News from the ETS Policy Information Center
Volume 9, Number 2 Educational Testing Service Summer 1999
"Figure 3 shows that, from ages 19-22, real
wage rates differ little, irrespective of the [GPA]
quartile in which the workers scored."
This issue was authored by Paul E. Barton,
Director, ETS Policy Information Center
(The original of this article, including charts, can be
accessed at
but requires Adobe Acrobat Reader to read. Its text is
reproduced here as a service to SQ website readers.)
Return to: Research Findings on Career Success
Go to: Home page
Comments to:
This Issue: Learn More, Earn More?
. Introduction
. The Effect of Grades
. The Effect of Achievement
. The Effect of Graduation
. Implications
Americans continually receive societal messages
telling them that learning more-and, particularly,
receiving degrees and certifications-generally leads
to higher earnings. Related information is published
regularly and is often communicated in terms of
average earnings for people with different levels
of education. Most high school students have read
about this relationship and have likely been told by
parents that they would have to get a good education
to make good money.
Perhaps less well known is that the earning differential
between those who have college degrees and
those who do not has been widening over the last
couple of decades. And real earnings have actually
declined for those with a high school diploma or less.
Economic incentives to learn beyond high school
abound, and economists attribute increasing post-high
school enrollments in educational institutions to
a growing market for advanced education. Economists
are the first to point out that labor markets play a
critical role in motivating young people to further
their education. However, there is a glaring exception
to this rule that has received practically no attention.
Students heading into the work force right after
high school do not receive market signals telling them
that learning more in high school will result in their
earning more upon leaving high school-up to, say,
age 20 or 21. This is important when trying to understand
student behavior and may have a lot to do with
students' motivation to complete high school, as well as
to take more rigorous courses and achieve good grades.
The Effect of Grades
Let's start with how the market treats high
school graduates who had good grades. Differences
in employment and earnings among
students who graduated from high school with
different grade-point averages (GPAs) and
who were not enrolled in college two years
later are summarized in Figure 1. The data
source is the National Center for Education
Statistics' NELS:88 survey, which began with
a national sample of eighth-graders in 1988
and tracked the students through 1994, when
they would have been out of high school for
two years, or 19- or 20-years-old.
Were those who had achieved better
grades more likely to be employed? Figure 1
shows that the percent of employed males was
slightly higher for those with high grades but
that the differences were not statistically significant.
For females, the middle group had a
higher percent employed than the low or high
group, and women with the highest GPAs
were no more likely to be employed than those
with the lowest GPAs.
The bottom half of Figure 1 shows differences
in earnings among students with different
GPAs. While it shows that male graduates
with the lowest GPAs had higher average
earnings than those with the highest GPAs,
the difference was not statistically significant.
For females, the monthly earnings were about
the same.
In sum, these data show that the labor
market does not generally reward high school
graduates who had good grades. Of course, the
meaning of grades can vary considerably from
school to school; so students with high GPAs
in one school may have what are considered
low GPAs at another school. Because of
this, we used a more rigorous comparison
approach, by also considering reading test
scores provided by NELS:88.
The Effect of Achievement
In Figure 2, we compare high school graduates
by the quartile in which they scored on
the reading test. The employment rate rose
for males from the lowest to the third quartile
but then dropped down considerably for
those scoring in the highest quartile. Thus,
the highest scoring male graduates were least
likely to be employed. For females, the differences
were small among the second, third,
and fourth quartiles, but the employment
rate was much lower for those scoring in the
lowest quartile.
In terms of monthly earnings, there was no
statistically significant difference between the
lowest and the highest quartiles for males. For
females, the earnings were almost identical.
We can also get a feel for earnings by using
the Armed Services Vocational Aptitude Battery
(ASVAB), given in the National Longitudinal
Survey of Youth by the U.S. Bureau of
Labor Statistics. In Figure 3, we can see workers'
average hourly rate of pay by test scores
for mathematics, science, and paragraph
comprehension. The data are for all civilian
workers ages 19-31, not just 19-year-old
high school graduates who are not enrolled in
school. (However, the results are similar if we
only look at high school graduates.)
as well as to take more rigorous courses and
achieve good grades.
Percentage Employed After Two Years (1994)
Average Earnings per Month After Two Years (1994)
Figure 1- The Difference Grades Make
Employment and Earnings of Male and Female 1992 High School Graduates
(not enrolled in school) Two Years After High School (1994), by Grade Point Average
Source: NELS:88 data (National Center for Education Statistics), calculated by the ETS Policy Information Center.
Percentage Employed After Two Years (1994)
Average Earnings per Month After Two Years (1994)
Figure 2 - The Difference Reading Achievement Makes
Employment and Earnings of Male and Female 1992 High School Graduates
(not enrolled in school) Two Years After High School (1994), by Reading Scores
Source: NELS:88 data (National Center for Education Statistics), calculated by the ETS Policy Information Center.
Figure 3 shows that, from ages 19-22, real
wage rates differ little, irrespective of the
quartile in which the workers scored. However,
after about age 22, the pay rates start to
separate, and those with higher scores start to
gain higher pay. The difficult question is why
this happens. John Bishop 1 has suggested a
couple of reasons.
One possibility, he says, is that high school
graduates who got higher grades or scores
entered jobs that had greater training opportunities;
so even though these jobs may not
have paid more initially, their related training
ultimately led to higher pay. The other possibility
Bishop discusses is that employers may
not fully realize the skills of young new hires
initially-but that they may over time. This
author would add that we don't know if
employers end up paying more for specific
academic knowledge or whether the same set
of characteristics that allows people to excel in
the school environment also enables them to
rise to the top in employment settings.
The Effect of Graduation
Overall, then, the previous discussion of findings
shows that the labor market does not
generally reward efforts to excel in school-or
that it does so only marginally-for those who
persevere to graduation.2 But in the first few
years after graduation, does it reward those
who have earned a high school diploma?
Figure 4 compares 19- and 20-year-olds
not enrolled in school, on the basis of whether
they dropped out of high school or got a
diploma (again using NELS:88 data). It shows
that males who graduated were somewhat
more likely to have a job (91 percent, com-pared
to 84 percent) but that females who
graduated were considerably more likely to
have one (81 percent, compared to 60 percent).
These differentials are less than what has
been reported in the annual census reports
(March Current Population Surveys). The
March 1999 report shows a differential of
24 percentage points in favor of high school
graduates from ages 18 to 20. For recent high
school graduates and dropouts, who are surveyed
by the census every October, the differential
has been narrowing-from 24 percent-age
points in 1980, to 22 in 1990, to 17 in
1996-suggesting less differentiation by
employers based on high school graduation
(October Current Population Surveys, Bureau
of the Census). So there seems to be some
variation in the size of the differential depending
on the time of the survey and the age of
the respondents.
Figure 4 shows that neither males nor
females who were working at ages 19 and 20
earned significantly different average monthly
wages based on whether or not they finished
high school. So while employers may often
prefer hiring youths with a high school
diploma, they will not, on average, initially
pay a 19- to 20-year-old more for having one;
however, a pay differential does emerge over
the years.
Although exactly comparable statistics do
not exist from 1975, we can safely say that
little or no progress has been made in increasing
the high school completion rate over the
last quarter century. In 1975, 83 percent of
those ages 19-20 had received a diploma or a
GED, compared to 85.9 percent in 1997, for
those ages 18-24. (Given that the 1997 age
group was larger than the 1975 one, these
percentages are likely about equal.)
1 John Bishop, "Achievement Test Scores and Relative Wages," in Workers and Their Wages: Changing Patterns in the United
States. Marvin Kosters (ed), Washington, D.C.: American Enterprise Institute Press, 1993.
2 This is based on only one explanatory variable. Further investigation, using multivariate
analyses, would be desirable.
Figure 3 - The Difference Age Makes
Average Hourly Pay Rate for Civilian Workers, by Age and Armed Services
Vocational Aptitude Battery (ASVAB) Score Quartile
Source: U.S. Department of Labor, Bureau of Labor Statistics, National Longitudinal Survey of Youth, 1973-93
Note: To control for differences in age at testing, individuals were assigned to age-specific performance quartiles for
each subject, based on their age at testing.
Figure 4 - The Difference A Diploma Makes
Employment and Earnings of Male and Female 19- and 20-Year Olds (not enrolled in school in 1994), With and Without a High School Diploma or GED From 1992
Source: NELS:88 data (National Center for Education Statistics), calculated by the ETS Policy Information Center.
In this analysis, however, we are more
interested in those who stay in high school
and graduate with a diploma-and the related
trend is disturbing when one considers reports
from The National Education Goals Panel
since 1988. In 1988, 80.3 percent of 18-24-
year-olds had regular high school diplomas; by
1992, this figure had risen to 81.2 percent.
Since then, there has been a steady decline: to
78.8 percent in 1994, to 77.5 percent in 1995,
to 76.4 percent in 1996, and to 74.9 percent in
1997. 3 This author has no explanation for the
recent decline but suggests that the labor
market is not sending the kind of signals in
the early years after high school that would
encourage youths to persevere.
While the lack of immediate labor market
reward for achievement in-or even finishing-
high school has been generally down-played,
or seemingly unnoticed, by economists
and educators, it has not likely gone unnoticed
by teenagers.4 Such youths are usually
attuned to how to earn the privileges and
rewards of adulthood-and they surely check
to see if putting extra time into studying is
generally worth the effort. They have friends
and older siblings who left school and went to
work, and they find out what these people
earn. They see the Help Wanted ads in the
papers and the signs in stores. They know
However, the plans and aspirations of
young people in this age range are also shaped
by forces other than peers. Teenagers know
how their parents and relatives fared and can
relate this information to how much education
these people received. Many also listen to the
advice of parents, who likely know that education
pays off ultimately, if not immediately.
We just don't know how the varying influences
and conflicting messages affect the decisions
young people make and the effort they expend.
One apparent fact, however, is that we
cannot consider a motivating force for high
school students to be the job market for youths
who have been out of school for two years or
so; we need to establish for them that the payoffs
may take several years to emerge. We
must also let these high-schoolers know that if
they are inclined to go on to college, they will
need to take more rigorous courses; and if they
aim for selective schools, they will also need to
get above-average grades in these courses.
This lack of differences in early employment
and earnings experience is not a recent development.
While the kind of data used in this
report was not available in the 1970s, enough
information was there to see that a similar
situation existed back then. In 1975, this
author wrote a book chapter entitled, "Youth
Employment and Career Entry,"5 which asks
the question, Is teenage employment different?
The answer determined was yes: that most 17-, 18-, and 19-year-olds are
working as sales clerks at the 5 and 10 or drug
store checkout counters, or in fast food chains.
3 Data for 1988-1997 are from various years of the October Current Population Surveys (unpublished data). Data for 1975 are
from Dropout Rates in the United States, 1991, National Center for Education Statistics, 1992.
4 However, John Bishop at Cornell has been calling our attention to the matter for more than a decade.
5 Paul E. Barton, "Youth Employment and Career Entry," in Labor Market Information for Youths, edited by Seymour Wolfbein,
Temple University School of Business Administration, Philadelphia, 1975, pp. 86-87.
Teenagers go into particular occupations
and industries that are teenage intensive,
and work their way into the kinds of jobs
adults hold as they pass the magic age of
20 or 21....
In raising the question of the access of
teenagers to adult-type jobs, the concern is
greatest with the school-leavers expecting to
enter the labor force on a full-time basis.
This review of the evidence strongly suggests
that teenagers are largely held back
from adult-type jobs, and that until age 20
or 21... a high school diploma (at least for
males) makes very little difference in the
early years of work experience.
Finishing high school does not, in the
labor market, coincide with entering economic
adulthood. Instead, says the 1975 report, "It
gives rise to a period where the youth is dislocated
from the education system and not yet
located in the employment world but remains
in a youth labor pool, aging and waiting."
The implication is that teenagers are waiting
to be duly compensated-since while they
are in this youth labor pool, they are all paid
about the same, irrespective of their academic
accomplishments. However, as they do when
considering candidates of any age, employers
of youth do give great weight to experience.
This can be seen in Figure 5, which shows
the employment rates for 1994 of those who
graduated high school in 1992.
Males who held a job in high school had a
higher post-graduation employment rate and
monthly earnings $350 higher than those
who did not. Females who worked in high
school had a post-graduation employment
rate 21 percentage points higher than those
who hadn't, and their monthly earnings were
about $250 higher.
Over the last 20 years the press has been
wont to report on studies that showed ill
effects of student employment on grades. But
actually, studies from this time frame show
that working a moderate number of hours
(20 or less) does not depress grades. At some
point, of course, work does affect schooling-
but common sense tells us this.
What has developed in post-World War II
America is a situation where those under 20
or 21 are pretty much treated alike in the
labor market. The age of entry into jobs that
pay more and have career ladders and fringe
benefits seems to be rising, and is likely into
the mid-20s now. But young graduates
who are not rewarded in the years after high
school for their academic efforts unlikely
realize this. As a result, raising high school
achievement levels and graduation rates are
difficult tasks.
The solution here is not obvious. One
encouraging effort now under way is the campaign
to get businesses to ask for high school
transcripts when they are hiring. In the past
this seldom happened, as John Bishop pointed
out in a study from the early 1990s.
Over the last two years the Business Coalition
for Education Reform has been urging
employers to ask for transcripts, as a way of
signaling to students that academics count in
the employment world; about 5,000 employers
to date (May 1999) have agreed to cooperate.
More must join the effort, though. A 1999
survey carried out by the Public Agenda Foundation
found that while 84 percent of high
school students say they would work harder
if they knew employers would look at their
transcripts, only 16 percent of employers say
they do indeed ask for transcripts.
Figure 5 - The Difference Working Makes
Employment and Earnings of Male and Female 1992 High School Graduates (not enrolled in school) Two Years After High School (1994), by Employment
in High School
Source: NELS:88 data (National Center for Education Statistics), calculated by the ETS Policy Information Center.
Overall, what we have is a disjuncture
between the age we complete public education
and the age of economic adulthood. This is
bound to cause a lot of problems-one of them
being high school students working well below
their potential. If nothing else, we need to
understand what has been happening over
the last three or more decades.
In other writings, this author has urged
employers and educators to collaborate by
providing our nation's youth with a combination
of schooling and work experience, and to
stretch out the high school years, if necessary,
to make sure we allow ample time for students
to master the academics. This kind of joint
effort has the potential to establish a better
link between learning and earning and, among
other benefits, to make the economic rewards
of school achievement tangible and clear.
We are not likely to fool youth into thinking
that learning in high school pays; in fact,
we should stop fooling ourselves. And if we tell
ourselves that all must go to college, we must
also keep in mind that only slightly more
than one in four 25- to 29-year-olds earns
a four-year college degree.
Those of us trying to bring about education
reform need to further examine how much
learning more actually results in earning
more. This dynamic may be critical to reaching
our ambitious national education goals.
Setting high standards for achievement certainly
seems the way to go, as does administering
fair and reliable tests that are aligned
to what is actually taught. But without consequences
in the work world that can be clearly
observed by high school students, forging
change will surely continue to be like rowing
upstream. In fact, more than 15 years of education
reform has seen little change in the
academic achievement of high school students
or in graduation rates-rates that have actually
been falling since 1992.
The previous discussion lends itself
to numerous implications and additional
.There is discussion about the "right" curriculum
to make high school students more
employable-giving them what employers
want them to have. But what employers
really seem to want is for hirees to be
older, and curriculum changes cannot age
high school graduates.
.Employers do recognize experience, as
seen in Figure 5. Yet, the school-to-work
movement's effort to incorporate worksite
training and experience into the high
school structure has resonated with only a
small proportion of employers. Can we do a
better job of building on the part-time work
experiences that teenagers are able to get
while going to school?
.We regularly read that employers are
dissatisfied with the education of young
people coming out of America's high
schools. It is, therefore, puzzling that
employers treat similarly those at both
the top and bottom of the grade and test-score
distribution scale, at least for the first
couple years after high school graduation.
.We have found many of our remedial youth
education and training programs to be
wanting in results. More specifically, findings
from controlled experiments and
evaluations using control groups showed
that, on follow up, those in the experimental
group did not fare better than those in
the control group. But at ages 18-21, or
thereabout, we know that all youth in the
work force are treated much the same.
What is the right standard of success,
then, in this kind of labor market?
.If a disjuncture does exist between the age
society stops providing education and the
age of maturity, as demanded by the labor
market for adult-type jobs, does a major
issue in need of public policy exist?
6 New York Times, February 17, 1999.
ETS Policy Notes is published by the
ETS Policy Information Center
Educational Testing Service
Rosedale Road, 04-R
Princeton, NJ 08541-0001
(609) 734-5694
Director, ETS Policy Information
Center: Paul E. Barton
Editor: Richard J. Coley
Copyright © 1999 by Educational
Testing Service. All rights reserved.
Educational Testing Service is an
Affirmative Action/Equal Opportunity
ETS, and the ETS logo are registered
trademarks of Educational Testing
The modernized ETS logo is a trade-mark
of Educational Testing Service.
04202-02536 Y79M7 Printed in the U.S.A.
I.N. 204930
Recent ETS Policy Information Center
reports can be downloaded from our Web
.Too Much Testing of the Wrong Kind;
Too Little of the Right Kind in K-12
.How Teachers Compare: The Prose,
Document, and Quantitative Skills
of America's Teachers
.Order in the Classroom: Violence,
Discipline, and Student Achievement
.Does it Compute? The Relationship
Between Educational Technology and
Student Achievement in Mathematics
Editor's Note
This issue was authored by
Paul E. Barton; the views he
expresses are his own.
We are indebted to Richard Fry
and Harold Wenglinsky of Edu-cational
Testing Service, and
John Bishop of Cornell Univer-sity
for their reviews. Reviewers
do not necessarily agree with all
statements and interpretations.
Harold Wenglinsky extracted
data for this report from
NELS:88. Janet Spiegel edited
the piece, Carla Cooper provided
the desktop publishing, and
Jim Chewning served as
production coordinator.
Return to: Research Findings on Career Success
Go to: Home page
Comments to:
Web (this website, 170+ pages on IQ and Success skills)

The ads below are placed by - they are not necessarily endorsed by this site