W 25069
W 25069
W 25069
We thank Dan Black, Hoyt Bleakley, John Bound, David Deming, Sue Dynarski, Jose Galdo,
Josh Goodman, Audrey Light, Peter Mueser, Sarah Turner, Ophira Vishkin, and Martin West for
helpful comments, along with seminar participants at Aarhus University, the 2014 Bergen-
Stavanger Workshop on Labour Markets, Families and Children, the 2013 CESifo education
group meetings, Bristol, Chicago (family economics), Cleveland Fed, Colorado, Cornell PAM,
Dartmouth, Guelph, HEC, Institute for Fiscal Studies, MDRC, Michigan CIERS, Mannheim,
National University of Singapore, Ohio State, the 2017 Ottawa-Carlton Graduate School in
Economics Launch Conference, Penn GSE, Seton Hall Conference on College Match, Stanford
CEPA, Toronto, UIC, Washington, Washington University in St. Louis, and Wilfred Laurier, and
students in the Winter 2017 versions of Economics 622 and Public Policy 713 at the University of
Michigan, particularly Ellen Stuart. This research was supported by NSF SES 0915467. Any
foolishness the careful reader might uncover is our own damn fault. Please be sure and tell us
about it. The views expressed herein are those of the authors and do not necessarily reflect the
views of the National Bureau of Economic Research.
NBER working papers are circulated for discussion and comment purposes. They have not been
peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies
official NBER publications.
© 2018 by Eleanor Wiske Dillon and Jeffrey A. Smith. All rights reserved. Short sections of text,
not to exceed two paragraphs, may be quoted without explicit permission provided that full
credit, including © notice, is given to the source.
The Consequences of Academic Match between Students and Colleges
Eleanor Wiske Dillon and Jeffrey A. Smith
NBER Working Paper No. 25069
September 2018
JEL No. I23,I26,J24
ABSTRACT
We consider the effects of student ability, college quality, and the interaction between the two on
academic outcomes and earnings using data on two cohorts of college enrollees. Student ability
and college quality strongly improve degree completion and earnings for all students. We find
evidence of meaningful complementarity between student ability and college quality in degree
completion at four years and long-term earnings, but not in degree completion at six years or
STEM degree completion. This complementarity implies some tradeoff between equity and
efficiency for policies that move lower ability students to higher quality colleges.
Jeffrey A. Smith
Department of Economics
University of Wisconsin
1180 Observatory Drive
Madison, WI 53706-1393
and IZA
and also NBER
econjeff@ssc.wisc.edu
1. Introduction
How students of varying ability sort into colleges of varying qualities has captured the attention
not only of academic researchers studying higher education but also of the policy literature, the
popular press, and the blogosphere. Much of the literature frames the discussion in terms of the
match between student ability and college quality, with relatively low ability students at relatively
high quality colleges labeled “overmatched,” and relatively high ability students at relatively low
quality colleges labeled “undermatched.” Until the last decade, the literature focused almost
exclusively on overmatch, particularly overmatch induced by racial and ethnic preference policies
at selective colleges. More recently, undermatch has moved into the spotlight via the widely read
studies by Bowen, Chingos and McPherson (2009) and Roderick et al. (2008). While the “match”
terminology has normative overtones, we use it simply to define one possible sorting mechanism.
The individual and social desirability of full assortative matching is among the empirical questions
we seek to address.
We ask whether college quality has different effects for students of different abilities. Many
studies conclude that increased college resources improve average student outcomes. These main
effects of college quality have relevance for mismatched students, since over- or under-matching
implies moving up or down the college quality distribution. However, despite the current ubiquity
of the match conversation, the literature offers few credible estimates of the interaction effects of
student-college match. The applied theory literature on college sorting, such as Rothschild and
White (1995) and Sallee, Resch, and Courant (2008), posit complementarities between student
ability and college quality; casual discussion about “fit” often presumes it. We look for evidence
of these complementarities or of alternative relationships. In the presence of any kind of
differential effects, resorting students via policy, even when respecting existing capacity
constraints, has the potential to produce gains or losses in both efficiency and equity. In contrast,
if the effects of college quality do not vary by student ability, then resorting can yield only equity
gains. Knowledge of the effects (if any) of academic match has clear value to students and parents
making decisions about college enrollment, and to researchers and policymakers concerned with
the design, operation, and effects of state university systems with diversified college quality
portfolios.
Our analysis contributes to the small but growing literature on the effects of academic
match at the college level in a number of important ways. First, we clarify the conceptual
1
distinction between the main effects of college quality and student ability and the match effects
that may or may not result from their interaction.1
Second, we present estimates from the 1979 and 1997 cohorts of the National Longitudinal
Survey of Youth (hereinafter NLSY-79 and NLSY-97). By considering two different datasets, we
can examine the stability of our estimates between cohorts of college students separated by over
two decades, during which post-secondary education in the United States changed in important
ways. We code our outcome and conditioning variables and design our analyses in the same way
for the two datasets in order to make our analysis across cohorts as compelling as possible. Our
analysis also implicitly replicates (in a broad sense) and extends the earlier analyses of the college
quality main effect in the NLSY-79 presented in Black et al. (2005).
Third, we examine a variety of outcome measures. With a couple of important recent
exceptions discussed in more detail below, the earlier literature focuses primarily on degree
completion. Bowen and Bok’s (1998) early finding of no apparent impact on degree completion
for overmatched students suggested to us that these students might find other ways to deal with
better-prepared colleagues and a high-pressure environment. For example, they might follow the
increasingly common path of increased time to degree, as highlighted in Bound et al. (2010). Or
they might follow scholarship athletes at some colleges in taking easy courses, as suggested in
journalistic exposés such as Steeg et al. (2008) and Ann Arbor News (2008). Or they might transfer
to another school that represents a better match. Our examination of transfers, highlighted by
Arcidiacono and Lovenheim (2016) as an understudied outcome, as well as of earnings in the years
immediately following college enrollment, tells us more about the mechanisms through which
college quality and ability affect educational and labor market outcomes. Our analysis of earnings
up to 11 years after initial enrollment quantifies the medium-term labor market effects of college
quality, ability and match. For the NLSY-79 cohort, we present estimates for earnings up to 30
years following college start, the first longer-term estimates of match effects.
Fourth, following Black and Smith (2006) and our earlier analysis of the determinants of
academic match in Dillon and Smith (2017), we use composite indices as our measures of student
ability and college quality. We expect these measures to embody substantially less measurement
error than the single measures (e.g. the student’s own SAT score and the average SAT score of the
entering class) commonly used in the literature and so to provide more accurate estimates.
1
Kurlaender and Grodsky (2013) provide similar clarification in the sociology literature.
2
Finally, we make an explicit case for our “selection on observed variables” identification
strategy, which we view as credible in our context. Relative to the small set of existing studies of
college match that use this identification strategy, we have a richer and more compelling set of
relevant conditioning variables. Moreover, the literature provides strong evidence of the
importance of factors likely conditionally unrelated to outcomes in driving college choice; these
factors provide exogenous variation in college quality. Unlike earlier papers, we show that our
estimates stabilize as we add marginal sets of conditioning variables.
We find substantial amounts of overmatch and undermatch in both cohorts, with a modest
decline from the NLSY-79 to the NLSY-97. Our examination of the effects of ability, college
quality, and their interaction reveals substantively strong and statistically significant main effects
of college quality and student ability on degree completion and earnings. These marginal effects
of college quality are always positive, even for relatively low ability students, indicating that
resorting policies like affirmative action likely pass a private benefit-cost test, on average, for the
affected students. College quality matters relatively more, compared to ability, in the later cohort.
We find evidence of a causal effect of the interaction of quality and ability, but only for
certain outcomes. We find clear evidence of complementarity between student ability and college
quality in time to degree in both cohorts: abler students benefit relatively more from college quality
for this outcome. We also find clear evidence of complementarity for long-run earnings outcomes
in the NLSY-79 cohort; this pattern shows up in the point estimates for the later cohort as well.
These patterns indicate an equity-efficiency tradeoff associated with policies that increase the
enrollment of (relatively) less able students at high quality colleges.
Looking at mechanisms, we find strong evidence of match-related transfer behavior:
overmatched students have a higher conditional probability of transferring down and
undermatched students have a higher conditional probability of transferring up. In contrast, and
unlike Arcidiacono et al. (2016), we find no match effects related to STEM (= Science, Technology,
Engineering and Math) degree completion.
We structure the remainder of the paper as follows: Section 2 briefly reviews the literature
on academic match. Section 3 describes our excellent data, with particular attention to the
construction of our student ability and college quality measures and to the outcomes we consider.
Section 4 lays outs our econometric framework. Section 5 considers identification and makes the
case for a causal interpretation of our estimates. Section 6 presents our main findings on the
3
consequences of academic match and compares results between the two cohorts and between our
analysis and key earlier studies. Finally, Section 7 offers our conclusions.
2. Literature
We view the study of student-college match as an extension of the broader literature on the causal
effect of college quality on student outcomes.2 If students benefit from greater college resources,
then we expect academic undermatch to be mechanically costly for students, even without any role
for student-college interactions, because it implies attending a lower-resource college. Overmatch
is likewise mechanically beneficial. We review the subset of the literature on college quality that
explicitly considers match.3
A few papers devote their full attention to the determinants of academic match. Our earlier
work, Dillon and Smith (2017), uses the same NLSY-97 data we employ here and finds important
roles for financial constraints in explaining patterns of match, as well as the in-state public college
options available to the student. We show that more informed students and parents, proxied by
variables such as parental education and the fraction of high school peers attending four-year
colleges, act as though they believe that the main effect of college quality dominates any negative
effects of academic overmatch. Smith et al. (2013) conduct a similar analysis using different data
sets and a different definition of match. Reassuringly, they obtain similar findings. Lincove and
Cortes (2016) use administrative data from Texas and find, among other interesting patterns, an
important role for “social matching,” which they define as attending a college with a high share of
students in one’s own racial or ethnic group. Hoxby and Avery (2012) emphasize the role of
information about college choices in driving match decisions for high achieving students from
disadvantaged backgrounds while Griffith and Rothstein (2009) highlight the role of geographic
distance for all students.
The studies most similar to ours examine academic match using “selection on observed
variables” identification strategies to deal with non-random selection of students into colleges of
2
Recent studies that examine the “main effect” of college quality include Black and Smith (2004), Bowen et al. (2009),
Cohodes and Goodman (2014), Cunha and Miller (2014), Dale and Krueger (2002, 2014), Hoekstra (2009), Hoxby
(2015), Long (2008, 2010), and Zimmerman (2014). All agree on a positive causal effect for at least some groups.
3
We focus here on papers that address academic match at the undergraduate level and that use U.S. data. See Sander
and Taylor (2012) for a survey of the related, tendentious, literature on academic match in law school.
4
varying qualities.4 None of these studies find meaningful match effects, though several identify
strong main effects of both student ability and college quality.5 Mattern et al. (2010) use data from
a large number of colleges and a relatively limited set of observed characteristics, with student and
college quality both measured using SAT scores and discretized into quartiles. They study the
effect of academic match on first year college GPA and persistence into the same college in the
second year. The analysis in Chingos (2012) resembles our own in imposing capacity constraints
but employs different data (the National Educational Longitudinal Study), cruder measures of
student ability and college quality, a linear specification in college quality and student ability, and
a less compelling set of conditioning variables. Black et al. (2005) look at the simple interaction
of student ability and college quality in the context of a parametric linear model of log wages
applied to the NLSY-79 data. We compare our results to theirs in Section 6.7. Finally, Bowen et
al. (2009) examine match effects on college completion using impressive administrative data from
various state university systems. They use relatively crude selectivity categories for universities, a
modest conditioning variable set, and high school GPA and/or ACT/SAT scores as their measure
of student ability.6
Light and Strayer (2000) also look at academic match using the NLSY-79, but employ an
empirical approach that differs substantially from ours. They consider two sequential choices. The
first choice, which they model as a multinomial probit, consists of either not attending four-year
college (which combines entering the labor force and attending a two-year college) or going to
college in one of four ordered quality quartiles. The second choice, which they model as a probit,
concerns college completion. To address the potential for non-random selection on unobserved
variables, they allow correlated errors between the two choices. Identification comes from
4
Alon and Tienda (2005) examine academic match using the High School and Beyond and National Educational
Longitudinal Study of 1988 (NELS:88) datasets. Unfortunately, they look only for effects of selectivity (their proxy
for college quality) conditional on ability rather than for effects of the interaction of selectivity and ability, which
renders their results difficult to interpret in terms of match effects.
5
The exception is Loury and Garman (1995), who find substantively important match effects on degree completion
(including negative effects of college quality for black students) and post-college earnings in their study that uses the
National Longitudinal Study of the High School Class of 1972. Their earnings estimates condition both on a much
less rich set of background variables and on several intermediate outcomes – college GPA, major, years of college –
and so correspond to a very different estimand than our own. Their completion estimates do not have the issue of
conditioning on intermediate outcomes and so remain a puzzle. A replication in light of the subsequent literature
would add value.
6
See in particular their Figures 10.5a, 10.5b, 11.1, 11.2 and 11.3. They use the terminology of match somewhat
differently than we do; in particular, they sometimes refer to what we call the main effect of quality as a mismatch
effect when it applies to mismatched students. Cunha (2009) critiques this study.
5
conditioning on observed variables, from some (not super plausible) exclusion restrictions, from
some restrictions on the coefficients on the interactions of student ability and college quality, and
from restrictions on the covariance matrix of the errors in the college choice model. Their estimates
reveal substantively important match effects; we say more about why their qualitative findings
differ from our own in Section 6.7.
Another genre of studies focuses primarily on academic overmatch. Bowen and Bok (1998)
find strong positive college quality effects on degree completion and earnings among black
students attending the selective schools included in the “College and Beyond” data, which likely
rule out negative net effects from overmatch within this group. Arcidiacono et al. (2012) study
Duke University, where the average African-American student starts out somewhat less prepared
academically than other students, presumably due to affirmative action, but somewhat more likely
to express a desire to major in the natural sciences, engineering, or economics. The authors find
that black students at Duke differentially migrate away from these majors toward majors in the
humanities or other social sciences and show that this pattern results almost entirely from their
differential preparation. Put into our conceptual framework, their findings suggest that relatively
undermatched black students at Duke adapt by changing majors within an institutional context
where essentially all students finish their degree. Arcidiacono et al. (2016) continue this line of
work by examining students at different University of California (UC) campuses entering school
between 1995 and 1997, years prior to the ban on affirmative action in admissions in that state.
The authors provide compelling evidence that overmatched minority students at UC schools who
intend a STEM major have lower probabilities of completing any degree at a UC school and of
graduating in STEM.7
A final group of papers uses “natural” experiments or discontinuities to isolate the
experiences of students on the margin of admission to a higher quality college. Hoekstra (2009)
began this strand of the literature by using a discontinuity in student SAT scores (conditional on
high school GPA) to examine the effects of admission to, and enrollment in, a state flagship
university. He finds large positive effects of flagship acceptance and attendance on earnings 10-
15 years after high school completion for men but surprisingly small effects for women.
Kurlaender and Grodsky (2013) exploit an unusual event in 2004 in which the admissions offices
7
The fine Arcidiacono, Aucejo, Husser and Spenner (2013) paper looking at mismatch and friendship networks in
college also sheds some light on potential mechanisms.
6
of UC schools initially offered their marginal acceptances deferred admission due to a “budget
crisis” but later repented and offered them regular admission. They find that marginal students
accumulate fewer credits compared to similar students at lower-ranked UC schools, but have
higher graduation probabilities. Thus, they find evidence of academic match effects on the
“intensive” course-taking margin but the college quality effect dominates for degree completion.
The estimates in this group of papers correspond to quite narrowly defined populations of marginal
students. Moreover, these findings shed only indirect light on match. The nature and extent of
mismatch (as conceived of in the broader literature) for students at a college quality margin
depends on the particular definition of match employed as well as the quality of the student’s next
best alternative and the homogeneity of student ability within each college. Still, at the very least,
the evidence from these papers stands at odds with large, negative consequences of what we might
call local overmatch.8
Overall, we view the literature as providing strong evidence of causal effects of college
quality and student ability on academic and labor market outcomes. In contrast, most but not all
of the literature finds little in the way of academic match effects, other than on intermediate
outcomes such as transfer and major choice.
3. Data
3.1. NLSY
We use the NLSY-79 data, which includes Americans who were ages 14 to 22 on January 1, 1979,
and the NLSY-97 data, which includes Americans who were 12 to 16 years old as of December
31, 1996. In both cohorts, participants were interviewed annually starting in 1979 and 1997,
respectively, and continuing through their college years. They have been interviewed biannually
since 1994 and 2011, respectively. We include the representative samples from each survey along
with the supplemental samples of blacks and Hispanics. 9 Most respondents in the NLSY-79
sample graduated high school and made their college choice between 1975 and 1983, while the
8
In other recent RD papers, Zimmerman (2014) and Goodman et al. (2017) find substantively important effects of
college quality (and/or type) on labor market outcomes toward the other end of the college quality spectrum, namely
the margin between low quality four-year colleges and two-year colleges.
9
In the NLSY-79 we omit the military and low-income white samples because both were dropped from the survey
before most respondents had completed college. We use custom probability of inclusion weights, constructed by the
NLSY, to combine the sampling groups in each survey, and to control for differing response rates by age, sex, and
race-ethnicity.
7
NLSY-97 sample did the same between 1998 and 2002. We focus on students who enroll in a four-
year college by age 21 (39% of high school graduates and GED holders in the NLSY-79 sample
and 36% in the NLSY-97 sample).
One of the strengths of the NLSY data for both cohorts lies in the rich set of individual and
family covariates it provides. 10 Using the restricted access geocode data provides additional
information on the identities of colleges attended and allows the use of contextual information
based on the respondent’s residential location. The following sections describe our ability and
college quality measures, as well as our outcome variables; Appendix Tables A1, A2, A3 and A4
describe the construction of our analysis sample and summarize our conditioning variables.
3.2. Ability
We follow Dillon and Smith (2017) in designing our measures of student ability and college quality
for the NLSY-97 sample and construct comparable measures for the earlier NLSY-79 cohort. Our
measures of student ability draw on the Armed Services Vocational Aptitude Battery (ASVAB).
In the 1997 cohort, 86% of respondents who started at a four-year college completed the test; the
corresponding number for the 1979 cohort is 93%. We use the method developed by Altonji et al.
(2012) to construct comparable measures of eight of the ASVAB test components common to the
two cohorts, adjusting for the transition between 1979 and 1997 from pen-and-paper to computer
adaptive testing and for the varying ages at which the respondents took the test. We do not use the
scores on the purely vocational components.
We then construct the first two principal components of these eight section scores. Our
primary measure of ability, which we call ASVAB1, equals each respondent’s percentile of the
first principal component within the sample distribution of college-bound respondents in their
NLSY cohort.11 As shown in Appendix Table A5, the first principal component explains 68% and
66% of the total variance in test scores across the eight sections for the 1979 and 1997 cohorts,
respectively. In both cohorts, the first component places the highest weight on academic subjects
10
The NLSY datasets also feature impressively high response rates, over 80 percent of the initial respondents in
most survey rounds. See https://www.nlsinfo.org/content/cohorts/nlsy79/intro-to-the-sample/retention-reasons-
noninterview and https://www.nlsinfo.org/content/cohorts/nlsy97/intro-to-the-sample/retention-reasons-non-
interview.
11
The ASVAB test is not a straightforward measure of “innate” ability because it includes the influences and training
that the student has experienced up to the point she takes the test. See Neal and Johnson (1996) for a more thorough
discussion of what the ASVAB test measures. We do not mind if the ASVAB also measures intrinsic motivation, as
argued by Segal (2012). More broadly, we use the term “ability” quite agnostically to mean the set of skills, innate or
acquired, that students possess around the time of the college choice.
8
such as arithmetic reasoning and paragraph comprehension. Not surprisingly giving the loadings,
the correlation between ASVAB1 and the respondent’s SAT (or rescaled ACT) score equals 0.79
in NLSY-79 and 0.80 in NLSY-97.
The second component of the ASVAB scores, which we call ASVAB2, explains a further
10-11% of the variance. As in Cawley et al. (2001), who perform a similar analysis using the
NLSY-79 data, the second component places the most weight on the two timed sections of the test:
numerical operations and coding speed. We include ASVAB2 as an additional control variable in
our multivariate analyses. To capture further dimensions of ability we also include high school
GPA and SAT scores along with multiple proxies for non-cognitive or socio-emotional skills.12
3.3. College quality
We construct a one-dimensional index of college quality by combining measures related to
selectivity and college resources. The available data limit us to using measures of inputs as proxies
for quality, but we note that Hoxby (2015)’s value-added estimates correlate with one important
component of our index, namely college selectivity. In particular, our index combines the mean
SAT or ACT score of entering students, the percent of applicants rejected, the average salary of
all faculty engaged in instruction, and the undergraduate faculty-student ratio. We combine data
from the U.S. Department of Education’s Integrated Post-Secondary Education Data System
(IPEDS) and U.S. News and World Report, using data from 1992 for the NLSY-79 and data from
2008 for the NLSY-97.13
Following Black and Smith (2004, 2006) and Black et al. (2005) we estimate the principal
components across these four measures of quality. We use the eigenvector of the first principal
component (reported in Appendix Table A6) to calculate a weighted average of the proxies
available for each college, then calculate percentiles of this average across colleges.14 We interpret
12
We follow Aucejo and James (2017) and include an index of petty anti-social behaviors before age 14 and early
sexual activity. Following Cadena and Keys (2015), we also include an indicator of whether the NLSY interviewer
rated the respondent as somewhat uncooperative in any of the first three rounds of interviews.
13
U.S. News and IPEDS collect many of the same statistics and often report identical values. Combining data from
the two sources gives us the most complete sample of colleges. We measure college quality somewhat later than the
years when each cohort entered college. Many of the component measures first become available in IPEDS in 1992
and 2008 is the earliest year for which we could obtain recent U.S. News data. In both cases, the stability of the
underlying proxies over time assuages any concerns about the modest temporal distance.
14
We weight the percentiles by full-time-equivalent undergraduates. In each reference year, our sample includes all
colleges that identify as offering four-year degrees in IPEDS that year, have at least one first-degree-seeking
undergraduate enrolled, and report at least two of our quality proxies. We exclude specialty institutions such as
nursing colleges and theological seminaries (Carnegie codes 51-60).
9
our index as an estimate of latent college quality, which we view as continuous and one-
dimensional. Combining multiple proxies into a single index measures latent quality with less error
than a single proxy. Our index reveals remarkable stability in college quality between our two
cohorts; weighted by full-time undergraduates the correlation equals 0.86.15 Our measure does not
capture differences in the quality that different students experience within the same university due
to e.g. quality differences across fields of study or participation in honors programs. Our index
also speaks only indirectly to absolute differences in college quality. In practice, the four individual
quality measures underlying our index increase modestly but steadily with the index for the bottom
90% of four-year colleges but more steeply for the top 10% of colleges. Figure 1 documents this
pattern for expenditures per student, a measure we do not include in our index but which strongly
correlates with its components. This very general scaling issue with latent indices, emphasized in
this literature by Bastedo and Flaster (2014), also applies to the two other most common proxies
for college quality in the literature: the mean SAT score of the entering class and the Barron’s
selectivity categories.
We analyze the quality of the first four-year college a student attends rather than the last,
as in Black et al. (2005) and some other studies. Our concern with academic match motivates this
choice, as treating the quality of the first college as the choice variable allows us to treat subsequent
transfer and completion choices, some of which may result from mismatch, as intermediate
outcomes on the way to earnings effects.
3.4 Sorting among colleges by student ability
To assess the degree of sorting across colleges by student ability we consider the joint distributions
of the student ability and college quality measures just described. As we weight the quality
percentile by student body size, a college in the nth quality percentile is the college that a student
in the nth ability percentile would attend under perfect assortative academic matching. 16 We define
academic mismatch as substantial deviations from this type of matching. One appealing feature of
our measure is the possibility of achieving perfect assortative matching without violating
institutional enrollment constraints. The academic match measures employed in other important
15
Our estimates also exhibit face validity; choosing one corner of one state at random, for the NLSY-97 cohort the
University of Michigan gets a 93, Michigan State a 74, Wayne State a 36 and Eastern Michigan a 28.
16
Our measure of student body size is full-time equivalent undergraduates.
10
studies in the literature, such as Roderick et al. (2008), Bowen et al. (2009), and Smith et al. (2013)
lack this feature.17
Table 1 gives the joint distributions of student ability and college quality for the 1979 and
1997 cohorts, with both variables discretized into quartiles. In both cohorts, students differentially
concentrate along the diagonal, which corresponds to academic match. The four diagonal cells
account for 34.3% of students in 1979 and 37.1% in 1997, rather than the 25% implied by random
sorting. The three upper right cells, corresponding to low ability students at high quality colleges,
account for 11.3% of students in 1979 and 10.2% in 1997, while the three lower left cells,
corresponding to high ability students at low quality colleges, account for 14.9% in 1979 and
13.2% in 1997.18 Thus, we find quite substantial departures from academic match in both cohorts.
Viewed longitudinally, our data (perhaps surprisingly given the recent policy focus on match)
reveal only a small, though meaningful, increase in assortative matching.19
3.5. Outcomes
We examine five educational outcomes: graduation within four or six years of starting, obtaining
a STEM degree, and transfer to a higher or lower quality college. The NLSY-79 survey did not
begin asking questions on the specific college attended until even the younger sample respondents
were several years into college, making it difficult to follow transfer behavior in the earlier
cohort.20 We therefore calculate the transfer outcomes only for the NLSY-97 cohort. We define
graduation as completing a four-year degree at any college. We define STEM degree completion
based on the last reported major(s) prior to graduation and code majors as STEM or non-STEM
using the (uncontroversial) system in Arcidiacono et al. (2016).21 Some restless students transfer
more than once; we code our transfer variable based on the first observed transfer and only count
17
Our measure also differs from that in Chetty et al. (2017). Rodriguez (2015) and House (2017) analyze various
measures of mismatch and find that the measure matters for the amount of mismatch measured.
18
Black and Smith (2004, Table 4) and Light and Strayer (2000, Table 3) present alternative estimates of the joint
distribution for the NLSY-79 cohort that tell the same basic story. See also Mattern et al. (2010, Table 1).
19
Statistically, a chi-squared test rejects the null of a common joint distribution in the two cohorts. Substantively, our
results comport with Hoxby (2009, Figure 1) and Chapter 1 of Herrnstein and Murray (1994), which show that much,
but not all, of the large increase in stratification by ability among colleges had played out by the time the NLSY-79
cohort entered college. Smith et al. (2013) reach a quite different conclusion from ours, arguing that undermatch
decreased dramatically between the two cohorts they consider. The different timing of their two cohorts, their (quite)
different definition of mismatch, and their inclusion of two-year schools stymie a detailed accounting of the differences
between their finding and ours.
20
Light and Strayer (2004) document the nature and extent of transfers in the NLSY-79 and consider their relationship
to wages. In keeping with the limited nature of the available data, they distinguish between two-year and four-year
colleges but not more finely by quality.
21
See Table A-1 of their on-line appendix.
11
transfers that change college quality by at least five percentiles (up or down) on our college quality
index. Transfers from any four-year college to any two-year college always count as a transfer
down.
On the labor market side, in the spirit of the program evaluation literature we examine the
level of real (2010$) earnings (rather than the log) in all years from the start of college, without
conditioning on degree completion.22 We look relative to the start of college rather than the end
because we want to capture the opportunity cost of college, because academic match may affect
the probability of working while in school, and because we want to capture effects on time to
degree. We average earnings in two-year intervals, using observed earnings for one year when the
value for the other equals zero or missing. We omit two-year intervals if the respondent did not
report non-zero earnings in either year. This reduces variance at minimal cost to sample size and
temporal fineness, as nearly everyone in our sample of four-year college attendees works in almost
every year. Furthermore, comparisons with the information on job spells suggest that a non-trivial
fraction of the zeros represent measurement error.
Table 2 summarizes these outcomes for our sample. The 1997 cohort has a higher
graduation rate, consistent with the pattern documented in Archibald, Feldman, and McHenry
(2015) that U.S. graduation rates reached a nadir for students starting college in the mid-1980s and
have recovered since then. The probability of graduating with a STEM degree has fallen a bit,
from 15% to 13%. Consistent with the somewhat earlier cohort studied by Goldrick-Rab (2006)
and with the Texans in Andrews et al. (2014), we find a great deal of transfer behavior among the
NLSY-97 students; 27% transfer at least once. Earnings increase both over the life cycle within
cohorts and between the 1979 and 1997 cohorts.
4. Econometric framework
To determine whether the data provide evidence of important interactions between ability and
college quality, we want to look flexibly at the (conditional) relationship between these two
variables and the outcomes of interest. Several econometric frameworks comport with this goal.
This section describes two: our preferred estimator based on a flexible polynomial approximation
22
The NLSY datasets offer two different earnings measures: a CPS-like measure based on a question about total
earnings the previous year and a constructed variable that builds on information about wages, hours, and weeks on
individual job spells. We use the CPS-like measure for both cohorts.
12
and an alternative estimator that uses indicators for bins of the discretized joint distribution of
ability and college quality.
For binary outcomes, we estimate probit models. In our preferred specification, we estimate
the conditional probability function as:
(1) Pr(Yi = 1| Ai , Qi , X i ) = Φ ( β 0 + β p ( Ai , Qi ) + β X X i )
In (1), Y denotes the binary outcome of interest, A denotes student ability, Q denotes college quality,
β p ( Ai , Qi ) denotes a flexible polynomial of ability and quality, and X denotes a vector of other
conditioning variables. For earnings, we estimate a parametric linear regression model using the
same specification by ordinary least squares. In both cases,
(2) β P ( Ai , Qi ) = β1 Ai + β 2 Ai2 + β 3 Qi + β 4 Qi2 + β 5 Ai Qi + β 6 Ai2 Qi + β 7 Ai Qi2
We chose this specification after a fairly rigorous round of statistical testing.23 The polynomial in
ability and quality becomes non-parametric once we promise to include higher-order terms (but
not too quickly!) on those happy occasions when our sample size increases. Model (1) then
becomes a partially linear model in which we non-parametrically estimate the effects of ability
and quality while conditioning parametrically on the other variables.
Polynomial approximations sometimes mislead, especially around the edges of the data.
As a sensitivity check, we implement a different semi-parametric framework that includes
indicators for combinations of college quality quartile and student ability quartile. We include
indicators for 15 of the 16 possible combinations, with ability and quality both in the lowest
quartile serving as the omitted category. This approach avoids the oft-observed instability of higher
order polynomials away from the center of the data but cannot capture any within-quartile
mismatch. In practice, the two estimators tell the same substantive story; see Appendix Table A7
for the estimates from the second approach for a subset of our outcomes.
The NLSY surveys include several measures of respondents’ cognitive skills. For ease of
interpretation we interact only ASVAB1 with college quality. We therefore want to concentrate
23
We conducted a series of specification tests, with and without additional covariates, starting with higher-order terms
in ability, quality, and their interactions and gradually moving towards more parsimonious specifications. For several
outcomes, these tests do not reject the exclusion of all ability-quality interaction terms. We include the most
parsimonious specification that still allows for non-linear interaction effects between ability and quality and report
tests of the joint significance of these interaction terms in our results. Our thorough search leads us to think that other
paths to functional form flexibility, such as substituting the log of earnings for the level or considering the levels of
our ability and quality measures rather than their percentiles, would do little to alter our qualitative findings.
13
the effects of any common component of ability in this variable. To accomplish this, we
orthogonalize the SAT score and GPA variables against ASVAB1 prior to including them in the
multivariate analyses.24
5. Identification
This section considers the case for interpreting our estimates as causal. We argue that we have a
sufficiently rich conditioning set that the remaining variation in college quality that serves to
identify our effects is uncorrelated with the error term in the outcome equation. To accomplish
this, we need two things. First, we need the observed covariates included in our model to capture,
either directly or as proxies, all the factors that affect both the college quality choice and the
outcomes we study. Second, in order to avoid identification via functional form, we need some
(conditionally) exogenous variation in college quality choices. Put differently, we need
instrumental variables to exist, even though we do not observe them, as they produce the
(conditional) variation in college quality we implicitly use in our estimation.
We divide our conditioning variables into four sets, each of which proxy for one broad
factor affecting educational choices: pre-college skill, student demographic and family
characteristics, neighborhood characteristics as of the first survey, and other social factors. We list
these variables and describe their construction in detail in Appendix Table A2. Our preferred
specification includes the first three sets of covariates; the fourth set provides a test of sorts,
described below, for our identification strategy. We never condition on whether the student
remains enrolled in college each year or whether they have completed a degree, which we view as
intermediate outcomes.
We make the case that our conditioning set suffices to solve the problem of non-random
selection into colleges in two ways. First, we can think about whether our conditioning set contains
those things (or compelling proxies for those things) that existing theory and empirical evidence
deem important. Much recent literature, e.g. Heckman and Kautz (2012), emphasizes the
importance of non-cognitive skills for educational and labor market outcomes. The broader
literature, including our own earlier study, illustrates the need to condition on family resources,
24
We experimented with two other ways of using the SAT and GPA variables: one set of analyses simply omitted
them while the other combined them with the ASVAB components to create a “super” ability index. Neither strategy
affects our qualitative conclusions. We do not use the super index as our primary ability measure because of the
large number of observations with missing information on SAT and/or GPA.
14
both intellectual and financial. More money makes many things about college easier, including
longer time-to-degree, more frequent visits home, and not having to work during school, and so
affects outcomes; it also surely affects the college quality choice. Parental education will correlate
with their knowledge of the college choice process and of how to succeed at college in both the
institutional and academic senses. Parental education also likely correlates with taste for education
and otherwise unobserved features of the student’s childhood environment that affect both
outcomes and college choice. Becker and Lewis (1973) highlight a quality-quantity tradeoff for
parents, so number of siblings may reflect both resources and preferences. We expect that our
county education variable will both help with measurement error in the direct parental resource
variables and proxy for primary and secondary school quality as well as peer pressure and
expectations.
The second way to think about our covariate set asks whether the marginal covariates make
any difference to the estimates. In Heckman and Navarro’s (2004) framework there exist multiple
unobserved factors on which we need to condition. As we increase the number of proxy variables
in our conditioning set, the amount of selection bias in our estimates should decrease to zero, so
long as we keep adding proxies for all factors. Turning this around, if we observe that the estimates
stabilize as we increase the richness of the conditioning set, this suggests we are doing a good job
of proxying for the unobserved factors, unless there exists an additional unobserved factor
uncorrelated with all of our covariates. Oster (2017) cautions that a finding of coefficient stability
means little if the newly added variables do not capture any (conditional) variation in the dependent
variable. We perform such analyses by adding sets of related variables to the conditioning set in
an order that roughly reflects our prior about their importance for solving the problem of non-
random selection into colleges of different qualities.
The literature suggests that plenty of exogenous variation exists in college quality choices
conditional on our observed covariates. First, differences in state college quality mix, admission
policies, and pricing strategies provide plausibly exogenous variation in the budget sets facing
students and their parents. Tienda and Niu (2010) and Daugherty et al. (2014) estimate large effects
of guaranteed admission to in-state public colleges through the Texas Top 10% rule on college
choices for eligible students. Cohodes and Goodman (2014) estimate similarly large effects of in-
state-specific scholarships in Massachusetts, lowering the average quality of college attended for
eligible students who chose in-state colleges over more selective outside options. Second,
15
distances to colleges of various qualities provide variation in the costs of attendance as in Card
(1995) and Currie and Moretti (2003). Third, what normally represents a sad feature of this
literature, namely the consistent finding that many students, parents, and high school guidance
counselors have little or no idea about how to choose a college, provides aid and comfort for our
identification strategy. Hoxby and Avery (2012) and Hoxby and Turner (2013) show the difference
a small amount of reliable information can make for many students. Similarly, the literature
provides many examples of small behavioral economics tricks having non-trivial effects on college
choices. Pallais (2015) finds that you can change college choices by changing the number of
colleges to which students can send their ACT scores for free and Bettinger et al. (2012) find that
having H & R Block help with the federal financial aid form can have real effects on college-going.
Scott-Clayton (2012) reviews the literature showing that students and parents often know very
little about the likely costs and benefits of college. Finally, both the descriptive and ethnographic
literature, such as Roderick et al. (2008), and the quantitative literature on sorting, such as Lincove
and Cortes (2016), suggest that many students choose among colleges for reasons unrelated to
academic match such as the football team or the presence of high school friends. While the value
students place on these non-academic traits may well unconditionally correlate with outcomes, we
expect that variation among students in the nature and extent of the trade-off between academic
and non-academic aspects of colleges that they face to produce useful, and random, conditional
variation in college quality choice.
What can we say about any remaining selection bias? Putting aside match for the moment,
two worries usually arise about the college quality main effect. First, students, their parents, and
college admissions officers may have access to information on student ability that we, the
researchers, do not. To the extent that those unobserved bits affect admissions, we would expect
an upward bias in the estimated effect of college quality because it proxies in part for higher
unobserved student ability or ambition (and we might expect this bias primarily at the upper end
of the college quality distribution, where “holistic” rather than rule-based admissions dominate).
Second, we might worry about measurement error in college quality, as in Black and Smith (2006).
Though our use of a quality index based on multiple proxies addresses this issue some
measurement error surely remains, which we expect will push the estimated effect toward zero. Of
course, we have no basis for arguing that these two biases cancel out in practice.
16
Now think about the interaction of college quality and student ability. If we overstate the
effect of a high quality college for all students, then overmatched students will look better than
they should relative to other students of the same ability. Similarly, undermatched students will
look relatively worse than they should. Thus, upward bias in the estimated effect of college quality
should lead us to understate the effects of overmatch and to overstate the effects of undermatch,
and so potentially to overstate the degree of complementarity between student ability and college
quality. Measurement error in ability and/or in college quality, in contrast, should attenuate our
estimates of the effects of both overmatch and undermatch; indeed, Griliches and Ringstad (1970)
highlight the particularly pernicious effects of measurement error in non-linear contexts such as
interactions.
17
important academic match effects, the effect of quality should vary with student ability. For
example, college quality might increase degree completion probabilities more for students lower
in the ability distribution. Our second key finding is that the effect of college quality on graduation
varies very little with student ability. Likewise, the effect of student ability is quite steady at
different points in the college quality distribution. Figure 2 plots the average predicted six-year
graduation probability at each percentile of college quality. It shows that at the 25th, 50th and 75th
percentiles of the ability distribution the probability of graduating within six years increases almost
linearly in college quality percentile for the NLSY-79 cohort. The NLSY-97 cohort shows some
evidence of a weak overmatching effect: the probability of graduating within six years increases
more slowly with college quality above the 60th percentile of colleges for students in the 25th
percentile of ability.
We can quantify the evidence for mismatch in our college completion results in two ways.
First, because our model nests a model with only main effects of college quality and ability, we
can test the restriction that all coefficients on the interactions of ability and college quality jointly
equal zero. The p-values from these tests appear in the third row in the bottom panel of Table 3.
The p-values of 0.554 for the NLSY-97 cohort and 0.363 for the NLSY-79 cohort indicate that the
restrictions implicit in the main-effects-only model cause little trouble for the data. Alternative
statistical tests for the absence of mismatch effects consider the null of equal average derivatives
with respect to student ability and with respect to college quality, at the 25th, 50th and 75th
percentiles of each. P-values for these tests appear in the last two rows of Table 3; they comport
with our interpretation based on the magnitudes above.
Second, we can look to Tables 4A and 4B, which compare the observed completion rate
with the completion rate implied by our model in a counterfactual world of perfect matching. We
obtain this value by predicting degree completion for every observation with their college quality
percentile recoded to match their ability percentile. Based on our model, we find that degree
completion rises less than one percentage point if we eliminate academic mismatch, moving from
59.9% to 60.4% for the younger cohort and from 49.7% to 50.3% for the older one. The negative
effect of moving lower ability students away from high quality colleges to their matched quality
level almost entirely cancels out the positive effect of moving higher ability students out of low
quality colleges. Chingos (2012) performs a similar calculation and also finds virtually no effect
of resorting students.
18
This net effect of eliminating academic mismatch masks large improvements in outcomes
from moving some students to higher-quality colleges. The last columns of Tables 4A and 4B
present a second counterfactual in which we ignore capacity constraints (and general equilibrium
considerations) and assume that all students attend a college in the 90th percentile of college quality.
Our model predicts that moving all students to a high-quality college would increase degree
attainment within six years by 9.1 percentage points to 58.8% for the older cohort and by 10.6
percentage points to 70.5% for the younger cohort. This increase might seem smaller than expected,
but student characteristics matter as well, and differ strongly between students presently at the 90th
percentile of college quality and those further down the distribution.
Now consider the results for graduation in four years, a standard that remains normative
but has become increasingly aspirational for many students, as documented in e.g. Bound et al.
(2010). The average derivative estimates for completion in four years resemble those for
completion in six years in sign, all positive, but differ in showing a clear and substantively
meaningful pattern of complementarity between student ability and college quality, one that gets
stronger for the NLSY-97 cohort. We can reject the nulls of equal mean derivatives for the 1997
cohort. In parallel (and substantively similar) results from linear probability models presented in
Appendix Table A8, we strongly reject the null of zero coefficients on the interaction terms and
the nulls of equal average derivatives with respect to ability and college quality for the 1997 cohort
(and come closer to rejection of these nulls in the 1979 cohort). Overall, we find serious evidence
of complementarities between ability and quality for on-time degree completion. In addition, the
difference between the two cohorts in the relative importance of student ability and college quality,
with college quality playing a smaller role for the older cohort, remains striking.
19
quality, with only modest differences between the effects at the 25th, 50th, and 75th percentiles. We
find substantively small effects, not statistically different from zero, of college quality at all levels
of student ability. The p-values for the null of zero interaction equal 0.791 and 0.708 for the NLSY-
97 and NLSY-79 cohorts. Our estimates predict that resorting to perfect academic match would
change STEM degree completion by less than 0.6 percentage points in either cohort.
Students may react to learning they have made a poor initial match by transferring to
another school. Again, this mid-course adjustment could delay graduation beyond four years. We
find evidence consistent with match effects when looking at transfer behavior in the NLSY-97
cohort. The third column of estimates in Table 5 corresponds to model (1) with transfer up as the
dependent variable, while the fourth column of estimates corresponds to transfer down. 25
Increasing a student’s ability percentile by 10 percentage points raises the probability that she will
transfer to a higher quality college by 1.4 percentage points if she starts at a 25th percentile college.
In contrast, student ability has virtually no effect on the probability of transferring to a higher
quality college if the student starts at a 75th percentile college. The second three rows show an
expected pattern: increasing the quality of the first college a student attends lowers the probability
that she will transfer to an even higher quality college, with a larger effect for students higher in
the ability distribution. The pattern of derivatives with respect to ability reflects students
preferentially transferring to better matches, while the pattern of derivatives with respect to quality
is partly mechanical.
We see the reverse patterns when considering transfers to lower quality colleges. More able
students transfer to a lower quality college less often, though only modestly and imprecisely so.
Increasing the quality of the first college attended raises the probability that students will transfer
down; these effects differ statistically from zero but not much by the level of student ability. Taken
together, these transfer results provide some support for the mismatch hypothesis along with a
strong (and again partly mechanical) main effect of college quality.
We cannot reject the null of only ability and quality main effects on transfer behavior: As
shown at the bottom of Table 5, the p-values equal 0.384 and 0.328 for transferring to a higher and
lower quality college, respectively. In Table 4B, we predict that eliminating initial mismatch would
25
Replacing the five percentile-point cutoff for a transfer up or down with a zero cutoff or a 10 percentile-point cutoff
yields qualitatively similar findings.
20
modestly decrease transfers to a higher quality college and modestly increase transfers to lower
quality colleges. Eliminating initial mismatch substantially decreases the transfer probability for
students who presently severely mismatch in their initial college choice but such students
constitute only a small fraction of the total. Since transfers often delay graduation, these moves
have real costs for students (and often for the taxpayer) in terms of more time in school and less
time in the labor force.
6.3 Earnings
Tables 6A and 6B present our estimates for the effects of ability and college quality on average
annual earnings during and after college. In years 2-3, both college quality and student ability have
generally negative effects on annual average earnings in both cohorts. In the NLSY-79, 2-3 years
after starting college a student at the 50th percentile of ability earns $208 less per year for each 10
percentile-point increase in the quality of first college attended. Students at less selective colleges
are more likely to have dropped out of college without a degree and begun working full time 2-3
years after starting college. Higher quality colleges may also require greater effort to keep up with
course work, limiting the time students have to work while still in college. Finally, near the top of
the college quality distribution, marginal increases in college quality may give students access to
more financial aid and reduce their need to work during college. The negative relationship between
ability and earnings likely reflects a similar short-run tradeoff between current earnings and
investment in skill accumulation, as well as access to more merit-based financial aid.
At 10-11 years after students begin college these patterns have completely reversed: both
college quality and student ability strongly raise average annual earnings. For a student of median
ability in the NLSY-97 cohort, each 10 percentile-point increase in the quality of the first college
is associated with an additional $1,480 of annual earnings. In keeping with the completion rate
estimates, we find much larger ability effects in the NLSY-79 cohort and smaller college quality
effects. For example, at the median of college quality, a 10 percentile point increase in student
ability increases earnings at 10-11 years by $915 in the NLSY-79, compared to just $417 for the
later cohort. While our average derivative estimates carry with them frustratingly large standard
errors, we find strong evidence that college quality increases future earnings throughout the ability
distribution.26
26
Our estimates for 6-7 years after college start (not shown) lie in between the 2-3 and 10-11 year estimates.
21
As with degree completion in four years, the estimates for earnings 10-11 years after
college start suggest a substantively important complementarity between college quality and
student ability, particularly for the NLSY-97 cohort. The average derivative of earnings with
respect to student ability has a much larger value, around $127 per percentile point for the NLSY-
97, for students at the 75th percentile of college quality than for those at the 25th percentile or at
the median. Similarly, the average derivative of earnings with respect to college quality increases
with student ability, from about $104 per percentile point at the 25th percentile of ability to about
$186 per percentile point at the 75th percentile of ability. Still, we cannot reject the nulls of no
interaction effects (as well as other nulls involving much larger interaction effects) or of equal
derivatives.
Table 4 shows that re-sorting the students in our data to eliminate mismatch in the NLSY-
97 cohort would increase mean earnings by about $1,328 10-11 years after beginning college. The
corresponding change for the NLSY-79 cohort is $694. While we do not put much stock in the
particular number given how far this scenario projects outside the data, and given the likely
importance of equilibrium effects of uncertain direction and magnitude (including the fact that
resorting the students would change the quality of all of the colleges as we measure it), the data
clearly do not shout out harmful effects of college quality, on average, for low ability students. At
the same time, our estimates suggest that policies that place some students with lower ability at
top colleges do impose some efficiency costs due to the complementarity between student ability
and college quality.
The NLSY-79 cohort, now into its fifties, allows us to examine earnings outcomes for
several decades after college start. These results appear in the last columns of Table 6A and in
Figure 3. The data provide large, positive, and generally statistically significant estimates of the
effect of student ability and college quality at all durations from 10-31 years after college start.
Even at long durations we can clearly rule out negative average effects of overmatch. We can also
rule out simple versions of the “college quality is just a signal” argument, as employer learning
would surely have overwhelmed college quality effects over the horizons we consider.27
In general, the average derivatives get larger as the time elapsed from college start increases,
sometimes quite substantially so. The average derivative with respect to quality for students at the
75th percentile of ability increases from $724 for a 10 percentile point increase in quality at 10-11
27
See Hershbein (2013) for subtler signaling theories of college quality.
22
years to $2,527 at 20-21 years to $4,048 at 30-31 years. As our data embody only one cohort, we
have no way of separating these increases into components due to age and period effects.
Additionally, a pattern consistent with complementarity between student ability and college
quality, which was fairly weak in the estimates of earnings 10-11 years after college start for the
NLSY-79 cohort, appears quite strongly in the longer-term follow-up estimates. As shown in the
bottom row of the table, for earnings 20-21 and 30-31 years out as we can reject the null of equal
average derivatives with respect to college quality at different levels of student ability. Finally, we
remind the reader of our imprecise estimates, and of the gentle decline in the sample size as
individuals gradually attrit from the panel.28
6.4 Subgroups
We consider subgroups defined by sex and by parental education, where we partition the latter into
“low” and “high” subgroups based whether or not at least one parent attended college. We think
of parental education as a proxy for several things, including tastes for college (and college quality)
as well as family resources. In the NLSY-79 cohort nearly half of college entrants have parents
with no more than a high school education, but by the NLSY-97 cohort only a quarter of college
entrants have parents with no college education. We lack the sample size to usefully examine finer
categories. Similarly, though of great substantive interest, we lack the sample sizes to present
meaningful subgroup estimates for African-Americans and Hispanics.
Tables 7 and 8 report the effects of student ability and college quality on earnings 10-11
years after starting college separately by subgroups.29 To limit the demands on our relatively small
sample, we estimate these effects by interacting the ability-quality polynomials with subgroup
indicators and continuing to estimate pooled coefficients on the other covariates. The main finding
from the pooled estimates, positive effects of both student ability and college quality at all levels,
generally holds for all sub-groups, with more volatility in point estimates, and predictably larger
standard errors. In both cohorts, the pattern of complementarity between student ability and college
quality is most apparent for the children of more educated parents, though the average effect of
student ability is larger for the children of less educated parents. We mostly find larger effects of
ability for female students than for male students, much larger in the NLSY-79 cohort. However,
28
All of the qualitative findings in Table 6A related to long-term earnings impacts for the NLSY-79 cohort persist if
we restrict ourselves to a balanced panel.
29
Completion rate estimates by subgroup tell broadly the same story as the pooled completion estimates, with more
volatility in individual point estimates due to smaller sample sizes.
23
the two main differences between the two cohorts hold for both men and women: student ability
plays a relatively larger role in determining earnings in the earlier NLSY-79 cohort and the
younger cohort displays more evidence of complementarity between ability and college quality in
degree completion (the latter in results not shown).
6.5 Identification
As promised in Section 5, we now consider some evidence regarding our identification strategy.
Tables 9 and 10 present estimates based on increasingly rich sets of conditioning variables for our
two most important outcomes: degree attainment within six years and earnings in years 10-11 after
starting college. The lower rows of each table indicate the set of included conditioning variables;
the categories correspond to those in Appendix Table A2. The estimates in column (4) of each
table correspond to those in Tables 3 and 6.
Overall, the tables reveal a substantial amount of movement in the coefficients when
moving from column (1) to column (2) by adding additional measures of ability and socio-
emotional skills, and when moving from column (2) to column (3), which corresponds to adding
demographics and family characteristics. 30 We see somewhat less movement (how much less
varies across outcomes and across derivatives) when we add neighborhood characteristics in
column (4). Finally, and in parallel to the similar analysis in the Black et al. (2005) study of college
quality, moving from column (4) to column (5) changes the estimates very little. With each
transition, including the last, the r-squared values meaningfully increase. These findings support a
causal interpretation of our estimates or, at least, suggest that any remaining biases would not
overturn our qualitative conclusions.
6.6 Comparing the NLSY-79 and NLSY-97 results
Our big picture stories apply to both cohorts: large amounts of overmatch and undermatch in the
unconditional joint distribution of student ability and college quality, substantively and statistically
significant positive main effects of student ability and college quality for college completion and
earnings in the medium and long terms, some evidence of match effects in time to degree, and no
substantive or statistical evidence of match effects for degree completion or for STEM degree
completion. This stability surprised us somewhat. In this section, we briefly remark on two specific
30
The estimated effects of ability and college quality on graduation rates in the NLSY-79 cohort display less
sensitivity to the conditioning set than the other outcomes we consider. We lack a good explanation for this pattern.
24
differences in the results: (1) the modest but not trivial reduction in mismatch between cohorts;
and (2) the relatively smaller role of college quality in determining outcomes for the NLSY-79.
As we noted in Section 3.4, the small decrease in mismatch between the cohorts comports
with some other evidence in the literature. Following Hoxby (2009), we suspect that it results from
ongoing reductions in the cost that students (especially high ability students) face in obtaining
information about admissions criteria, real (as opposed to posted) prices, and optimal strategy.
Reductions in transportation and communication costs likely also play a role.
One reason why college quality may have smaller measured effects on outcomes in the
NLSY-79 cohort is that colleges varied less in our input-based measures of quality for this earlier
cohort. As shown in Figure 1, each percentile increase in our college quality index corresponds to
a somewhat smaller change in expenditure per student in 1992 than in 2008. We may also suffer
from attenuation bias due to measurement error in matching students to their first college attended.
As noted in Section 3.3, the NLSY-79 did not ask students the name of the college(s) they attended
until 1984, part way through most students’ college enrollment. We use the first reported college,
which we suspect functions as an excellent proxy for first college attended, but memory lapses
combined with the occasional transfer representing a large change in college quality could yield
some mis-measurement.
6.7 Comparisons with other studies
Two earlier published papers, Light and Strayer (2000) [hereinafter LS] and Black, Daniel and
Smith (2005) [hereinafter BDS] estimate college quality effects interacted with ability using the
NLSY-79 data. The estimates from the LS probit model of degree attainment (they do not look at
earnings) that appear in the two right-most columns of their Table 8 correspond most closely to
our own.31 These estimates assume, implicitly, selection on observed variables; i.e. they frame
them as a sensitivity analysis in which they shut down their apparatus for dealing with selection
on unobserved variables. Unlike us, they find that ability does not always increase degree
attainment across all college quality quartiles, nor does college quality monotonically increase
degree completion across all ability quartiles.32 In the latter case their estimates comport with the
31
The corresponding completion probabilities, which we find easier to interpret, appear in their Table 12. Because the
model underlying their Table 12 assumes independent errors, the distribution of unobserved variables does not depend
on the choice of college and college quality in the first period. Thus, we interpret the three rows for the “overall sample”
for each ability quartile as three independent simulations of the same parameter values.
32
Though substantively different, our estimates do not quite differ statistically from theirs. For example, in their Table
12 students in AFQT quartile 1, roughly our ASVAB quartile 1, suffer a reduction in college completion probabilities
25
predictions of standard theories of academic mismatch. Several differences between the LS setup
and our own strike us as potential candidates to account for the difference in findings: (1) they
treat transfers as dropouts; (2) they restrict some of the interactions between ability quartile and
college quality quartile to have zero coefficients on a priori grounds; (3) they condition on
variables that we think plausibly endogenous, namely living at home, receipt of financial aid, and
actual tuition paid; and (4) their remaining covariate set represents (in essence) a modest subset of
our own, which raises the possibility of residual selection bias not present in our analysis.33
The analysis in BDS, not surprisingly given the authorial overlap, differs less drastically
from our own. Qualitatively, we reach similar conclusions. While they examine mismatch only for
their log wage outcome, and not for degree completion, they find strong main effects of both
student ability and college quality for both degree attainment and log wages. Their Appendix Table
7 presents estimates from a parametric linear model with hourly wages as the dependent variable
and a rich covariate set similar to our own (other than in its inclusion of years of schooling), along
with main effects in ability and college quality and interactions between college quality and their
versions of ASVAB1 and ASVAB2. They offer separate estimates for men and women; for both
groups and both interactions they obtain estimated coefficients near zero and far from statistical
significance.34
As far as we know, just three other studies consider the persistence of the earnings effects
of college quality over a very long interval after college start; none of these studies consider match
effects. Turner (2002) considers earnings of men in the Panel Study of Income Dynamics (PSID)
who completed a BA by 1975. She finds large and growing effects of college quality from 1975
to 1992 and provides suggestive evidence that the increases primarily represent period effects
rather than lifecycle effects. The average SAT score of entering students proxies for college quality.
of around 0.07 from moving from their first to second quartiles of college quality. The corresponding estimate in our
alternative specification using quartile indicators presented in Appendix Table A7 is an increase of 0.025 with a
standard error of 0.049. The comparison is complicated by the fact that they do not present standard errors on their
predictions and the fact that the two estimates, which rely on the same data, presumably have a non-zero covariance.
33
Other differences seem to us as a priori less likely to account for a large portion of the difference, e.g. (1) LS measure
college quality differently than we do; (2) LS measure student ability a bit differently than we do, relying on the
Armed Forces Qualifying Test (AFQT) score, a weighted average of four ASVAB component scores; and (3) their
sample differs somewhat from ours, as indicated by their sample size of 2,754 compared to ours of 2,441.
34
We also used our data to replicate their specification and then marched, one change at a time, from their setup to
our setup. When we did what they did, we got estimates that look very much like what they got. Key differences result
from using the first college rather than the last college attended, which reduces the estimated effect of college quality
somewhat, and from including the county conditioning variables, which also reduce the estimated effect of college
quality.
26
As discussed in her note 13, the lack of a compelling proxy for student ability in the PSID hampers
a causal interpretation of her estimates, which rely on a “selection on observed variables”
identification strategy. Figure 1 of Black et al. (2005) shows impacts on log wages for the NLSY-
79 for calendar years 1987 to 1998, about 15 years after starting college. They find persistent,
stable, and substantively and statistically meaningful effects of college quality for both men and
women using a selection on observed variables identification strategy (and conditioning on years
of schooling).
Dale and Krueger (2014) link social security earnings data to two cohorts of the “College
and Beyond” dataset that includes students entering a non-random sample of relatively high quality
colleges in the fall of 1976 and 1989. They present estimates of log earnings impacts through 2007,
or 31 years after college enrollment for the older cohort and 18 years after for the younger one.35
Dale and Krueger (2014) present estimates using both a “selection on observed variables”
identification strategy that includes test scores and a strategy that attempts to deal with any
remaining selection on unobserved variables by conditioning on the average SAT score of the
schools to which each student applied, which they call their “self-revelation” model. 36 The first
identification strategy finds persistent and sizeable effects of college quality on later earnings for
all groups; in marked contrast, the “self-revelation model” estimates reveal such impacts only for
black and Hispanic students and those from disadvantaged family backgrounds.
Finally, in a broad sense our results coincide with the descriptive analysis presented in
Chetty et al. (2017), who find using US income tax data linked across generations that within
college quality tiers, average child income varies only very modestly with parental income. This
pattern conflicts with strong negative effects of overmatch as students with relatively low income
parents within a quality tier will also have relatively lower average ability as we define it.
35
The earnings measure is actually the median value of log annual earnings over five-year intervals, excluding
individuals with low enough values to suggest only marginal labor market attachment.
36
This identification strategy has issues of its own; see e.g. Hoxby (2009) for discussion.
27
comparability in coding and conditioning. In both cohorts, we find strong evidence that college
quality and student ability increase the probability of degree completion and later earnings. The
relative importance of college quality increases in the later cohort; this finding parallels that of
Castex and Dechter (2014), who document a similar change in the relative importance of ability
and years of schooling as determinants of wages in the two NLSY cohorts.
For college students and their families, our most salient conclusion is that increasing
college quality increases graduation rates and earnings at all points in the ability distribution.37 At
the margin, all students will benefit in expectation from attending colleges with more resources
and stronger peers. In Dillon and Smith (2017) we find that well-informed and well-resourced
students seek to attend higher quality colleges, even if they will be “overmatched” at these
institutions. Our current work validates this unconditional pursuit of college quality. Policies
targeted at increasing the representation of certain groups of students at high-quality colleges will
(on average) benefit the targeted students, but may do little to improve overall outcomes if the total
number of seats at high-quality colleges remains unchanged.
The simple and compelling applied theory models in Rothschild and White (1995) and
Sallee, Resch and Courant (2008) posit strong complementarities between student ability and
college quality, which can justify the observed long-term increase in college match described in
e.g. Hoxby (2009). We find modest but substantively important support for these theories. We can
reject uniform effects of college quality across the ability distribution for some but not all of the
outcomes we examine. The effects of college quality vary with student ability in the production of
graduation in four years (but not six), particularly for the NLSY-97 cohort, for transfers, and for
long-term (but not immediately post-college) earnings. The interaction effects we find do not
overwhelm the uniformly positive main effects. These results suggest modest efficiency gains from
better sorting the strongest students into the top colleges, and some efficiency costs to policies that
weaken this sorting.
Less prepared students appear to adjust to the demands of more intense colleges by slowing
their studies, leading to smaller gains in four-year graduation rates than their higher-ability
classmates but similar gains in six-year graduation rates. In contrast to some other papers, we do
not find similar evidence of adjustment by shifting out of STEM majors. One interpretation of our
37
With the possible statistically insignificant and empirically irrelevant exception of students in the 25th percentile
of ability attending colleges in the top 10% of the quality distribution.
28
findings on earnings is that the networking and recruiting benefits of attending a higher-quality
college benefit all students in their first job, but higher-ability students build more successfully on
these early gains as they move through their careers, perhaps because of greater skill acquisition
in college.
We conclude with five caveats. First, we interpret our estimates in partial rather than
general equilibrium terms; as such, they apply primarily to moving around small numbers of
students. Second, we pay for the plausibility of the conditional independence assumption with
modest sample sizes. Particularly in the context of high variance outcomes such as earnings, some
of the patterns we find show up more clearly in the estimates than in the statistical tests. Third,
measurement error remains a concern in multiple senses. While using multiple proxies for student
ability and college quality reduces measurement error, it does not eliminate it. Less trivially, we
know that individual students at larger colleges experience very different parts of what their
institutions have to offer; for example, faculty research and teaching quality may differ across
departments. Thus, even if our quality measure does well at capturing the average quality of a
college, it may embody substantial measurement error at the student level at which our analysis
operates.
Fourth, we consider only undergraduate mismatch. Our results may not generalize to
contexts, such as law schools, that provide students with fewer dimensions on which to respond to
an environment that proves too challenging or not challenging enough. In law school, for example,
students cannot easily change majors or take fewer courses. For this reason, mismatch, particularly
overmatch, might have very different overall effects in these contexts than in ours.
Finally, this paper considers only academic match. As noted in Smith (2008), other types
of mismatch between students and their undergraduate institutions represent an important omission
from most of the literature. Perhaps the most obvious concerns mismatch in terms of social class
or socio-economic status, or what an economist might prefer to call (at the cost of losing some
nuance in interpretation) family resources. Recent scholarly books such as Armstrong and
Hamilton’s (2012) Paying for the Party and Radford’s (2013) Top Student, Top School? highlight
this form of mismatch, as does Tom Wolfe (2004) in his novel of college life entitled I Am
Charlotte Simmons. Because mismatch on social class will likely correlate with academic
mismatch, it represents a potentially confounding treatment in our context.
29
References
Alon, Sigal and Marta Tienda (2005): “Assessing the `Mismatch’ Hypothesis: Differences in College
Graduation Rates by Institutional Selectivity,” Sociology of Education, 78, pp. 294-315.
Altonji, Joseph, Prashant Bharadwaj, and Fabian Lange (2012): “Changes in the Characteristics of
American Youth: Implications for Adult Outcomes,” Journal of Labor Economics, 30(4), pp. 783-828
Andrews, Rodney, Jing Li and Michael Lovenheim (2014): “Heterogeneous Paths through College:
Detailed Patterns and Relationships with Graduation and Earnings,” Economics of Education Review, 42,
pp. 93-108.
Ann Arbor News (2008): “Academics and Athletics: A Four-Day Ann Arbor News Series on the
University of Michigan.” http://www.mlive.com/wolverines/academics/stories/
Archibald, Robert, David Feldman, and Peter McHenry (2015): “A Quality-Preserving Increase in Four-
Year College Attendance,” Journal of Human Capital, 9(3), pp. 265-297.
Arcidiacono, Peter, Esteban Aucejo and V. Joseph Hotz (2016): “University Differences in the
Graduation of Minorities in STEM Fields: Evidence from California,” American Economic Review,
106(3), pp. 525-562.
Arcidiacono, Peter, Esteban Aucejo, Andrew Hussey, and Kenneth Spenner (2013): “Racial Segregation
Patterns in Selective Universities,” Journal of Law and Economics, 56(4), pp. 1039-1060.
Arcidiacono, Peter, Esteban Aucejo and Ken Spenner (2012): “What Happens After Enrollment? An
Analysis of the Time Path of Racial Differences in GPA and Major Choice,” IZA Journal of Labor
Economics, 1:5.
Arcidiacono, Peter, and Michael Lovenheim (2016): “Affirmative Action and the Quality-Fit Trade-Off,”
Journal of Economic Literature, 54(1), pp. 3-51.
Armstrong, Elizabeth and Laura Hamilton (2013): Paying for the Party: How College Maintains
Inequality, Cambridge, MA: Harvard University Press.
Aucejo, Esteban and Jonathan James (2017): “Catching Up to Girls: Understanding the Gender Imbalance
in Educational Attainment Within Race,” Unpublished manuscript, Arizona State University.
Bastedo, Michael and Allyson Flaster (2014): “Conceptual and Methodological Problems in Research on
College Undermatch,” Educational Researcher, 42(3), pp. 92-99.
Becker, Gary and H. Gregg Lewis (1973): “On the Interaction between the Quantity and Quality of
Children,” Journal of Political Economy 81(2, part 2), pp. S279-S288.
Bettinger, Eric, Bridget Terry Long, Philip Oreopoulos, and Lisa Sanbonmatsu (2012): “The Role of
Application Assistance and Information in College Decisions: Results from the H&R Block FAFSA
Experiment,” Quarterly Journal of Economics, 127(3), pp. 1-38.
Black, Dan, Kermit Daniel, and Jeffrey Smith (2005): “College Quality and Wages in the United States,”
German Economic Review, 6(3), pp. 415-443.
30
Black, Dan and Jeffrey Smith (2004): “How Robust is the Evidence on the Effects of College Quality?
Evidence from Matching,” Journal of Econometrics, 121(1), pp. 99-124.
Black, Dan and Jeffrey Smith (2006): “Evaluating the Returns to College Quality with Multiple Proxies
for Quality,” Journal of Labor Economics, 24(3), pp. 701-728.
Bound, John, Michael Lovenheim and Sarah Turner (2010): “Why Have College Completion Rates
Declined? An Analysis of Changing Student Preparation and Collegiate Resources,” American Economic
Journal: Applied Economics, 2(3), pp. 129-157.
Bowen, William and Derek Bok (1998): The Shape of the River: Long-Term Consequences of
Considering Race in College and University Admissions, Princeton, NJ: Princeton University Press.
Bowen, William, Matthew Chingos, and Michael McPherson (2009): Crossing the Finish Line:
Completing College at America’s Public Universities, Princeton, NJ: Princeton University Press.
Cadena, Brian and Ben Keys (2015): “Human Capital and the Lifetime Costs of Impatience,” American
Economic Journal: Economic Policy, 7(3), pp. 126-153.
Card, David (1995): “Using Geographic Variation in College Proximity to Estimate the Return to
Schooling.” In L. Christophides, E. Grant and R. Swidinsky, eds., Aspects of Labour Market Behavior:
Essays in Honor of John Vanderkamp. Toronto: University of Toronto Press, pp. 201-222.
Castex, Gonzalo, and Evgenia Kogan Dechter (2014): “The Changing Roles of Education and Ability in
Wage Determination.” Journal of Labor Economics 32(4), pp. 685-710.
Cawley, John, James Heckman, and Edward Vytlacil (2001): “Three Observations on Wages and
Measured Cognitive Ability,” Labor Economics, 8(4), pp. 419-442.
Chetty, Raj, John Friedman, Emmanuel Saez, Nicholas Turner, and Danny Yagan (2017): “Mobility
Report Cards: The Role of College in Intergenerational Mobility.” Unpublished manuscript.
Chingos, Matthew (2012): “Graduation Rates at America’s Universities: What We Know and What We
Need to Know.” In A. Kelly and M. Schneider, eds., Getting to Graduation: The Completion Agenda in
Higher Education. Baltimore, MD: Johns Hopkins University Press, pp. 48-70.
Cohodes, Sarah and Joshua Goodman (2014): “Merit Aid, College Quality, and College Completion:
Massachusetts' Adams Scholarship as an In-kind Subsidy,” American Economic Journal: Applied
Economics, 6(4), pp. 251-85.
Cunha, Flavio (2009): “Crossing the Finish Line: A Review Article,” Journal of Human Capital, 3(4), pp.
354-378.
Cunha, Jesse and Trey Miller (2014): “Measuring Value-Added in Higher Education: Possibilities and
Limitations in the Use of Administrative Data.” Economics of Education Review 42, pp. 64-77.
Currie, Janet and Enrico Moretti (2003): “Mother’s Education and the Intergenerational Transmission of
Human Capital: Evidence from College Openings,” Quarterly Journal of Economics, 118(4), pp. 1495-
1532.
31
Dale, Stacy Berg and Alan Krueger (2002): “Estimating the Payoff of Attending a More Selective
College: An Application of Selection on Observables and Unobservables,” Quarterly Journal of
Economics, 2002, 107(4), pp. 1491-1527.
Dale, Stacy Berg and Alan Krueger (2014): “Estimating the Effects of College Characteristics over the
Career Using Administrative Earnings Data,” Journal of Human Resources, 49(2), pp. 323-358.
Daugherty, Lindsay, Paco Martorell, and Isaac McFarlin (2014): “Percent Plans, Automatic Admissions,
and College Outcomes,” IZA Journal of Labor Economics, 3:10.
Dillon, Eleanor and Jeffrey Smith (2017): “Determinants of the Match between Student Ability and
College Quality,” Journal of Labor Economics, 35(1), pp. 45-66.
Goldrick-Rab, Sara (2006): “Following Their Every Move: An Investigation of Social-Class Differences
in College Pathways,” Sociology of Education, 79(1), pp. 61-79.
Goodman, Joshua, Michael Hurwitz, Jonathan Smith (2017): “Access to Four-Year Public Colleges and
Degree Completion.” Journal of Labor Economics, 35(3), pp. 825-867.
Griffith, Amanda and Donna Rothstein (2009): “Can’t Get There From Here: The Decision to Apply to a
Selective College,” Economics of Education Review, 28, pp. 620-628.
Griliches, Zvi and Vidar Ringstad (1970): “Error-in-the-Variables Bias in Nonlinear Contexts,”
Econometrica, 38(2), pp. 368-370.
Heckman, James and Tim Kautz (2012): “Hard Evidence on Soft Skills,” Labour Economics 19(1), pp.
451-464.
Heckman, James and Salvador Navarro (2004): “Using Matching, Instrumental Variables and Control
Functions to Estimate Economic Choice Models,” Review of Economics and Statistics, 86(1), pp. 30-57.
Herrnstein, Richard and Charles Murray (1994): The Bell Curve: Intelligence and Class Structure in
American Life. New York: NY: Free Press.
Hershbein, Brad (2013): “Worker Signals among New College Graduates: The Role of Selectivity and
GPA,” W.E. Upjohn Institute for Employment Research Working Paper 13-190.
Hoekstra, Mark (2009): “The Effect of Attending the Flagship State University on Earnings: A
Discontinuity-Based Approach,” Review of Economics and Statistics, 91(4), pp. 717-724.
House, Emily (2017): “Finding the Fit: Exploring Postsecondary Undermatch in Tennessee.”
Unpublished doctoral dissertation, University of Michigan.
Hoxby, Caroline (2009): “The Changing Selectivity of American Colleges.” Journal of Economic
Perspectives, 23(4), pp. 95-118.
Hoxby, Caroline and Sarah Turner (2013): “Expanding College Opportunities for High-Achieving, Low
Income Students,” SIEPR Discussion Paper 12-014.
32
Hoxby, Caroline and Christopher Avery (2012): “The Missing `One-Offs’: The Hidden Supply of High-
Achieving, Low Income Students,” NBER Working Paper No. 18586.
Kurlaender, Michal, and Eric Grodsky (2013): “Mismatch and the Paternalistic Justification for Selective
College Admissions.” Sociology of Education 86(4): 294-310.
Light, Audrey and Wayne Strayer (2000): “Determinants of College Completion: School Quality or
Student Ability?” Journal of Human Resources, 35(2), pp. 299-332.
Light, Audrey and Wayne Strayer (2004): “Who Receives the College Wage Premium? Assessing the
Labor Market Returns to Degrees and College Transfer Patterns.” Journal of Human Resources 39(3):
746-773.
Lincove, Jane and Kalena Cortes (2016): “Match of Mismatch? Automatic Admissions and College
Preferences of Low- and High-Income Students,” NBER Working Paper No. 22559.
Long, Mark (2008): “College Quality and Early Adult Outcomes,” Economics of Education Review, 27,
pp. 588-602.
Long, Mark (2010): “Changes in the Returns to Education and College Quality,” Economics of Education
Review, 29, pp. 338-347.
Loury, Linda Datcher and David Garman (1995): “College Selectivity and Earnings,” Journal of Labor
Economics, 13(2), pp. 289-308.
Mattern, Krista, Emily Shaw, and Jennifer Kobrin (2010): “Academic Fit: Is the Right School the Best
School or is the Best School the Right School?” Journal of Advanced Academia, 21(3), pp. 368-391.
Neal, Derek and William Johnson (1996): “The Role of Premarket Factors in Black-White Wage
Differences,” Journal of Political Economy, 104(5), pp. 869-895.
Niu, Sunny Xinchun and Marta Tienda (2010): “The Impact of the Texas Top Ten Percent Law on
College Enrollment: A Regression Discontinuity Approach,” Journal of Policy Analysis and
Management, 29(1), pp. 84-110.
Oster, Emily (2017): “Unobservable Selection and Coefficient Stability: Theory and Evidence.” Journal
of Business and Economic Statistics, forthcoming.
Pallais, Amanda (2015): “Small Differences that Matter: Mistakes in Applying to College,” Journal of
Labor Economics, 33(2), pp. 493-520.
Radford, Alexandria (2013): Top Student, Top School? How Social Class Shapes Where Valedictorians
Go to College. Chicago: University of Chicago Press.
Roderick, Melissa, Jenny Nagaoka, Vanessa Coca, and Eliza Moeller with Karen Roddie, Jamiliyah
Gilliam, and Desmond Patton (2008): From High School to the Future: Potholes on the Road to the
Future. Chicago, IL: Consortium on Chicago School Research.
Rodriguez, Awilda (2015): “Tradeoffs and Limitations: Understanding the Estimation of College
Undermatch,” Research in Higher Education, 56, pp. 566-594.
33
Rothschild, Michael and Lawrence White (1995): “The Analytics of the Pricing of Higher Education and
Other Services in Which the Customers are Inputs,” Journal of Political Economy 103(3), pp. 573-586.
Sallee, James, Alexandra Resch, and Paul Courant (2008): “On the Optimal Allocation of Students and
Resources in a System of Higher Education,” The B.E. Journal of Economic Analysis and Policy:
Advances, 8(1), Article 11.
Sander, Richard and Stuart Taylor (2012): Mismatch: How Affirmative Action Hurts Students It’s
Intended to Help, and Why Universities Won’t Admit It, New York: Basic Books.
Scott-Clayton, Judith (2012): “Information Constraints and Financial Aid Policy,” NBER Working Paper
No. 17811.
Segal, Carmit (2012): “Working When No One is Watching: Motivation, Test Scores, and Economic
Success,” Management Science, 58(8), pp. 1438-1457.
Smith, Jeffrey (2008): “Heterogeneity and Higher Education.” In M. McPherson and M. Schapiro,
eds. Succeeding in College: What it Means and How to Make it Happen. New York: College Board, pp.
131-144.
Smith, Jonathan, Matea Pender, and Jessica Howell (2013): “The Full Extent of Student-College
Academic Undermatch,” Economics of Education Review, 23, pp. 247-261.
Steeg, Jill, Jodi Upton, Patrick Bohn, and Steve Berkowitz (2008): “College Athletes [sic] Studies Guided
Toward `Major in Eligibility’.” USA Today, Updated 11/19/08.
http://usatoday30.usatoday.com/sports/college/2008-11-18-majors-cover_N.htm
Turner, Sarah (2002): “Changes in the Return to College Quality,” Unpublished manuscript, University of
Virginia.
Wolfe, Tom (2004): I Am Charlotte Simmons, New York: Farrar, Straus, and Giroux.
Zimmerman, Seth (2014): “The Returns to College Admission for Academically Marginal Students,”
Journal of Labor Economics, 32(4), pp. 711-754.
34
Table 1A: Joint distribution of college quality and ability—NLSY-79
Ability 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Total
Quartiles (lowest) (highest)
1st Quartile 10.6 7.0 4.7 2.7
(lowest) (42.6) (27.9) (18.7) (10.8) (100.0)
[37.6] [27.9] [17.2] [13.8] [25.0]
2nd Quartile 7.7 7.0 6.4 3.9
(30.8) (28.0) (25.7) (15.5) (100.0)
[27.3] [28.0] [23.7] [19.9] [25.1]
3rd Quartile 5.9 6.0 8.4 4.7
(23.6) (24.1) (33.5) (18.8) (100.0)
[20.8] [24.0] [30.8] [23.9] [24.9]
4th Quartile 4.0 5.0 7.6 8.3
(highest) (16.1) (20.0) (30.6) (33.2) (100.0)
[14.3] [20.0] [28.2] [42.4] [25.0]
Ability 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Total
Quartiles (lowest) (highest)
1st Quartile 12.0 7.0 4.3 1.8
(lowest) (48.0) (27.8) (17.0) (7.2) (100.0)
[44.0] [25.7] [18.2] [8.1] [25.0]
2nd Quartile 7.4 8.0 5.5 4.1
(29.6) (31.9) (21.9) (16.6) (100.0)
[27.0] [29.4] [23.4] [18.7] [25.0]
3rd Quartile 4.9 6.9 7.1 6.1
(19.6) (27.7) (28.2) (24.5) (100.0)
[18.0] [25.6] [30.1] [27.6] [25.0]
4th Quartile 3.0 5.3 6.6 10.1
(highest) (12.0) (21.0) (26.6) (40.4) (100.0)
[11.0] [19.4] [28.3] [45.5] [25.0]
Each cell contains the overall percentage, (the row percentage), and [the column percentage]. College quality is
measured by the 4-factor index. Ability is measured by the first principal component of the ASVAB scores.
Percentages are weighted as described in the text. A Pearson’s chi-squared test rejects equality of the ability-quality
distributions across cohorts (p-value = 0.078).
35
Table 2: Summary of College Outcomes
NLSY-79 NLSY-97
4-year college starters 2,497 2,071
Graduate within 6 years 50% 60%
In 5 years 42% 53%
In 4 years or less 26% 34%
Complete STEM degree 15% 13%
Transfer 27%
Transfer to a higher quality college 7%
Transfer to a lower quality college 16%
Labor force participation 2-3 years after entering 95% 92%
Men 94% 89%
Women 95% 94%
Labor force participation 10-11 years after entering 95% 92%
Men 96% 93%
Women 94% 91%
Average earnings 2-3 years after entering $8,371 $9,382
Men $9,556 $10,842
Women $7,228 $8,246
Average earnings 10-11 years after entering $39,022 $42,828
Men $45,362 $48,830
Women $32,627 $37,970
All percentages are of all four-year college starters. Two-year average earnings are calculated among those who
worked in at least one of the target years. Weighted as described in the text.
36
Table 3: Effect of College Quality and Ability on Degree Attainment
NLSY 79 NLSY 97
Within 6 Within 4 Within 6 Within 4
years years years years
∂Outcome/∂A Q = p25 0.487* 0.265* 0.310* 0.141*
(0.055) (0.056) (0.058) (0.061)
Q = p50 0.495* 0.310* 0.264* 0.252*
(0.055) (0.053) (0.060) (0.059)
Q = p75 0.467* 0.346* 0.313* 0.347*
(0.057) (0.048) (0.064) (0.058)
∂Outcome/∂Q A = p25 0.223* 0.027 0.330* 0.241*
(0.058) (0.048) (0.060) (0.067)
A = p50 0.273* 0.102 0.360* 0.323*
(0.062) (0.057) (0.056) (0.056)
A = p75 0.210* 0.141* 0.308* 0.416*
(0.064) (0.058) (0.065) (0.051)
Observations 2,441 2,441 2,071 2,071
R-squared 0.129 0.096 0.201 0.180
Pr(interaction=0) 0.363 0.510 0.554 0.360
Pr(∂Outcome/∂A equal) 0.806 0.451 0.440 0.033
Pr(∂Outcome/∂Q equal) 0.304 0.207 0.537 0.068
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality as
described in the text. Standard errors in parentheses. * indicates that the estimated effect is statistically significant at
5%. The final three rows present the p-statistics from Wald tests that the coefficients on the interaction terms of the
ability-quality polynomial are jointly equal to zero, that the mean marginal effects of ability are equal across the
three percentiles of college quality, and that the mean marginal effects of college quality are constant across the
three percentiles of ability.
37
Table 4A: Counterfactuals from Re-assigning Students to Colleges, NLSY-79
If all students attend a If all students attend a
Actual outcome matched college 90th pctile college
Graduate within 6 years 49.7 50.3# 58.8#
(0.15) (0.19) (0.36)
Graduate within 4 years 26.0 27.4# 29.5#
(0.13) (0.18) (0.35)
Major in STEM 15.5 16.0# 14.4#
(0.12) (0.17) (0.25)
Earnings 2-3 years after 8,383 8,354 8,190#
starting college (25) (29) (89)
Earnings 10-11 years after 39,014 39,708# 40,135#
starting college (86) (116) (220)
Earnings 20-21 years after 64,330 67,169# 69,768#
starting college (182) (279) (563)
Earnings 30-31 years after 79,201 84,639# 88,589#
starting college (295) (406) (907)
Counterfactual outcomes are calculated using the estimates reported in Tables 3, 5, and 6. Annual earnings are in
2010 dollars. Bootstrapped standard errors from 200 draws in parentheses. # indicates that the counterfactual
predictions are statistically different from the observed outcomes with 5% confidence.
38
Table 5: Effect of College Quality and Ability on Intermediate Outcomes
NLSY 79 NLSY 97
STEM STEM Transfer Transfer
degree degree up down
∂Outcome/∂A Q = p25 0.384* 0.250* 0.139* -0.074
(0.059) (0.061) (0.060) (0.039)
Q = p50 0.390* 0.236* 0.050 -0.025
(0.052) (0.048) (0.033) (0.055)
Q = p75 0.361* 0.237* 0.014 -0.064
(0.047) (0.040) (0.011) (0.064)
∂Outcome/∂Q A = p25 0.014 0.036 -0.132* 0.090
(0.018) (0.031) (0.031) (0.047)
A = p50 0.063 0.046 -0.165* 0.079*
(0.036) (0.041) (0.041) (0.039)
A = p75 0.079 0.037 -0.262* 0.108*
(0.052) (0.052) (0.061) (0.028)
Observations 2,396 2,067 2,044 2,044
R-squared 0.145 0.114 0.111 0.039
Pr(interaction=0) 0.708 0.791 0.384 0.328
Pr(∂Outcome/∂A equal) 0.688 0.960 0.098 0.363
Pr(∂Outcome/∂Q equal) 0.239 0.920 0.106 0.512
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality as
described in the text. Standard errors in parentheses. * indicates that the estimated effect is statistically significant at
5%.
39
Table 6A: Effect of College Quality and Ability on Earnings, NLSY-79
Year 2-3 Year 10-11 Year 20-21 Year 30-31
∂Earnings/∂A Q = p25 -3,118* 8,674* 28,372* 30,299*
(1,160) (3,379) (7,675) (11,487)
Q = p50 -3,365* 9,150* 30,619* 40,513*
(1,276) (3,562) (8,030) (12,278)
Q = p75 -2,858* 12,827* 42,637* 62,602*
(1,341) (3,588) (8,052) (12,302)
∂Earnings/∂Q A = p25 -1,797 4,183 12,945 12,242
(1,227) (2,839) (6,691) (9,718)
A = p50 -2,075 5,752 27,397* 29,541*
(1,141) (3,057) (8,098) (11,896)
A = p75 -1,759 7,243* 25,270* 40,482*
(1,039) (3,145) (7,808) (11,222)
Observations 2,094 2,228 1,792 1,593
R-squared 0.093 0.141 0.186 0.199
Pr(interaction=0) 0.816 0.710 0.035 0.083
Pr(∂Earnings/∂A equal) 0.842 0.502 0.219 0.077
Pr(∂Earnings/∂Q equal) 0.871 0.712 0.015 0.044
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality as
described in the text. Standard errors in parentheses. * indicates that the estimated effect is statistically significant at
5%. Annual earnings are in 2010 dollars.
40
Table 7: Earnings 10-11 Years after Starting College, by Parents’ Education
NLSY 79 NLSY 97
Parents H.S. Parents some Parents H.S. Parents some
only college only college
∂Earnings/∂A Q = p25 17,427* 2,151 9,498 3,252
(5,336) (4,055) (11,359) (4,154)
Q = p50 15,743* 5,495 15,805 3,303
(5,190) (4,665) (11,458) (4,678)
Q = p75 15,417* 12,815* 17,449 12,039
(4,985) (5,016) (11,324) (5,053)
∂Earnings/∂Q A = p25 7,902* -76 12,274 9,519
(3,765) (4,165) (7,077) (4,915)
A = p50 7,331 4,952 7,600 14,741
(4,641) (3,987) (7,419) (4,764)
A = p75 5,442 9,236* 19,246 17,464
(5,201) (4,071) (10,480) (4,506)
Observations 2,199 2,199 1,712 1,712
R-squared 0.145 0.145 0.139 0.139
Pr(interaction=0) 0.969 0.527 0.612 0.424
Pr(∂Earnings/∂A equal) 0.881 0.180 0.688 0.211
Pr(∂Outcome/∂Q equal) 0.921 0.246 0.406 0.402
N, subgroup 1,076 1,123 421 1,291
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality
as described in the text. Standard errors in parentheses. * indicates that the estimated effect is statistically
significant at 5%. Annual earnings are in 2010 dollars. Students are divided into those who grew up with
at least one parent who had some college education or who had no parents with more than a high school
education.
41
Table 8: Earnings 10-11 Years after Starting College, by Sex
NLSY 79 NLSY 97
Women Men Women Men
∂Earnings/∂A Q = p25 10,060* 7,077 4,944 2,120
(4,276) (4,663) (4,438) (6,369)
Q = p50 9,231* 8,108 7,547 294
(4,242) (5,276) (4,887) (6,953)
Q = p75 10,874* 13,926* 12,711 11,724
(4,106) (5,368) (5,105) (7,340)
∂Earnings/∂Q A = p25 5,856 2,755 9,788 11,904
(3,161) (4,751) (5,075) (6,401)
A = p50 6,542 4,977 13,967 16,481
(3,525) (5,013) (4,924) (6,650)
A = p75 5,821 7,973 17,338 20,086
(3,994) (4,536) (4,702) (6,694)
Observations 2,228 2,228 1,732 1,732
R-squared 0.142 0.142 0.138 0.138
Pr(interaction=0) 0.942 0.892 0.674 0.768
Pr(∂Earnings/∂A equal) 0.847 0.461 0.478 0.229
Pr(∂Earnings/∂Q equal) 0.933 0.688 0.464 0.650
N, subgroup 1,171 1,057 966 766
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality
as described in the text. Standard errors in parentheses. * indicates that the estimated effect is statistically
significant at 5%. Annual earnings are in 2010 dollars.
42
Table 9A: Covariate Set Comparisons for Graduating within 6 Years, NLSY-79
(1) (2) (3) (4) (5)
∂Outcome/∂A Q = p25 0.494* 0.514* 0.476* 0.487* 0.486*
(0.049) (0.051) (0.054) (0.055) (0.055)
Q = p50 0.508* 0.527* 0.489* 0.495* 0.490*
(0.049) (0.051) (0.055) (0.055) (0.055)
Q = p75 0.528* 0.532* 0.486* 0.467* 0.458*
(0.048) (0.049) (0.055) (0.057) (0.057)
∂Outcome/∂Q A = p25 0.224* 0.236* 0.222* 0.223* 0.233*
(0.056) (0.054) (0.055) (0.058) (0.058)
A = p50 0.299* 0.316* 0.292* 0.273* 0.280*
(0.062) (0.060) (0.061) (0.062) (0.062)
A = p75 0.248* 0.255* 0.235* 0.210* 0.210*
(0.062) (0.061) (0.062) (0.064) (0.064)
Additional ability measures Yes Yes Yes Yes
Demographics and family Yes Yes Yes
Neighborhood Yes Yes
Additional covariates Yes
Observations 2,497 2,497 2,497 2,441 2,441
R-squared 0.095 0.113 0.123 0.129 0.135
Pr(interaction=0) 0.502 0.355 0.363 0.363 0.333
Pr(∂Outcome/∂A equal) 0.862 0.944 0.940 0.806 0.783
Pr(∂Outcome/∂Q equal) 0.195 0.124 0.187 0.304 0.280
Table 9B: Covariate Set Comparisons for Graduating within 6 Years, NLSY-97
(1) (2) (3) (4) (5)
∂Outcome/∂A Q = p25 0.330* 0.358* 0.306* 0.310* 0.288*
(0.060) (0.058) (0.059) (0.058) (0.059)
Q = p50 0.264* 0.316* 0.260* 0.264* 0.255*
(0.061) (0.058) (0.060) (0.060) (0.060)
Q = p75 0.341* 0.362* 0.304* 0.313* 0.295*
(0.066) (0.063) (0.064) (0.064) (0.064)
∂Outcome/∂Q A = p25 0.479* 0.393* 0.355* 0.330* 0.332*
(0.057) (0.059) (0.058) (0.060) (0.059)
A = p50 0.508* 0.420* 0.380* 0.360* 0.361*
(0.055) (0.056) (0.055) (0.056) (0.056)
A = p75 0.436* 0.371* 0.330* 0.308* 0.319*
(0.067) (0.064) (0.064) (0.065) (0.065)
Additional ability measures Yes Yes Yes Yes
Demographics and family Yes Yes Yes
Neighborhood Yes Yes
Additional covariates Yes
Observations 2,071 2,071 2,071 2,071 2,071
R-squared 0.127 0.173 0.195 0.201 0.212
Pr(interaction=0) 0.225 0.586 0.606 0.554 0.704
Pr(∂Outcome/∂A equal) 0.237 0.536 0.491 0.440 0.618
Pr(∂Outcome/∂Q equal) 0.411 0.615 0.591 0.537 0.608
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality.
Standard errors in parentheses. * indicates that the estimated effect is statistically significant at 5%.
Annual earnings are in 2010 dollars. Covariate categories defined in Appendix Table A2.
43
Table 10A: Covariate Set Comparisons for Earnings 10-11 Years after Start, NLSY-79
(1) (2) (3) (4) (5)
∂Earnings/∂A Q = p25 12,376* 10,613* 8,369* 8,674* 8,385*
(2,981) (3,169) (3,322) (3,379) (3,434)
Q = p50 13,391* 11,495* 8,913* 9,150* 8,979*
(3,259) (3,461) (3,547) (3,562) (3,573)
Q = p75 17,932* 15,856* 13,338* 12,827* 12,916*
(3,165) (3,378) (3,515) (3,588) (3,569)
∂ Earnings/∂Q A = p25 5,942* 6,583* 6,641* 4,183 3,602
(2,749) (2,750) (2,683) (2,839) (2,798)
A = p50 7,526* 8,096* 8,500* 5,752 5,516
(3,064) (3,007) (2,931) (3,057) (3,013)
A = p75 10,336* 10,677* 10,296* 7,243* 6,990*
(3,100) (3,049) (3,008) (3,145) (3,155)
Additional ability measures Yes Yes Yes Yes
Demographics and family Yes Yes Yes
Neighborhood Yes Yes
Additional covariates Yes
Observations 2,278 2,278 2,278 2,228 2,228
R-squared 0.057 0.068 0.134 0.141 0.145
Pr(interaction=0) 0.510 0.553 0.533 0.710 0.648
Pr(∂Earnings/∂A equal) 0.321 0.357 0.336 0.502 0.448
Pr(∂Earnings/∂Q equal) 0.531 0.579 0.599 0.712 0.637
Table 10B: Covariate Set Comparisons for Earnings 10-11 Years after Start, NLSY-97
(1) (2) (3) (4) (5)
∂Earnings/∂A Q = p25 3,306 3,769 2,871 3,581 3,732
(3,620) (3,631) (3,835) (3,891) (3,927)
Q = p50 4,451 5,087 3,369 4,173 3,929
(4,111) (4,043) (4,234) (4,283) (4,356)
Q = p75 14,481* 15,225* 12,085* 12,667* 12,731*
(4,400) (4,481) (4,453) (4,458) (4,478)
∂ Earnings/∂Q A = p25 14,158* 12,877* 12,334* 10,393* 9,764*
(4,123) (4,205) (4,032) (4,089) (4,106)
A = p50 20,075* 19,064* 16,831* 14,820* 14,284*
(4,149) (4,134) (4,005) (4,065) (4,110)
A = p75 24,461* 23,489* 20,671* 18,634* 17,868*
(3,928) (3,969) (3,912) (3,986) (3,994)
Additional ability measures Yes Yes Yes Yes
Demographics and family Yes Yes Yes
Neighborhood Yes Yes
Additional covariates Yes
Observations 1,732 1,732 1,732 1,732 1,732
R-squared 0.075 0.079 0.130 0.136 0.141
Pr(interaction=0) 0.176 0.172 0.262 0.281 0.259
Pr(∂Earnings/∂A equal) 0.085 0.082 0.137 0.149 0.134
Pr(∂Earnings/∂Q equal) 0.155 0.145 0.283 0.289 0.291
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality.
Standard errors in parentheses. * indicates that the estimated effect is statistically significant at 5%.
Annual earnings are in 2010 dollars. Covariate categories defined in Appendix Table A2.
44
Figure 1: Expenditures per Student as a Function of the College Quality Index
Instructional expenditure per student from iPEDS and U.S. News and World Report. College quality indices
calculated as described in the text.
45
Figure 2: Effect of College Quality on Graduating within 6 Years
a. NLSY-79 Cohort
b. NLSY-97 Cohort
46
Figure 3: Long-term Earnings Estimates for NLSY-79
a. Effects of College Quality
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality as
described in the text. Markers indicate that the estimated effect is statistically significant at 5%. A subset of these
estimates is presented in Table 6A.
47
APPENDIX
48
where available or from the youth survey (98.6% from parent
survey). The NLSY79 did not give parents a separate survey. First
quartile is omitted group.
Siblings Number of siblings reported by the respondent in the NLSY 79 or
children age 18 and under living at the respondent’s address in
1997 for the NLSY-97.
Parental education Indicators for the highest educational attainment of either of the
respondent’s resident parents (or only parent in single parent
households) as reported in the first survey waves for each cohort.
We include at most one resident mother and father figure using the
following prioritization: biological, adopted, step, or foster. High
school diploma is the omitted category.
Neighborhood
Regional indicators Indicators for region of the U.S. (Northeast, South, Midwest,
West) where the respondent lived in 1979 or 1997. Midwest is the
omitted category.
Rural Indicator that the respondent did not live within a Metropolitan
Statistical Area (MSA) in 1979 or 1997.
% Adults w/college deg. in county The share of the over-25 population that has a 4-year college
degree in the county where the respondent lived in the first year of
each survey, from the 1972 and 1994 County and City Databooks,
respectively.
Additional covariates
Overweight/Obese Indicators that the respondent was overweight or obese (using BMI
and CDC definitions) in 1979 or 1997.
Religious observance per year How many times per year the respondent attended religious
services in 1979 or 1997 (entered in regressions as indicators for
each range of values offered in the survey).
Count of enriching resources Count of educational resources the respondent said he or she had
regular access to at home in the 1979/1997 surveys. In the NLSY-
79, these resources are: a magazine subscription, a newspaper
subscription, and a library card. In the NLSY-97 these resources
are: a computer, a dictionary, and a quiet place to study.
Had contact with biological Indicator that respondent had ever lived with each biological
mother/father parent for at least three months by the age of 18 (NLSY-79) or had
any contact with each biological parent by 1997 (NLSY 97).
49
Appendix Table A3: Starting Student Characteristics, NLSY-79
Notes: This table describes the characteristics of students attending each college quality quartile. All statistics are
weighted as described in text. Ability percentiles are among 4-year college starters, with the ASVAB measures
adjusted by age when taking the test. Family, neighborhood, and additional characteristics are measured in the first
1979 survey. Parents’ education is the maximum over resident parents. % of adults over the age of 25 in home
county with a four-year college degree is taken from 1970 census.
50
Appendix Table A4: Starting Student Characteristics, NLSY-97
Notes: This table describes the characteristics of students attending each college quality quartile. All statistics are
weighted as described in text. Ability percentiles are among 4-year college starters, with the ASVAB measures
adjusted by age when taking the test. Family, neighborhood, and additional characteristics are measured in the first
1997 survey. Parents’ education is the maximum over resident parents. % of adults over the age of 25 in home
census district with a four-year college degree is taken from 1990 census.
51
Appendix Table A5: Principal Components of the ASVAB
NLSY-79 NLSY-97
1st 2nd 1st 2nd
Component Component Component Component
Eigenvalue 5.46 0.84 5.25 0.88
Total variance explained 0.68 0.10 0.66 0.11
Eigenvectors:
General Science 0.37 -0.28 0.37 -0.32
Arithmetic Reasoning 0.37 -0.17 0.38 -0.05
Word Knowledge 0.38 -0.08 0.37 -0.23
Paragraph Comprehension 0.37 0.05 0.38 -0.09
Numerical Operations 0.33 0.51 0.31 0.55
Coding Speed 0.30 0.63 0.28 0.62
Mathematics Knowledge 0.37 -0.10 0.38 0.09
Mechanical Comprehension 0.32 -0.47 0.34 -0.37
Note: Following Altonji, Bharadwaj, and Lange (2011), we adjust scores on each test component the age of the
respondent when they took the test by calculating age-specific percentiles and then assigning each student the score
that corresponds to their percentile for 16 year olds.
52
Appendix Table A6: Principal Components of the College Quality Proxies
1992 2008
Eigenvalue 2.46 2.56
Variance explained 61% 64%
Eigenvectors:
Mean SAT 0.56 0.55
Rejection rate 0.47 0.46
Faculty/Student ratio 0.45 0.47
Avg. faculty salaries 0.51 0.52
Calculated from the set of four-year colleges in IPEDS in each year that report all four college quality proxies (1,157
institutions in 1992, 1,346 in 2008). In both years, quality measures that are missing in IPEDS are filled in where
possible using data from U.S. News and World Report.
53
Appendix Table A7: Effect of College Quality and Ability on College Outcomes, Quartile Dummies
NLSY-79 NLSY-97
Graduate Earnings, Graduate Earnings,
within 6 years 10-11 years within 6 years 10-11 years
ASVAB q1, Quality q2 0.025 3,401 0.133* 5,862
(0.049) (2,243) (0.044) (3,114)
ASVAB q1, Quality q3 0.038 8,122* 0.162* -791
(0.056) (3,301) (0.054) (2,929)
ASVAB q1, Quality q4 0.178* -1,581 0.179* 10,566*
(0.069) (3,233) (0.072) (5,112)
ASVAB q2, Quality q1 0.119* 4,835 0.145* 8,278*
(0.050) (2,517) (0.045) (2,547)
ASVAB q2, Quality q2 0.199* 7,009* 0.220* 7,597*
(0.052) (2,558) (0.044) (3,098)
ASVAB q2, Quality q3 0.271* 8,708* 0.269* 7,884*
(0.052) (2,674) (0.049) (3,343)
ASVAB q2, Quality q4 0.271* 4,760 0.424* 10,966*
(0.061) (3,398) (0.060) (3,642)
ASVAB q3, Quality q1 0.173* 4,231 0.099 3,940
(0.055) (2,381) (0.053) (3,720)
ASVAB q3, Quality q2 0.200* 9,174* 0.283* 4,026
(0.056) (2,860) (0.046) (2,774)
ASVAB q3, Quality q3 0.309* 11,899* 0.325* 10,225*
(0.051) (2,801) (0.046) (3,375)
ASVAB q3, Quality q4 0.365* 11,054* 0.371* 17,048*
(0.060) (3,346) (0.049) (3,909)
ASVAB q4, Quality q1 0.367* 10,307* 0.278* 3,895
(0.066) (3,395) (0.061) (3,152)
ASVAB q4, Quality q2 0.398* 10,811* 0.328* 8,648*
(0.060) (3,807) (0.051) (3,334)
ASVAB q4, Quality q3 0.431* 10,893* 0.364* 15,162*
(0.053) (3,030) (0.049) (3,851)
ASVAB q4, Quality q4 0.442* 14,556* 0.487* 22,685*
(0.054) (3,231) (0.044) (3,750)
Observations 2,441 2,228 2,071 1,732
R-squared 0.127 0.145 0.201 0.138
Pr(constant slope) 0.782 0.400 0.702 0.040
Standard errors in parentheses. * indicates that the estimated effect is statistically significant at 5%. The final row
reports the p-statistic from a Wald test of whether the differences in the effects of adjacent quality quartiles are
constant across ability quartiles (i.e. A2Q2-A2Q1 = A1Q2-A1Q1).
54
Appendix Table A8: Degree Attainment, Linear Probability Model
NLSY 79 NLSY 97
Within 6 Within 4 Within 6 Within 4
years years years years
∂Outcome/∂A Q = p25 0.503* 0.276* 0.312* 0.155*
(0.059) (0.054) (0.060) (0.056)
Q = p50 0.515* 0.336* 0.280* 0.279*
(0.061) (0.056) (0.059) (0.058)
Q = p75 0.481* 0.388* 0.312* 0.365*
(0.062) (0.054) (0.062) (0.065)
∂Outcome/∂Q A = p25 0.220* 0.031 0.344* 0.238*
(0.056) (0.046) (0.058) (0.060)
A = p50 0.274* 0.105 0.376* 0.352*
(0.063) (0.056) (0.056) (0.057)
A = p75 0.211* 0.144* 0.330* 0.455*
(0.063) (0.059) (0.062) (0.060)
Observations 2,441 2,441 2,071 2,071
R-squared 0.166 0.102 0.245 0.209
Pr(interaction=0) 0.153 0.203 0.650 0.030
Pr(∂Outcome/∂A equal) 0.751 0.211 0.630 0.015
Pr(∂Outcome/∂Q equal) 0.202 0.102 0.477 0.017
Mean marginal effects are calculated from the coefficients of a polynomial of ability and college quality as
described in the text. Standard errors in parentheses. * indicates that the estimated effect is statistically significant at
5%. The final three rows present the p-statistics from Wald tests that the coefficients on the interaction terms of the
ability-quality polynomial are jointly equal to zero, that the mean marginal effects of ability are equal across the
three percentiles of college quality, and that the mean marginal effects of college quality are constant across the
three percentiles of ability.
55