the supplemental fixed-effects analyses
using the BPS96/01 account for this timeinvariant factor.
Student characteristics. I include a
series of dummies for major, with engineering
(which has the lowest average GPA) as the
reference.13 I also account for enrollment
status (full-year or part-year), intensity (fulltime or part-time), and residency (in-state or
out-of-state). Finally, I include a categorical
measure of employment status during the academic year. I define full-time employment as
35 hours or more per week and part-time as
less than 35 hours. No employment is the reference category.
Institution characteristics. Because
student GPAs are higher on average at private
institutions, I include a measure of institution
control (public versus private) (Rojstaczer
2002). When accounting for student quality,
increasing selectivity depresses GPA but
increases the likelihood of graduation (Alexander and Eckland 1977; Alon and Tienda
2005). I therefore include institutions scores
on the 1992 Barrons Admissions Competitiveness Index. Barrons Index ranges from
noncompetitive (1) to most competitive (6)
and is based on four criteria: SAT/ACT scores
of students accepted in the previous year, GPA
required for admission, class rank required for
admission, and percentage of applicants
accepted the previous year. I also include a
measure of tuition cost, because it directly
shapes how much money parents provide.14
Alternative funding. In the final set of
analyses for each dependent variable, I
include measures of aid from alternative
sourcesgrants and scholarships, loans,
work-study, and other aid (e.g., veteran benefits).15 I pulled information on alternative aid
sources from federal and institutional files, as
part of the NPSAS. Because educational
funds are often derived from student employment, I also include a measure of money
earned starting June 30 of the year in question
and ending July 1 of the following year.
Analytic Approach
The article is organized in two main sections.
First, I examine the influence of parental
financial aid on student GPA. Then, I move to
its effects on degree completion. For each
outcome variable I proceed as follows. I estimate a bivariate model, in which parental aid
is used to predict the outcome of interest. I
then include student sociodemographics and
80
parental SESbecause these variables are

central to most education and stratification
researchand move to a full model adding
family structure, academic ability, student
characteristics, and institution characteristics.
Next, I account for alternative funding,
including grants and scholarships, loans,
work-study, other aid (e.g., veteran benefits),
and funds earned through student employment. I treat parental aid both as a relative
measurein which it is divided by the total
aid packageand as an absolute measure
(alongside the absolute amounts derived from
alternative sources). Finally, for the GPA
analyses, I present supplemental, fixed-effects
estimates as a way of cross-checking patterns
and addressing the possibility that unique
characteristics of college students without
support are driving the findings.
RESULTS
Student GPA
Table 2 presents coefficients for the regression of GPA on parental aid (logged) and
other explanatory variables. Model 1 indicates that at the bivariate level, an increase in
parental aid is associated with a decrease in
student GPA (b = 2.182, p < .001).16 This
effect is apparent even though parental aid
operates as a proxy for a host of different
social processes linked to parental background.
Model 2 helps to disentangle effects of
social class from parental aid with the addition of controls for sociodemographics and
parental SES. The model again indicates that
as parental aid increases, student GPA
decreases (b = 2.233, p < .001). At the same
time, parental income and education work at
cross-purposes with parental investments. As
income increases, so does student GPA, and
having a parent with an advanced degree (as
opposed to a high school degree or less) is
associated with a significant increase in GPA.
Model 2 also shows that women and older
students have significantly higher GPAs.
White students (the reference category) have
the highest GPAs, followed by Asians, students of other races, and Blacks. Students
who identify as Hispanic have significantly
lower GPAs than other respondents.
Model 3 adds variables capturing family
structure, academic ability, student characteristics, and institution characteristics. Notably,
the significant negative effect of parental aid
on student GPA persists and, if anything, is
magnified (b = 4.570, p < .001).17 Here, the
positive effect of parental education is no
longer apparent. In fact, having a parent with
some college or a college degree significantly
decreases student GPA. As a long tradition of
educational research suggests, the positive
effect of parental education on student performance operates primarily indirectly through a
number of mechanisms such as advantages on
standardized tests like the SAT and greater
likelihood of admission to more selective
institutions. Net of these benefits, greater
selectivity in who attends college among
first-generation students, as opposed to those
with college-educated parents, becomes visible (see Torche 2011).
Students with married parents have, on
average, higher GPAs. As SAT score increases,
so does student GPA. All majors have significantly higher GPAs than the reference category of engineering.18 Students enrolled
full-year and full-time have higher GPAs.
Out-of-state students, who are typically from
wealthier families or recruited for merit, also
achieve greater GPAs. Although working
part-time during the academic year has no
adverse effects on GPA, full-time employment does. Students at private institutions
have significantly higher GPAs; however, as
selectivity increases, student GPA decreases.19
These findings indicate that parental aids
negative effect is not due to family, student,
or institution characteristics. It is possible,
however, that parental aids effects may not
be the same for students from different social
class backgrounds. I thus estimate a model
with an interaction term that allows the effect
of parental aid to vary by income. The interactive variable is significant, indicating some
heterogeneity.20
Hamilton
81
Table 2. Regression Coefficients for Student GPA on Parental Aid (Log) and Explanatory
Variables, B&B93 (N = 10,870)
Parental Aid (log)

Sociodemographics
Female
Age
Race
Black
Asian
Other race
Hispanic
Parental SES
Income
Independent status
Education
Some college
College degree
Advanced degree
Family Structure
Parents married
Academic Ability
SAT/ACT score
Major
Computer science
Education
Health
Humanities
Life sciences
Math
Physical sciences
Other major
Enrolled full-year
Full-time intensity
Full-time
Part-time
Private institution
Selectivity
Institution cost (log)
Model 1
Model 2
Model 3
2.182***
2.233***
4.570***
16.788***
1.165***
17.506***
1.084***
32.273***
7.040**
10.402**
11.503***
18.268***
3.517
6.231*
5.415*
3.802***
.239
3.125***
2.227
1.276
.141
5.799***
2.748*
4.319**
1.970
3.448**
.819
.096***
6.690**
8.458**
20.420***
15.432***
14.815***
9.163***
10.820**
7.864*
8.156***
8.276**
5.631**
10.554***
6.345***
2.222*
7.209***
1.593
12.103***
3.856***
1.605
Note: Omitted categories are White, high school degree or less, engineering, and no employment during
the school year.
*p < .05; **p < .01; ***p < .001 (two-tailed test).
82

3.2
Student GPA
3.15
3.1
3.05
3
2.95
2.9
2.85
0
12
16
20
24
28
32
Parental Aid in Thousands of Dollars
Income = 5K
Income = 15K
Income = 60K
36
40
Income = 37K
Income = 90K
Figure 1. Estimated Effect of Parental Aid on Student GPA, B&B93 (N = 10,870)
Note: Model includes controls for student sociodemographics, parental SES, family structure, academic
ability, student characteristics, and institution characteristics.
Figure 1 illustrates this finding, depicting

the relationship between parental aid and
GPA at selected values of parental income,
holding all other explanatory variables at the
mean. Income values correspond to 10, 25,
50, 75, and 90 percent of the parental income
distribution. As the graph demonstrates, the
most notable differences are in GPAs of students with no aid, with the advantage going to
students in the top half of the income distribution. For example, students whose parents
make $90,000 annually and receive no aid
have an average GPA of around 3.15, whereas
students whose parents make $5,000 start
below 3.05. However, as aid increases, the
curves begin to converge. By $16,000 in aid,
all students are pulled below the 3.00 mark
a critical threshold for many graduate programs and employers. Across values of
parental income, the highest level of parental
aid is associated with an average GPA of
around 2.95.21
Patterns across the selected incomes, while
slightly different, are highly consistent. The
curve for the most privileged students is the
steepest, but everyone experiences a significant reduction in GPAparticularly in the
first $8,000 of aid. Regardless of class background, the toll parental aid takes on GPA is
modest. Yet, any reduction in student GPA
due to parental aidwhich is typically offered
with the best of intentionsis both surprising
and important. Even small disparities in GPA
are magnified in an increasingly competitive
labor market and disadvantage graduates
when their records are considered next to
those without such deductions.
Alternative funding. Next, I look at how
alternative aid shapes the relationship between
parental aid and GPA, net of explanatory variables. Table 3 presents regression coefficients
for GPA on a relative measure of parental aid
(Model 1). Here parental aid is divided by the
total amount of funding, including grants and
scholarships, loans, work-study, other forms
of aid (e.g., veteran benefits), and funds
earned through student employment.
Because parental aid largely determines
how much alternative aidparticularly nonmerit-based aidstudents receive, the absolute and relative measures are highly
correlated (r = .828, p < .001) and have a
similar effect. Model 1 indicates that as the
Hamilton
83
Table 3. Regression Coefficients for Student GPA on Relative and Absolute Parental Aid,
Alternative Aid, and Explanatory Variables, B&B93 (N = 10,870)
Model 1: Relative
Parental Aid/Total Funding
Parental Aid (log)
Alternative Funding
Grants and scholarships (log)
Loans (log)
Work-study (log)
Other forms of aid (log)
Student employment (log)
Sociodemographics
Female
Age
Race
Black
Asian
Other race
Hispanic
Parental SES
Income (log)
Independent status
Education
Some college
College degree
Advanced degree
Family Structure
Parents married
Academic Ability
SAT/ACT score
Major
Computer science
Education
Health
Humanities
Life sciences
Math
Physical sciences
Other major
Enrolled full-year
Full-time intensity
Full-time
Part-time
Private institution
Selectivity
10.943***
Model 2: Absolute
3.637***
7.421***
3.949***
.069
3.580
.517
17.536***
1.103***
17.299***
1.074***
18.179***
3.574
6.256*
5.467*
19.155***
3.961
6.737*
5.946**
3.021***
2.661
4.128***
3.105*
2.906*
4.557**
2.335
2.650*
4.327**
1.943
3.080**
.794
3.362**
.492
.096***
.093***
6.542**
8.434**
20.477***
15.450***
14.503***
9.122***
10.581**
7.931*
7.953***
7.881**
5.516*
10.742***
6.299***
6.966**
8.269**
20.907***
15.932***
15.024***
9.437***
10.484**
8.457*
8.581***
8.361**
5.998**
9.661***
5.785***
2.771**
2.065
7.848***
2.488*
7.977***
2.063
11.827***
4.003***
2.299*
10.284***
3.791***
2.578*
the school year.
84
proportion of funding derived from parents

increases, student GPA decreases (b =
10.943, p < .001). Effects of other independent variables also remain much the same. In
this model, institutional cost reaches significance. As with selectivity, it has a negative
impact on GPA.
Model 2 presents absolute measures of all
funding forms in thousands of dollars
(logged). The negative effect of parental aid
persists (b = 3.637, p < .001). Here independent status positively affects GPA and
reflects the fact that independent students
receive less parental aid overall. Loans are the
only other aid source to have a similar effect
on student GPA (b = 3.949, p < .001), and
the coefficients are not significantly different.
On the other hand, as money for grants and
scholarships increases, so does student GPA
(b = 7.421, p < .001). Funds from work-study,
other sources of aid, and student employment
do not significantly affect student GPA.
These findings indicate that the negative
effects of parental aid and loans do not extend
to all forms of aid. In fact, funds from most
aid sources are not associated with losses to
GPAand they may even benefit students.
There is thus variation in the effect of a dollar
depending on funding source. I return to this
issue in the conclusion.
Supplemental analyses. Cross-sectional data are limited in the extent to which
they can distinguish effects of family funding
processes from spurious effects of student
characteristics. For example, it may be that
the association between parental aid and student performance is due to greater academic
talent and motivation among less privileged
students who make it to college, in comparison to the less select pool of more privileged
students who often receive more parental aid.
It is also possible that by the final year of college, many students with little parental
funding and low GPAs have simply dropped
out, creating the appearance that lower levels
of parental aid lead to higher student GPAs.
To address these possibilities and confirm
patterns established in the B&B93 analyses,22
I turn to the BPS96/01 panel data, which follow a group of students who started college in
1996, and estimate fixed-effects models in
which GPA is regressed on parental aid. Here,
I examine the effect of parental aid within the
educational careers of individual students,
effectively controlling for the potential impact
of differences between students. I include all
students, even those who do not persist in college to senior year, so as to avoid potentially
excluding low achieving students who receive
little parental aid.
I use a dichotomous indicator of aid
whether parents paid any tuition and fees
and student-reported GPA, because these
measures are available across all three
waves.23, 24 The full model includes year and
college majora potentially time-varying
factor that influences student GPA. I also
incorporate a variable that marks respondents
as transfer students if they left their original
institution. This is a rough indicator of changing institutional context, because the panel
data do not offer detailed measures of institution characteristics for all three time points.
As Table 4 indicates, when accounting for
time-invariant unobserved heterogeneity there
is still a significant, negative effect of parental
aidin this case, providing any help with
tuitionon GPA. The effect is apparent at the
bivariate level (b = 21.603, p < .001) and
when controlling for time-varying factors (b =
4.990, p < .001). As expected, student GPA
improves with time. Transferring is also associated with a higher GPA and may be due to
moving to a less selective school or entering an
institution that provides a better fit, motivating
academic focus. In addition, most majors have
a positive effect on GPA, as compared to engineering. These results provide strong evidence
that selectivity processes are not driving the
negative relationship between parental aid and
GPA, and that this relationship is not an artifact
of using the B&B93 dataset.25
Degree Completion
It is possible that parental aid works in the
opposite direction or simply has no effect
Hamilton
85
Table 4. Coefficients from Fixed-Effects Models Regressing Student GPA on Parental Aid,
Year, Major, and Transfer Status, BPS96/01 (N = 7,210)
Parents Paid Any Tuition and Fees

Year
1998
2001
Major
Computer science
Education
Health
Humanities
Life sciences
Math
Physical sciences
Other major
Undeclared
Transfer Student
Model 1
Model 2
21.603***
4.990***
22.700***
28.862***
12.600***
8.967**
6.186
6.826
11.955
18.075***
8.967*
16.656***
10.595**
22.246***
14.325***
7.854*
36.798***
Note: Omitted categories are 1996 and engineering.

with regard to degree completion. As my

focus is now on persistence, I turn to
BPS90/94, the only large-scale, nationally
representative dataset that follows a sample
of first-year students and includes a continuous measure of parental aid. Table 5 provides
coefficients from a logistic regression model
of BA completion. I start with a bivariate
model (Model 1) that indicates a significant,
positive relationship between parental aid and
BA completion within five years (b = .615,
p < .001). In this case, class advantages likely
magnify the benefits of parental aid. The
challenge is to see if parental aids positive
effects remain when accounting for parental
class background.
As Model 2 indicates, even net of sociodemographics and parental SES, parental aid
significantly increases the likelihood of
obtaining a bachelors degree (b = .432, p <
.001). Being female also raises ones chance
of graduating, but being older reduces the
odds. Blacks are less likely to graduate than
Whites, and Asians are more likely. Here the
effects of parental income are captured primarily by a significant negative coefficient

for independent status, which is linked to both
lower parental SES and lower levels of parental aid. Students whose parents have a college
or advanced degree are significantly more
likely to graduate than students whose parents
have a high school degree or less.
Model 3 adds controls for family structure,
academic ability, student characteristics, and
institution characteristics. Parental aid still
significantly boosts the odds of graduating
(b = .144, p < .01). Consistent with prior
work, racial effects are not apparent when
controlling for a full array of direct and indirect effects of parental SES (e.g., attending a
more selective and higher cost institution
both of which are associated with greater
odds of obtaining a degree).
Increases in first-year GPA also raise the
likelihood of persisting to graduation.26 Students with parents married to each other are
significantly more likely to graduate. Most
majors are at least as likely, or more likely, to
86
Table 5. Logistic Regression Coefficients for BA Completion on Parental Aid (Log) and
Explanatory Variables, BPS90/94 (N = 3,810)
Parental Aid (log)

Sociodemographics
Female
Age
Race
Black
Asian
Other race
Hispanic
Parental SES
Income
Independent status
Education
Some college
College degree
Advanced degree
Family Structure
Parents married
Academic Ability
SAT/ACT score
1989 to 1990 GPA
Major
Education
Health
Humanities
Life sciences
Math
Physical sciences
Other major
Undecided
Enrolled full-year
Full-time intensity
Full-time
Part-time
Private institution
Selectivity
Model 1
Model 2
.615***
.432***
.144**
.299***
.072***
.273**
.041*
.279*
.387*
.405
.118
.001
.304
.550
.090
.078
.568**
.086
.579**
.137
.333***
.522***
Model 3
.152
.216
.283*
.210*
.026
.000
.007***
.724***
.616**
.249
.399*
.666**
.271
.871*
.991***
.328
.413*
.749***
.709***
.451***
.134
.055
.084
.343**
.192***
.866***
the school year.
Hamilton
87
Predicted Probability of Obtaining

Bachelors Degree
90%
85%
80%
75%
70%
65%
60%
55%
50%
45%
40%
12
16
20
24
28
32
36
40
Parental Aid in Thousands of Dollars

Figure 2. Estimated Effect of Parental Aid on BA Completion, BPS90/94 (N = 3,810)
Note: Model includes controls for student sociodemographics, parental SES, family structure, academic
ability, student characteristics, and institution characteristics.
obtain a BA than engineering majors. The

only exception is students who are undecided
as of year one. Students enrolled full-year and
full-time are also more likely to graduate.
Notably, student employment, at any level,
does not influence the odds of graduating.
Finally, being at a private institutiononly
when selectivity and institution cost are
accounted foris associated with a decrease
in the likelihood of graduating.
Unlike for GPA, effects of parental aid on
degree completion do not vary by parental
income or education. Figure 2 presents this
relationship graphically, holding all explanatory variables at the mean. It is clear this is a
case of diminishing returns: the greatest
effects of parental aid come with the first
several thousand dollars. In fact, students
with no parental aid have a 56.4 percent predicted probability of graduating and students
who receive $4,000 in aid have a 62 percent
predicted probabilitynearly a 6 percent
increase. At this point, the benefits associated
with additional aid increase only incrementally and begin to plateau around $12,000,
where students have a 65.2 percent predicted
probability of graduating. At $40,000, the
predicted probability is 68.8 percenta gain
of around 3 percent for over three times the

money.
Alternative funding. Finally, I look at
how alternative aid influences the relationship
between parental aid and BA completion. Table
6 displays regression coefficients for a relative
model of parental aid (Model 1) and an absolute
model (Model 2) including all forms of funding
separately in thousands of dollars logged. As in
the GPA analyses, relative and absolute measures of parental aid are highly correlated (r =
.887, p < .001). Both models include the
explanatory variables discussed earlier.
Model 1 demonstrates that when considering parental aid as a percentage of the total
funding package, it has a significant and
positive effect on graduating (b = .444, p <
.01). In Model 2, using an absolute measure
of parental aid, the positive effect is again visible (b = .164, p < .01). Work-study funds also
have a significant and positive effect on BA
completion that is not significantly different
from that of parental aid (b = .487, p < .05).
In contrast, student employment negatively
affects graduation within five years (b =
.225, p < .01). All other aid forms show no
effects on graduating.
88
Table 6. Logistic Regression Coefficients for BA Completion on Relative and Absolute

Parental Aid, Alternative Aid, and Explanatory Variables, BPS90/94 (N = 3,810)
Parental Aid/Total Funding

Parental Aid (log)
Alternative Funding
Grants and scholarships (log)
Loans (log)
Work-study (log)
Other forms of aid (log)
Student employment (log)
Sociodemographics
Female
Age
Race
Black
Asian
Other race
Hispanic
Parental SES
Income (log)
Independent status
Education
Some college
College degree
Advanced degree
Family Structure
Parents married
Academic Ability
SAT/ACT score
1989 to 1990 GPA
Major
Education
Health
Humanities
Life sciences
Math
Physical sciences
Other major
Undecided
Enrolled full-year
Full-time intensity
Full-time
Part-time
Private institution
Selectivity
Model 1: Relative
Model 2: Absolute
.444**
.164**
.127
.085
.487*
.138
.225**
.269**
.040
.243**
.036
.006
.307
.548
.083
.066
.261
.421
.116
.078
.588**
.147***
.418
.151
.210
.273*
.158
.225
.285*
.207*
.021
.210*
.000
.007***
.000
.006***
.717***
.611**
.244
.395*
.666**
.771***
.616**
.235
.397*
.670**
.275
.865*
.987***
.355
.854*
.980***
.330
.402*
.754***
.718***
.421**
.383
.430*
.728***
.645***
.399**
.129
.124
.078
.065
.125
.096
.325*
.195***
.893***
.388**
.191***
.797***
the school year.
Hamilton
89
CONCLUSIONS
Parents often assume that one of the best
ways they can direct their money is toward
their childrens college education. However,
we have little empirical knowledge of how
parental funds shape student performance
during college. Within sociology, this intellectual gap can be attributed to a legacy of
research from status attainment, human capital, and cultural perspectives detailing the
positive impact of parental investments on
childrens academic, occupational, and economic outcomes. An alternative framework
derived from moral hazard theory suggests
that parental investments may actually provide an educational disincentive for children
by offloading costs of poor performance on to
parents.
In this article, I assessed effects of parental
financial investments on two key student outcomes: GPA and bachelors degree completion.
The findings provide mixed support for the
above perspectives. On the one hand, results
from BA completion analyses are most compatible with a more-is-more approach, as expected
within sociology of education. Although the
pattern is one of diminishing returns, parental
investments significantly increase students
likelihood of obtaining a BA, net of explanatory variables including first-year GPA and
accounting for money provided by alternative
funding sourcesgrants and scholarships,
loans, work-study, other aid (e.g., veteran benefits), and student employment.
Results from GPA analyses are surprising
and run counter to general expectations of how
parental investments should operate; they are,
however, consistent with a more-is-less
approach. As parental aid increases, student
GPA decreases, even net of sociodemographics, parental SES, family structure, academic
ability, student characteristics, and institution
characteristics. Although the effect is not linear,
and, ironically, the most harm comes from initial aid, increasing investments provide a gradual drag on student GPA. This pattern remains
when accounting for alternative sources of
funding. Supplemental analyses using fixed-
effects models confirm that more is not more

when it comes to college GPA; these analyses
suggest that results are not driven by unique
characteristics of students who reach college
without any parental assistance.
The patterns presented here highlight the
importance of separating parental social class
from the provision of parental funds. Class
advantages can obscure parental aids negative effects on GPA while magnifying parental funds positive influence on degree
completion. This is a problem in much
research on college financial aid, which uses
indicators of parental SES but fails to include
parental aid. In addition, the level of available
resources is not always the same as the
amount deployed. Some parents provide more
than they can reasonably afford, while others
intentionally hold back. Previous work thus
misses one mechanism that may help to
explain variation in educational outcomes
within social class groups.
Such seemingly contradictory findings
highlight the importance of examining multiple educational outcomes at once. Without
looking at BA completion, the story is one in
which parental investments are detrimental to
student performance. Without looking at student GPA, parental investments appear
entirely unproblematic. Instead, the situation
is more complex. Reconciling this set of findings requires moving beyond either a more-ismore or a more-is-less framework.
Although they come to diametric conclusions, sociological research on parental investmentsparticularly within status attainment
and human capital traditionsshares basic
assumptions with moral hazard theory. Both
approaches suggest that students are able and
willing to maximize the resources that parents
provide. As utilitarian actors, students either
use parental investments to increase their
human capital and optimize their future socioeconomic status, or they strategically limit
academic efforts to the extent that they can
avoid personal responsibility for the economic
costs of their behavior.
Neither of these situations fit this studys
findings. Students with parental funding are
90
not performing at higher levels, which would

suggest optimization of funds. Nor are they
performing at such low levels that they risk
academic probation or dropping out
expected results if students were to fully
capitalize on a lack of financial responsibility.
Instead, students with parental support are
best described as staying out of serious academic trouble but dialing down their academic efforts.
Satisficing in Higher Education

Parental funding may be best understood as
enabling satisficing behaviors among students. The concept of satisficing was first
elaborated by Simon (1955, 1957) but has
since expanded to the fields of social psychology, political sociology, social cognition, and
organizational behavior, among others.
Satisficing originates in the theory of bounded
rationality. This theory relaxes the notion that
humans are fully utilitarian actors who
encounter life decisions as if on a buffet,
where each item comes with full information
and each choice is independent of the next.
Instead, bounded rationality recognizes that
individuals often operate with incomplete
information about the consequences of their
decisions and have conflicting desires. As a
result, people tend to satisficethat is, to
meet the criteria for adequacy on multiple
fronts, rather than optimize chances for a particular outcome.
In this case, initial investments create conditions for students to lessen their scholarly
efforts. At a certain point, however, investments have a diminishing effect on GPA, as
students may not be willing to risk leaving the
institution or letting their grades drop so low
as to encounter serious consequences. Parental funds may encourage partial satisfaction
of opportunities for human capital acquisition
by staying in school, and partial satisfaction
of reduced financial responsibilities by reducing (but not abandoning) academic focus.
Satisficing is a micro-level explanation of
how college students respond to their parents
provision of resources. Satisficing does not,
however, occur in a cultural or social vacuum.

Characteristics of young adulthood, as
opposed to childhood and adolescence, may
enable satisficing behaviors among students.
In fact, Rosenfeld (2007) has dubbed the
young adulthood years, many of which are
spent in college, as an age of independence
in which youth have unprecedented social
and geographic distance from their families
of origin. Far from being the passive receptacles that much parental investment literature
assumes, young adults have a great deal of
power over how parental educational
resources are used. Instead of optimizing
parental resources in ways consistent with
academic excellence, reduced monitoring and
accountability to parents may free college
students to pursue courses of action that may
not reflect parental wishes.
Institutional context also matters. Colleges
have long offered an alternative, primarily
social, pathway to which students may devote
their attentions (Armstrong and Hamilton
2013; Horowitz 1987). Higher education
scholars have recently noted ways that student
involvement in the social side of college life
may shape academic achievement (Armstrong
and Hamilton 2013; Arum and Roksa 2011;
Stevens et al. 2008). Parental funds provide
the time, money, and proximity (i.e., living on
or near campus) necessary to delve deeply
into college peer cultures. Social integration
has positive effects on persistence (Pascarella
and Terenzini 2005), but being too involved in
campus social life may have potential negative effects on student performance. As Arum
and Roksa (2011) argue, many of todays college students are academically adriftthat
is, in college but focused on recreational rather
than academic pursuits. Time-use studies support this conclusion. Todays college students
spend an average of 28 hours a week on
classes and homework combinedless time
than an average high school student spends in
school alonebut devote a full 41 hours a
week to social and recreational purposes
(Brint et al. 2010).
Channeling parental funds into social experiences is likely not the result desired by most
Hamilton
91
parents. Unlike moral hazard, however, the

concept of satisficing does not require parents
and students interests to be at odds. In fact,
qualitative research suggests that parents are
also exposed to a cultural discourse of college
as a fun, best years of your life, coming-of-age
experience. Some parents may be onboard
with an approach that values the social, even
to the detriment of the academic, as long as
their children manage to leave with a degree.
A small percentage of such parents may draw
on social ties and cultural capital to help compensate for deficits in their childrens college
performance (Hamilton 2010). The impetus
behind investments at this point in the educational trajectory may vary markedly from
those that help children get to college.
These results indicate the importance of
recognizing variation in effects of parental
investments across educational contexts as
well as across life stages. As Coleman (1988)
suggests, parents financial capital is necessary
but not sufficient for childrens development of
human capital. Coleman focuses on the family
relationships that mediate these investments,
but the social, cultural, and institutional contexts in which they are deployed matter as
well. Emerging cross-cultural scholarship on
variation in social capitals effects has begun to
push parental investment literature in this
direction (e.g., Cheng, Martin, and Werum
2007). To date, however, financial investments
have been treated as largely immutable and
their positive effects inevitable.
Differential effects across types of student
aid highlight the fact that money is a social
medium that gains meaning through exchange,
rather than an impersonal, static resource
(Zelizer 1994). Parents are likely to find it
difficult to hold students to seemingly impersonal performance standards or to create a
sense of responsibility for the aid they provide. Loans may share similar qualities, especially if students assume parents will cover
their payments. Other types of aid require
students to do something (e.g., work, serve in
the military, or perform well in high school)
to receive funds, and many grants and scholarships have academic standards to motivate
performance. These monies may come with a
sense of having been earned rather than

bestowed (even if temporarily so, in the case
of loans).
Consideration of the meaning carried by
financial investments suggests not all parental
aid will result in satisficing. This may be the
average effect of such contributions, but we
can imagine that parents who intentionally
counter the assumption of a no-stringsattached gift may achieve quite different
results. In addition, the propensity to satisfice
likely varies with other aspects of students
personalities and identities. Some students
may be unlikely to satisfice, regardless of
how they are funded. Propensity to satisfice
may be linked to variation in cultural understandings regarding the purpose of college
and the central tasks of young adulthood.
Parental support for college, while not
entirely determined by class, is the aid form
most linked to privilegerather than need or
merit. Here we see that students receiving
high levels of parental aid enjoy an advantage
in college completion that, for many, likely
outweighs the detrimental effects to GPA.
Reliance on individual families for college
funding thus likely advantages privileged students in terms of access and persistence.
Indeed, research shows that the large class
gap in college completion has grown as federal and state support for higher education has
declined (Bailey and Dynarski 2011). At this
historical moment, disparities in the ability to
fund a young-adult life stage may be one of
the central mechanisms through which class
inequalities are reproduced.
Future research will require far richer data
on parentchild interactions in higher education. Existing financial information, including
recent reports of the specific amounts parents
spend on college (much less why, how, and for
what) is quite limited. Other forms of educational investments that are standard fare in K to
12 research, such as frequency of conversations
with children, are not even assessed in most
national postsecondary datasets. Parenting may
look different at this level of education, as it is
sometimes accomplished primarily through
wireless communications and financial transfers. However, as my findings demonstrate,
92
parents play an active role in shaping young

adults educational experiencesperhaps in
ways they do not intend.
Funding
This research was supported by a grant from the American Educational Research Association, which recieves
funds for its AERA Grants Program from the National
Science Foundation and the National Center for Education Statistics of the Institute of Education Sciences (U.S.
Department of Education) under NSF Grant #DRL0634035. Opinions reflect those of the author and do not
necessarily reflect those of the granting agencies.
Acknowledgments
I wish to thank Elizabeth A. Armstrong, Kyle Dodson,
Neil Gross, Eliza Pavalko, Brian Powell, Patricia McManus,
Rob Robinson, and Regina Werum for comments.
Notes
1. One recent exception is Pell Grant expenditures,
which have nearly doubled since 2008. It is unlikely,
however, that this rate of increase will be sustained
in the coming years (College Board 2012).
2. EFC formulas rely on family income, assets, benefits, size, and the number of family members
attending college. The definition of family is based
on legal dependency status.
3. In earliest usage, moral hazard referred to individuals
whose problematic behaviors made them the scourge
of the insurance industry. The term was later reframed
as inefficiencies that can occur when individuals can
displace risk, rather than deficiencies of the individuals themselves. However, the concept is still not
politically neutral and has been used to push conservative legislation and public policies (Baker 1996).
Here I use the term to name a process, andto the
extent possibledivest it of political leanings.
4. Results also remain consistent when not using
imputed values for parental aid.
5. I rounded sample sizes to the nearest 10 students,
per U.S. Department of Education Institute of Education Sciences guidelines for restricted data use.
6. Including these cases in imputation yields nearly
identical results.
7. I randomly selected this dataset for ease of presentation when providing descriptions of the data.
8. The B&B93 and BPS90/94 report transcript data
normalized on a 4.0 scale. A dummy variable indicating the small number of cases in which the original
GPA was reported on a different scale is not significant when included in supplemental analyses.
9. The B&B93 only includes a measure of cumulative
GPA. This captures current as well as prior performance. Higher education research indicates that, if
anything, parents invest more in higher performing
college students (Powell and Steelman 1995). Thus, if

we find a significant negative effect of parental aid on
GPA, these analyses may be underestimating effects
of parental aid. In the case of a significant positive
effect, effects of parental aid may be overestimated.
Fixed-effects analyses also address this issue.
10. This variable is student reported. Supplemental
analyses using cases with categorical parentreported data show the same patterns.
11. For the B&B93 and BPS90/94 analyses, the BIC
statistic indicates that a log-linear measure is
preferred.
12. Multicollinearity among covariates in the models
for GPA and BA analyses does not exceed conventional standards when using the variance inflation
factor (VIF).
13. Unlike the B&B93, the BPS90/94 does not include
computer science as a separate major but does
include an undecided category (as a significant proportion of the first-year population had not yet
declared a major).
14. For m = 1 of the B&B93, average tuition for a private school with a Barrons score of six was
$15,166, and average tuition at an in-state public
school with a Barrons score of one was $1,836.
15. Loans include subsidized, unsubsidized, and parental PLUS loans.
16. This is the only model in which analyses using a
sample produced by listwise deletion (excluding
cases with missing values on any variables) generate different results. At the bivariate level, the
parental aid coefficient is not significant. Imputed
analyses are consistent, however, with bivariate
analyses of the larger sample produced when only
cases missing on parental aid and GPA are excluded.
A conservative reading is that the negative effect of
parental aid at the bivariate level is obscured by the
positive effect of other class advantages.
17. The relationship between parental aid and student
GPA remains regardless of how parental aid is
specified.
18. Use of broader major categories (e.g., STEM
majors) does not alter the results. The pattern also
remains when looking among the third of students
who took both science/engineering and advanced
math classes.
19. Analyses examining effects of parental aid within
institutions produce the same patterns.
20. Parental aid and parental education interaction
terms are not significant.
21. The funds needed to provide parental aid can come
from a variety of sources, of which income is just
one. Parents can also take out loans in their own
names, draw on savings, and pool economic
resources from relatives. Parents may thus offer
funds that exceed income earned in a given year.
22. I replicated resultsincluding controlsacross
multiple waves of the B&B and BPS datasets.
These datasets cannot be merged due to different
wording of the parental funding items.
Hamilton
93
23. Nearly a third of students experienced a funding

change during their time in college.
24. Self-reported GPA data are not ideal but are less problematic here than in cross-sectional data where GPA
inflation varies with key explanatory factors across
individuals. GPA is also not assessed consistently
across waves. Students were asked to report on a 4.0
scale in 1996, but then asked if they have mostly As,
mostly As and Bs, and so on in 1998 and 2001. I
used the metric that NCES uses to move between a
categorical and numerical grade scale, and I assigned
the GPA midpoints (multiplied by 100) across all
three waves. Analyses for just 1998 and 2001 show
the same significant pattern, and similar results are
produced when 1996 data are converted to the categorical measure used in 1998 and 2001. If anything,
because spaces between data points are much larger
than when using a 4.0 scale, these results provide a
conservative estimate of effects of parental aid.
25. Using the BPS96/01, I also estimated randomeffects models, using first-year GPA as a time-lagged
variable and including the full set of controls. The
parental aid coefficient is still negative and significant, suggesting against reverse causality.
26. The positive effect of SAT score on college completion is entirely captured by this measure.
References
Acocella, Joan. 2008. The Child Trap: The Rise of Overparenting. The New Yorker (http://www.newyorker
.com/arts/critics/books/2008/11/17/081117crbo_
books_acocella).
Alexander, Karl L. and Bruce K. Eckland. 1977. High
School Context and College Selectivity: Institutional
Constraints in Educational Stratification. Social
Forces 56:16688.
Alon, Sigal and Marta Tienda. 2005. Assessing the
Mismatch Hypothesis: Differences in College Graduation Rates by Institutional Selectivity. Sociology
of Education 78:249315.
Amato, Paul. 2001. Children of Divorce in the 1990s:
An Update of the Amato and Keith (1991) MetaAnalysis. Journal of Family Psychology 15:35570.
Armstrong, Elizabeth A. and Laura T. Hamilton. 2013.
Paying for the Party: How College Maintains Inequality. Cambridge, MA: Harvard University Press.
Arum, Richard and Josipa Roksa. 2011. Academically
Adrift: Limited Learning on College Campuses. Chicago: University of Chicago Press.
Bailey, Martha J. and Susan M. Dynarski. 2011. Gains
and Gaps: Changing Inequality in U.S. College Entry
and Completion. National Bureau of Economic
Research, Working Paper No. 17633.
Baker, Tom. 1996. On the Genealogy of Moral Hazard.
Texas Law Review 75:23792.
Baum, Sandy and Patricia Steele. 2007. Trends in Student Aid. College Board Trends in Higher Education
Series. Washington, DC: College Board.
Baum, Sandy and Jennifer Ma. 2010. Trends in College

Pricing. College Board Trends in Higher Education
Series. Washington, DC: College Board.
Becker, Gary S. 1964. Human Capital. New York:
National Bureau of Economic Research.
Blau, Peter and Otis Dudley Duncan. 1967. The American Occupational Structure. New York: Wiley.
Bodvarsson, Orn B. and Rosemary L. Walker. 2004. Do
Parental Cash Transfers Weaken Performance in College? Economics of Education Review 23:48395.
Bourdieu, Pierre. 1984. Distinction: A Social Critique
of the Judgment of Taste. Cambridge, MA: Harvard
University Press.
Bowen, William and Harold Shapiro. 1998. Forward.
Pp. ixxi in The Student Aid Game: Meeting Need
and Rewarding Talent in American Higher Education, edited by M. McPherson and M. O. Schapiro.
Princeton, NJ: Princeton University Press.
Brint, Steven, John Aubrey Douglass, Gregg Thomson,
and Steve Chatman. 2010. Engaged Learning in a
Public University: Trends in Undergraduate Experience. Berkeley, CA: Center for Studies in Higher
Education.
Brown, Sherry M. 2002. Strategies that Contribute
to Nontraditional/Adult Student Development and
Persistence. PAACE Journal of Lifelong Learning
11:6776.
Buchmann, Claudia and Thomas A. DiPrete. 2006. The
Growing Female Advantage in College Completion: The
Role of Family Background and Academic Achievement. American Sociological Review 71:51541.
Burton, Lorelle J., Janet A. Taylor, David G. Dowling, and
Jill Lawrence. 2009. Learning Approaches, Personality and Concepts of Knowledge of First-Year Students:
Mature-Age versus School Leaver. Studies in Learning, Evaluation, Innovation, and Development 6:6581.
Cabrera, Alberto F., Amaury Nora, and Maria B. Castaneda. 1993. College Persistence: Structural Equations Modeling Test of an Integrated Model of Student
Retention. Journal of Higher Education 64:12339.
Camara, Wayne J. and Gary Echternacht. 2000. The
SAT I and High School Grades: Utility in Predicting
Success in College. The College Board: Research
Notes RN-10, July.
Carnevale, Anthony P. and Jeff Strohl. 2010. How
Increasing College Access Is Increasing Inequality,
and What to Do about It. Pp. 71183 in Rewarding
Strivers: Helping Low-Income Students Succeed in
College, edited by R. D. Kahlenberg. New York: The
Century Foundation Press.
Charles, Camille Z., Vincent C. Roscigno, and Kimberly C. Torres. 2007. Racial Inequality and College
Attendance. Social Science Research 36:32952.
Cheng, Simon, Leslie Martin, and Regina E. Werum.
2007. Adult Social Capital and Track Placement of
Ethnic Groups in Germany. American Journal of
Education 114:4174.
Coleman, James S. 1988. Social Capital in the Creation
of Human Capital. American Journal of Sociology
94:95120.
94
College Board. 2012. Pell Grants: Total Expenditures,

Maximum and Average Grants, and Number of
Recipients over Time. Trends in Higher Education. Retrieved December 6, 2012 (http://trends
.collegeboard.org/student-aid/figures-tables/fed-aidpell-grants-total-expenditures-maximum-and-average-grants-and-number-recipients-over-time).
Curtis, Gary and Carole E. Nummer. 1991. To Work or
Not to Work: That Is the Question. Journal of Student Financial Aid 21:1626.
Danziger, Sheldon and Cecilia Elena Rouse, eds. 2007.
The Price of Independence: The Economics of Early
Adulthood. New York: Russell Sage.
Davila, Alberto and Marie T. Mora. 2004. The Scholastic Progress of Students with Entrepreneurial Parents. Economics of Education Review 23:28799.
Deil-Amen, Regina and Ruth Lopez Turley. 2007. A
Review of the Transition to College Literature in
Sociology. Teachers College Record 109:232466.
DesJardins, Stephen L., Dennis A. Ahlburg, and Brian P.
McCall. 2002. Simulating the Longitudinal Effects of
Changes in Financial Aid on Student Departure from
College. Journal of Human Resources 37:65379.
DesJardins, Stephen L., Dennis A. Ahlburg, and Brian
P. McCall. 2006. The Effects of Interrupted Enrollment on Graduation from College: Racial, Income,
and Ability Differences. Economics of Education
Review 25:57590.
Downey, Douglas B. 1995. When Bigger Is Not Better: Family Size, Parental Resources, and Childrens
Educational Performance. American Sociological
Review 60:74661.
Dynarski, Susan. 2003. Does Aid Matter? Measuring
the Effect of Student Aid on College Attendance and
Completion. American Economic Review 93:27988.
Fain, Paul. 2009. At Public Universities: Less for
More. New York Times, October 26 (http://www.
nytimes.com/2009/11/01/education/edlife/01public-t
.html?_r=1).
Fischer, Mary J. 2007. Settling into Campus Life: Differences by Race/Ethnicity in College Involvement and
Outcomes. Journal of Higher Education 78:12561.
Flint, Thomas A. 1992. Parental and Planning Influences
on the Formation of Student College Choice Sets.
Research in Higher Education 33:689708.
Flint, Thomas A. 1997. Intergenerational Effects of
Paying for College. Research in Higher Education
38:31344.
Furstenberg, Frank F., Jr., Ruben G. Rumbaut, and Richard A. Settersten. 2005. On the Frontier of Adulthood: Emerging Themes and New Directions. Pp.
328 in On the Frontier of Adulthood, edited by R.
A. Settersten, F. F. Furstenberg Jr., and R. Rumbaut.
Chicago: University of Chicago Press.
Gambetta, Diego. 1987. Were They Pushed or Did They
Jump? Individual Decision Mechanisms in Education. Cambridge, UK: Cambridge University Press.
Hamilton, Laura T. 2010. Strategies for Success: Parental Funding, College Achievement, and the Transition
to Adulthood. PhD dissertation, Department of Sociology, Indiana University, Bloomington, IN.

Hamilton, Laura, Simon Cheng, and Brian Powell. 2007.
Adoptive Parents, Adaptive Parents: Evaluating the
Importance of Biological Ties for Parental Investment. American Sociological Review 72:95116.
Hays, Sharon. 1996. The Cultural Contradictions of
Motherhood. New Haven, CT: Yale University Press.
Heimer, Carol A. 1988. Social Structure, Psychology,
and the Estimation of Risk. Annual Review of Sociology 14:491519.
Henry, Gary T. and Ross Rubenstein. 2002. Paying for
Grades: Impact of Merit-Based Financial Aid on
Educational Quality. Journal of Policy Analysis and
Management 21:93109.
Horowitz, Helen Lefkowitz. 1987. Campus Life: Undergraduate Cultures from the End of the Eighteenth Century to
the Present. Chicago: University of Chicago Press.
Jencks, Christopher, Marshall Smith, Henry Acland,
May Jo Bane, David Cohen, Herbert Gintis, Barbara
Heyns, and Stephanie Michelson. 1972. Inequality: A
Reassessment of the Effect of Family and Schooling
in America: New York: Basic Books.
Jones, Ethel B. and John D. Jackson. 1990. College
Grades and Labor Market Rewards. Journal of
Human Resources 25:25366.
Kalenkoski, Charlene Marie and Sabrina Wulff Pabilonia.
2008. Parental Transfers, Student Achievement, and
the Labor Supply of College Students. U.S. Bureau
of Labor Statistics, Working Paper 416.
Kim, Doo Hwan and Barbara L. Schneider. 2005. Social
Capital in Action: Alignment of Parental Support in
Adolescents Transition to Postsecondary Education.
Social Forces 84:11811206.
Kuncel, Nathan R., Marcus Crede, and Lisa L.
Thomas. 2005. The Validity of Self-Reported
Grade Point Averages, Class Ranks, and Test
Scores: A Meta-Analysis and Review of the Literature. Review of Educational Research 75. doi:
10.3102/00346543075001063.
Lareau, Annette. 2011. Unequal Childhoods: Class,
Race, and Family Life, Second Edition with an
Update a Decade Later. Berkeley: University of California Press.
Little, Roderick J. A. and Donald B. Rubin. 2002. Statistical Analysis with Missing Data. Hoboken, NJ: Wiley
Interscience.
Loury, Linda D. and David Garman. 1995. College
Selectivity and Earnings. Journal of Labor Economics 13:289308.
Lucas, Samuel R. 1996. Selective Attrition in a Newly
Hostile Regime: The Case of 1980 Sophomores.
Social Forces 75:51133.
McPherson, Michael S. and Morton O. Schapiro. 1998.
The Student Aid Game: Meeting Need and Rewarding
Talent in American Higher Education. Princeton, NJ:
Princeton University Press.
Mullen, Ann L., Kimberly A. Goyette, and Joseph A.
Soares. 2003. Who Goes to Graduate School? Social
Hamilton
95
and Academic Correlates of Educational Continuation

after College. Sociology of Education 76:14369.
Newman, Katherine and Sofya Aptekar. 2007. Sticking
Around: Delayed Departure from the Parental Nest in
Western Europe. Pp. 207230 in The Price of Independence: The Economics of Early Adulthood, edited by
S. Danziger and C. E. Rouse. New York: Russell Sage.
Olson, Lorayn and Rachel A. Rosenfeld. 1984. Parents
and the Process of Gaining Access to Student Financial Aid. Journal of Higher Education 5:45580.
Osgood, Wayne, Gretchen Ruth, Jacquelynne Eccles, Janis
Jacobs, and Bonnie Barber. 2005. Six Paths to Adulthood: Fast Starters, Parents without Careers, Educated
Partners, Educated Singles, Working Singles, and Slow
Starters. Pp. 32055 in On the Frontier of Adulthood:
Theory, Research, and Public Policy, edited by R. A.
Settersten, F. F. Furstenberg, and R. G. Rumbaut. Chicago: University of Chicago Press.
Pascarella, Ernest T. and Patrick T. Terenzini. 2005.
How College Affects Students: A Third Decade of
Research. San Francisco: Jossey-Bass.
Paulsen, Michael B. and Edward P. St. John. 2002.
Social Class and College Costs: Examining the
Financial Nexus between College Choice and Persistence. Journal of Higher Education 73:189236.
Pike, Gary R., George D. Kuh, and Ryan C. MassaMcKinley. 2008. First-Year Students Employment,
Engagement, and Academic Achievement: Untangling the Relationship between Work and Grades.
NASPA Journal 45:56082.
Powell, Brian and Lala Carr Steelman. 1995. Feeling
the Pinch: Child Spacing and Constraints on Parental
Economic Investments in Children. Social Forces
73:146586.
Price, Derek V. 2004. Borrowing Inequality: Race, Class,
and Student Loans. Boulder, CO: Lynne Rienner.
Rojstaczer, Stuart. 2002. National Trends in Grade Inflation, American Colleges and Universities. Retrieved
December 6, 2012 (http://gradeinflation.com).
Rosenfeld, Michael J. 2007. The Age of Independence:
Interracial Unions, Same-Sex Unions, and the
Changing American Family. Cambridge, MA: Harvard University Press.
Royston, Patrick. 2005. Multiple Imputation of Missing
Values: Update. The Stata Journal 5:114.
Sax, Linda J. and Casandra E. Harper. 2007. Origins of
the Gender Gap: Pre-college and College Influences
on Differences between Men and Women. Research
in Higher Education 48:66994.
Settersten, Richard A. and Barbara E. Ray. 2010. Not
Quite Adults: Why 20-Somethings Are Choosing a
Slower Path to Adulthood, and Why Its Good for
Everyone. New York: Bantam Books.
Sewell, William H. and Robert M. Hauser. 1976. Causes
and Consequences of Higher Education: Models
of Status Attainment Process. Pp. 928 in Schooling and Achievement in American Society, edited by
W. H. Sewell, R. M. Hauser, and D. L. Featherman.

New York: Academic Press.
Simon, Herbert A. 1955. A Behavioral Model of Rational
Choice. Quarterly Journal of Economics 59:99118.
Simon, Herbert A. 1957. Models of Man, Social and
Rational: Mathematical Essays on Rational Human
Behavior. New York: Wiley.
Stater, Mark. 2009. The Impact of Financial Aid on
College GPA at Three Flagship Public Institutions.
American Educational Research Journal. doi:
10.3102/0002831208329903.
Steelman, Lala Carr and Brian Powell. 1989. Acquiring Capital for College: The Constraints of Family Configuration. American Sociological Review
54:84455.
Stevens, Mitchell L., Elizabeth A. Armstrong, and Richard Arum. 2008. Sieve, Incubator, Temple, Hub:
Empirical and Theoretical Advances in the Sociology
of Higher Education. Annual Review of Sociology
34:12751.
Teachman, Jay D., Kathleen Paasch, and Karen Carver.
1997. Social Capital and the Generation of Human
Capital. Social Forces 75:134359.
Torche, Florencia. 2011. Is a College Degree Still the
Great Equalizer? Intergenerational Mobility across
Levels of Schooling in the United States. American
Journal of Sociology 117:763807.
Turley, Ruth N. Lopez and Matthew Desmond. 2011.
Contributions to College Costs by Married,
Divorced, and Remarried Parents. Journal of Family
Issues 32:76790.
von Hippel, Paul T. 2007. Regression with Missing
Ys: An Improved Strategy for Analyzing Multiply
Imputed Data. Sociological Methodology 37:83117.
Warner, Judith. 2006. Perfect Madness: Motherhood in
the Age of Anxiety. New York: Riverhead Books.
Zelizer, Viviana A. Rotman. 1985. Pricing the Priceless
Child: The Changing Social Value of Children. New
York: Basic Books.
Zelizer, Viviana A. Rotman. 1994. The Social Meaning of
Money. New York: Basic Books.
Laura T. Hamilton is an Assistant Professor at the University of California-Merced. Her dissertation, Strategies
for Success: Parental Funding, College Achievement,
and the Transition to Adulthood, received the 2011 ASA
Dissertation Award honorable mention. Her book with
Elizabeth A. Armstrong, Paying for the Party: How College Maintains Inequality, examines how the organization
of public universities disadvantages all but the most affluent. Her other work explores parental investments in
adoptive youth (see Adoptive Parents, Adaptive Parents in American Sociological Review) and the
intersection of gender, class, and sexuality among college
students (see Gendered Sexuality in Young Adulthood
in Gender & Society).

American Sociological Review 2013 Hamilton 70 95

Uploaded by

Copyright:

Available Formats

American Sociological Review 2013 Hamilton 70 95

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

American Sociological Review 2013 Hamilton 70 95

Uploaded by

Copyright:

Available Formats

American http://asr.sagepub.

More Is More or More Is Less? Parental Financial Investments during College

American Sociological Association

>> Version of Record - Jan 30, 2013

Downloaded from asr.sagepub.com by guest on September 1, 2014

More Is More or More Is Less?