Educational Attainment in

the United States: 2015

Population Characteristics
Current Population Reports
By Camille L. Ryan and Kurt Bauman
March 2016
P20-578

This report provides a portrait of educational attain- had a bachelor’s degree or higher in 2015.3 Asians
ment in the United States based on data collected from were more likely than non-Hispanic Whites to have at
the Current Population Survey (CPS). The report exam- least a bachelor’s degree.
ines educational attainment of the adult population by •• Asians and non-Hispanic Whites were more likely to
demographic and social characteristics such as age, hold a bachelor’s degree or higher compared with
sex, race and Hispanic origin, and disability status, as Blacks and Hispanics.
well as differences in educational attainment between
•• Native adults were more likely to have a high school
the native and the foreign born.1 Historical data are also
education or higher but were no more likely than
included to present some general trends over time.
foreign-born adults to hold an advanced degree.
HIGHLIGHTS •• Adults without a disability were more likely to hold
•• In 2015, almost 9 out of 10 adults (88 percent) had at a bachelor’s degree or more than adults with a
least a high school diploma or GED, while nearly 1 in 3 disability.
adults (33 percent) held a bachelor’s or higher degree.2
DEMOGRAPHIC PORTRAIT OF EDUCATIONAL
•• The percentage of women who had a bachelor’s ATTAINMENT
degree or higher (33 percent) was not statistically dif-
In 2015, the majority (88 percent) of adults were at
ferent than the percentage of men (32 percent) with
least high school graduates and more than half (59
this level of education.
percent) had completed some college or more (Table 1).
•• Educational attainment varied by race and Hispanic One out of three adults (33 percent) reported they had
origin. More than half of Asians aged 25 and older a bachelor’s degree or more education, and 12 percent
reported an advanced degree, such as a master’s, pro-
fessional, or doctorate degree. Educational attainment
varied by age, sex, race and Hispanic origin, nativity,
and disability status.

1
The adult population is defined as the population 25 and older for 3
Federal surveys now give respondents the option of reporting more
the purposes of this analysis. than one race. Therefore, two basic ways of defining a race group are
2
The percentage before rounding can be found in Table 1 of this report. possible. A group such as Asian may be defined as those who reported
Although the estimate found in the table (32.5 percent) can be rounded Asian and no other race (the race-alone or single-race concept) or as
to 33 percent, this estimate is statistically different than 33 percent. All those who reported Asian regardless of whether they also reported
comparisons in this report are statistically significant at the 90 percent another race (the race-alone or in-combination concept). This report
level unless stated otherwise. The estimates are based on responses from shows data using the first approach (race alone). Use of the single-race
a sample of the population and may differ from actual values because population does not imply that it is the preferred method of presenting or
of sampling variability or other factors. As a result, apparent differences analyzing data. For further information, see the 2010 Census Brief,
between the estimates for two or more groups may not be statistically Overview of Race and Hispanic Origin: 2010 (C2010BR-02) at
significant. <[Link]
Table 1.
Educational Attainment of the Population Aged 25 and Older by Age, Sex, Race and
Hispanic Origin, and Other Selected Characteristics
(Numbers in thousands)
High school Some college or Associate’s Bachelor’s
Advanced degree
graduate or more more degree or more degree or more
Characteristic Margin Margin Margin Margin Margin
of error1 of error1 of error1 of error1 of error1
Total Percent (±) Percent (±) Percent (±) Percent (±) Percent (±)
Population 25 and older . . . . . . . . . 212,132 88.4 0.3 58.9 0.5 42.3 0.5 32.5 0.5 12.0 0.3

Age
25 to 34. . . . . . . . . . . . . . . . . . . . . . . 43,006 90.5 0.6 65.0 0.9 46.5 0.9 36.1 1.0 10.9 0.6
35 to 44. . . . . . . . . . . . . . . . . . . . . . . 39,919 88.7 0.5 62.8 0.9 46.7 1.0 36.3 1.0 13.8 0.7
45 to 64. . . . . . . . . . . . . . . . . . . . . . . 83,213 89.4 0.4 59.0 0.7 42.6 0.7 32.0 0.7 12.1 0.5
65 and older. . . . . . . . . . . . . . . . . . . . 45,994 84.3 0.7 49.7 0.9 34.1 0.9 26.7 0.8 11.3 0.7

Sex
Male. . . . . . . . . . . . . . . . . . . . . . . . . . 101,888 88.0 0.4 57.6 0.7 41.2 0.7 32.3 0.6 12.0 0.4
Female. . . . . . . . . . . . . . . . . . . . . . . . 110,245 88.8 0.3 60.1 0.6 43.4 0.6 32.7 0.6 12.0 0.4

Race and Hispanic origin

White alone. . . . . . . . . . . . . . . . . . . . 168,420 88.8 0.3 59.2 0.6 42.8 0.6 32.8 0.6 12.1 0.3
Non-Hispanic White alone . . . . . . . 140,638 93.3 0.3 63.8 0.6 46.9 0.7 36.2 0.7 13.5 0.4
Black alone . . . . . . . . . . . . . . . . . . . . 25,420 87.0 0.9 52.9 1.4 32.4 1.4 22.5 1.2 8.2 0.7
Asian alone. . . . . . . . . . . . . . . . . . . . 12,331 89.1 1.2 70.0 1.9 60.4 2.0 53.9 2.0 21.4 1.5
Hispanic (of any race) . . . . . . . . . . . . 31,020 66.7 1.1 36.8 1.0 22.7 0.9 15.5 0.7 4.7 0.4

Nativity Status
Native born . . . . . . . . . . . . . . . . . . . . 175,519 91.8 0.3 61.3 0.5 43.3 0.6 32.7 0.6 11.9 0.3
Foreign born . . . . . . . . . . . . . . . . . . . 36,613 72.0 1.0 47.6 1.1 37.6 1.1 31.4 1.1 12.5 0.7

Disability Status
With a disability. . . . . . . . . . . . . . . . . 28,052 78.6 0.9 41.6 1.2 24.9 1.0 16.7 0.9 5.7 0.5
Without a disability. . . . . . . . . . . . . . . 183,351 89.9 0.3 61.5 0.5 45.0 0.6 34.9 0.5 12.9 0.3
1
A margin of error is a measure of an estimate’s variability. The larger the margin of error in relation to the size of the estimate, the less reliable the estimate.
When added to and subtracted from the estimate, the margin of error forms the 90 percent confidence interval.
Source: U.S. Census Bureau, 2015 Current Population Survey.

Age. The overall increase in edu- more education compared to 36 degree and 12 percent of each sex
cational attainment documented percent of adults 25 to 34 years old held an advanced degree.4
over the past 6 decades occurred and 32 percent of adults aged 45
Race and Hispanic Origin. Educa-
as younger (and more educated) to 64 years (Table 1).
tional attainment also varied
cohorts replaced older, less edu-
Sex. Educational attainment differed by race and Hispanic origin.
cated cohorts in the adult popula-
between men and women. In 2015, Non-Hispanic Whites reported the
tion. In 2015, the 65 and older age
about 90 percent of both men and highest percentage of adults with
group reported lower levels of high
women had completed high school at least a high school education
school and college attainment than
or more. However, a higher per-
all younger age groups. Among 4
The percentage of men who held a bach-
centage of women had completed elor’s degree or higher was not statistically
adults aged 65 and older, 84 per-
at least some college. Sixty percent different from the percentage of women. Data
cent had completed high school from the American Community Survey (ACS),
of women had some college or released by the Census Bureau in September
or more education compared to
more education compared to 58 2015, show that women 25 years and older
91 percent of adults aged 25 to have a higher rate of college completion than
percent of men. Thirty-two percent men. The ACS is able to measure smaller
34 and 89 percent of adults aged
of men and 33 percent of women differences in the population due to its larger
35 to 44 years or 45 to 64 years sample size. See the blog entitled Women
had completed at least a bachelor’s Now at the Head of the Class, Lead Men in
(Table 1). In addition, 27 percent of
College Attainment at <[Link]
the population aged 65 and older .gov/2015/10/07/women-now-at-the-head
reported a bachelor’s degree or -of-the-class-lead-men-in-college
-attainment/>.

2 U.S. Census Bureau

(93 percent). Asians reported the completed a high school education adults with less than a high school
highest percentage of those with a or more. In addition, the percent- education (28 percent) compared
bachelor’s or higher degree age of those who had completed to natives (8 percent). However,
(54 percent). Hispanics reported the a bachelor’s degree or more was on the opposite end of the educa-
lowest percentage at every level about twice as large for those with- tional spectrum, the foreign born
from high school graduate or more out a disability compared to those were just as likely as the native
(67 percent) to advanced degrees with a disability. Thirty-five percent population to hold an advanced
(5 percent). of adults without a disability held degree (13 percent versus 12 per-
at least a bachelor’s degree com- cent respectively, not significantly
Disability Status. Adults who
pared to 17 percent of adults with different).5
reported a disability had lower lev-
a disability.
els of educational attainment com-
pared to those without a disability. Nativity Status. The educational
Seventy-nine percent of adults who attainment of the native and 5
Educational attainment differences
had a disability had at least a high foreign-born populations differ in between the native and foreign born arise
for many reasons. For example, educational
school diploma while 90 percent distribution (Figure 1). The foreign attainment categories in other countries do
of those without a disability had born had a higher proportion of not necessarily equate to the categories used
to measure attainment in the United States.

Figure 1.
Educational Attainment of the Population Aged 25 and Older by Nativity: 2015

Percent
35 Native Foreign born

30

25

20

15

10

0
Less than High school Some college or Bachelor's Advanced
high school completion associate's degree degree degree

Source: U.S. Census Bureau, 2015 Current Population Survey.

U.S. Census Bureau 3

HISTORICAL TRENDS two levels of education: complet- 1940 to 2015. In 1940, 5 percent
IN EDUCATIONAL ing high school or higher (regular of adults held a bachelor’s degree
ATTAINMENT high school diploma or GED) and or higher. By 2015, this percentage
completing a bachelor’s degree or had increased more than five fold
Data on educational attainment
higher (Figure 2). In 1940, one- to 33 percent.
were first collected by the
fourth of the population aged 25
U.S. Census Bureau in the 1940 In 1947, a higher portion of people
and older had completed high
Census. Since that time, data aged 25 to 29 had completed high
school. By 1967, over 50 percent
collection in the CPS has allowed school or higher than the total
of this population had reached this
a consistent, annual tracking of population 25 years and older.
level. This percentage continued to
the educational attainment of the Fifty-one percent of the population
increase, reaching 75 percent by
population.6 The increase in edu- aged 25 to 29 had completed high
1986 and 88 percent in 2015.
cational attainment can be seen at school compared to 33 percent of
6
See Educational Attainment in the United
The percentage of the adult popu- the total population 25 years and
States: 2007 (P20-560), available on the lation with a bachelor’s degree or over. However, the rates for these
Census Bureau Web site at <[Link]
/prod/2009pubs/[Link]>.
higher also increased steadily from two age groups began to converge

Figure 2.
Percentage of the Population 25 Years and Over Who Completed
High School or College by Age Group: Selected Years 1940–2015
Percent
100
25 to 29 years,
High school completion
90

80

70
25 years and older,
High school completion

60

50

40

25 to 29 years,
30 College completion

20
25 years and older,
College completion
10

0
1940 1950 1960 1970 1980 1990 2000 2010 2015

Note: Data for every individual year are not available for years prior to 1964.
Source: U.S. Census Bureau, 1947–2015 Current Population Survey and 1940 Decennial Census.

4 U.S. Census Bureau

in the 1990s. By 2015, the propor- completed a bachelor’s degree or adults with a bachelor’s degree
tion of those who had completed higher stayed within the range of or higher in all years. In 1988, 38
high school was 88 percent for the 21 to 25 percent. The rate for this percent of Asians held at least a
total population 25 years and older age group climbed to 36 percent in bachelor’s degree compared to 21
and 91 percent for adults aged 25 2015. The rate of college comple- percent of non-Hispanic Whites, 11
to 29. tion for the population 25 years percent of Blacks, and 10 percent
and older grew to 33 percent in of Hispanics.8 In 2015, a majority
The younger population was ahead
2015. of Asians 25 years and older had
of the general population in the
a bachelor’s degree or higher (54
rate of college completion for much TRENDS BY RACE AND percent). More than one-third of
of the period since 1940. In 1976, HISPANIC ORIGIN non-Hispanic Whites had a bach-
24 percent of people aged 25 to 29
Educational attainment has elor’s degree or higher (36 percent),
had a bachelor’s degree or higher,
increased for all race and Hispanic 22 percent of Blacks had this level
while 15 percent of the popula-
origin groups (Figure 3).7 Asians of education, as did 15 percent of
tion 25 and older had that level
had the highest percentage of Hispanics.
of education. After that point, the
attainment level of younger people
flattened out. During the 20 years 7
Changes in the CPS questions and data 8
The percentage of Blacks with a bach-
from 1975 to 1995, the propor- collection procedures limit the ability to reli- elor’s degree was not statistically different
ably report on Hispanic origin prior to 1973 from the percentage of Hispanics with a
tion of 25 to 29 year olds who had and Asian racial identification prior to 1988. bachelor’s degree.

Figure 3.
Percentage of the Population 25 Years and Older With a Bachelor's Degree
or Higher by Race and Hispanic Origin: 1988 to 2015

Percent
60

Asian

50

40 White,
non-Hispanic

30

Black

20
Hispanic

10

0
1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

Source: U.S. Census Bureau, 1988–2015 Current Population Survey.

U.S. Census Bureau 5

Although educational attainment The gap between non-Hispanic the presence of the foreign-born
has increased over time for all Whites and Asians ranged between population. In the previous section,
race groups and Hispanics, the 13 and 20 percentage points and it was observed that the gap in col-
gap between these groups has ended the period where it began lege completion between the total
remained the same for some and (17 percentage points in 1988 and population of Blacks and Hispanics
increased for others. In 1988, the 18 percentage points in 2015, not has grown over time. On the other
percentage of Hispanics with a statistically different). hand, once the influence of the
bachelor’s degree or higher was foreign-born population is taken
about the same as that of Blacks. EDUCATION BY RACE, away, it can be seen that native
By 2015, the percentage had HISPANIC ORIGIN, AND Hispanics have kept up with Blacks
increased for both groups, but the NATIVITY and others. Native Blacks and
gap between the two groups had The Current Population Survey first native Hispanics have had similar
also grown. In 2015, 15 percent of collected data about nativity in growth in the percentage with a
Hispanics had this level of educa- 1994. While education levels of the bachelor’s degree (Figure 4). The
tion compared to 22 percent of foreign born differ from those of percentage of native Blacks with at
Blacks—a difference of 7 percent- natives, these differences vary by least a bachelor’s degree rose from
age points. The gap between race and Hispanic origin. 12 percent in 1994 to 21 percent in
Whites and Blacks remained stable, 2015. In 1994, 10 percent of native
between 11 and 14 percentage Trends in education among Hispanics held a bachelor’s degree
points from 1988 to 2015. Hispanics are greatly influenced by

Figure 4.
Percentage of the Native Population Aged 25 Years and Older With a Bachelor's
Degree or Higher by Race and Hispanic Origin: 1994 to 2015

Percent
60

Asian

50

40 White,
non-Hispanic

30

Black
20
Hispanic

10

0
1994 1997 2000 2003 2006 2009 2012 2015

Source: U.S. Census Bureau, 1994–2015 Current Population Survey.

6 U.S. Census Bureau

 Figure 5.
Percentage of Population 25 Years and Older With a Bachelor's Degree or Higher
by Race, Hispanic Origin, and Nativity: 2015
Percent
60 Native Foreign born

50

40

30

20

10

0
White, non-Hispanic Black Asian Hispanic

Source: U.S. Census Bureau, 2015 Current Population Survey.

or higher. By 2015, this percentage population with a bachelor’s degree Hispanics were the only group
had doubled to 20 percent. or higher (55 percent) was not where the percentage of the native
statistically different from the per- population with a bachelor’s degree
The percentage with a college
centage of the foreign-born Asian or higher was higher than the per-
education differs by nativity within
population (54 percent). Among centage of the foreign-born popula-
race and Hispanic origin groups
White non-Hispanics and Blacks, tion with this level of education.
(Figure 5). Among Asians, non-
foreign-born adults were more Twenty percent of native Hispanics
Hispanic Whites, and Blacks, the
likely to hold a bachelor’s or higher had a college education compared
percentage of the foreign-born pop-
than native adults, with a difference to 12 percent of foreign-born
ulation with a bachelor’s degree or
of 12 percentage points for non- Hispanics.
higher was as high or higher than
Hispanic Whites and 10 percentage
the percentage of native adults
points for Blacks.
with that same level of education.
The percentage of the native Asian

U.S. Census Bureau 7

 Figure 6.
Percentage of the Population 25 Years and Older With a Bachelor's Degree
or Higher by Sex: 1967 to 2015
Percent
45

40

35

30

Men
25

20
Women

15

10

0
1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

Source: U.S. Census Bureau, 1967–2015 Current Population Survey.

COLLEGE COMPLETION again in 1990.9 In 2000, the gap to 32 percent and women’s to 33
AMONG MEN AND WOMEN had fallen to 4 percentage points, percent.10
and to 1 point in 2010 (both men
Historically, a higher percentage of The convergence in college comple-
and women rounding to 30 per-
men have held a bachelor’s degree tion between men and women in
cent). In 2015, the percentage
or higher compared to women the United States is the culmination
of men aged 25 and older with a
(Figure 6). In 1967, 13 percent of a long process of educational
bachelor’s degree or higher was not
of men 25 years and older held a advance led by younger women.
statistically different from that of
bachelor’s degree or higher com- An earlier report noted that young
women, with men’s rate rounding
pared to 8 percent of women. women aged 25 to 29 began to
However, the gap has narrowed have higher college attainment
over time. From 1970 through rates than young men in 1996.11
1990, the gap was in the range of 6 By contrast, prior to 1986, young
to 8 percent—a 6-percentage point 9
The gap between men and women
difference in 1970, a 7-percentage in 1967, 1970, 1980, and 1990 were not 10
As previously discussed (see footnote
point gap in 1980, and 6 points statistically different from one another with 4, p. 2), 2014 ACS data show women with
the exception of 1967 versus 1980. The 2000 higher levels of college completion than men.
and 2010 gaps were significantly different 11
Nicole Stoops, Educational Attainment
from earlier gaps and from each other, with in the United States: 2003, Washington, DC,
the exception of 1967 versus 2000 which U.S. Census Bureau, 2004. <[Link]
was not significant. .gov/prod/2004pubs/[Link]>.

8 U.S. Census Bureau

 Figure 7.
Percentage of the Population Aged 25 to 29 With a Bachelor's or Higher Degree,
by Sex: 1967 to 2015

Percent
45

40 Women

35
Men

30

25

20

15

10

0
1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

Source: U.S. Census Bureau, 1967–2015 Current Population Survey.

men had higher rates of college The story for women is somewhat percent to 36 percent with a bach-
completion. From 1967 to 1977, different. The attainment of women elor’s degree.
men aged 25 to 29 had a rate of aged 25 to 29 rose along with men
college completion that was 5 to 8 up to 1976–1977 and paused, but COLLEGE COMPLETION BY
percentage points higher than that did not significantly fall back in the SEX, RACE, AND HISPANIC
of women (Figure 7). Young men’s following years. The rate of college ORIGIN
college completion reached 27 per- completion climbed back above It is clear that women, especially
cent in 1976 and 1977, but subse- the 1977 level in 1989. Indeed, the younger women, have gained rela-
quently fell, and did not rise above entire period shown here is char- tive to their male counterparts in
27 percent again until 2011.12 This acterized by growth in women’s college completion over the period
marks a 35-year period when bach- bachelor’s attainment. In at least 32 from 1967 to 2015. It is interest-
elor’s degree attainment by young of the 44 years from 1972 to 2015, ing to see if these trends for young
men failed to climb overall. young women were ahead of where women were the same across
they had been 5 years earlier, and race and Hispanic origin groups.
in no year was there a significant Because sample size becomes
5-year decline. During the 37-year smaller as we interact more and
12
Bachelor’s attainment of men aged 25
to 29 was significantly lower than the level of period, 1976 to 2010, when young more variables, we use a broader
1976 and 1977 from 1980 through 1995. In men were at or below 27 percent
1978 and 1979, there was no significant dif- age group (25 to 34 rather than
ference from the levels of 1976 or 1977. completion, women went from 20

U.S. Census Bureau 9

25 to 29) and use the average level statistically significant). During a White, Black, and Hispanic origin
of college attainment over a 3-year large part of this 16-year period, groups.15 In addition, the gaps for
period to show these trends (Figure the line on the graph was lower for non-Hispanic Whites, Blacks, and
8). The figure shows women’s col- non-Hispanic Whites, meaning the Hispanics were not significantly
lege attainment relative to men degree to which men were ahead of different from each other, with the
for each race and Hispanic origin women was greater among non- non-Hispanic White gap differing
group—a rise in the curve show- Hispanic Whites than among Blacks significantly from the Black gap
ing an increase in women’s attain- or Hispanics.13 only in 1996 and from the Hispanic
ment compared to men. During gap only in 1993 (in both cases
In 1990, Asian women were less
the period from 1976 to 1989, the chance of having a bachelor’s
likely than Asian men to have a col-
non-Hispanic White women aged degree relative to women of the
lege degree, and the gap was larger
25 to 34 were less likely than non- same group was higher for non-
than for non-Hispanic White, Black,
Hispanic White men of that age to Hispanic Whites).
or Hispanic women.14 From 1990 to
have a college degree. However,
2000, the gap between women and Since 2000, 25- to 34-year-old non-
this was not true for 25 to 34 year
men was small for non-Hispanic Hispanic White, Black, and Hispanic
old Black or Hispanic women, who
were generally not behind their 13
The difference was significant in 12 15
Young non-Hispanic White women had
male counterparts (the exception out of 16 years when comparing Blacks and higher rates of bachelor’s degree attainment
non-Hispanic Whites and in 10 out of 16 years than non-Hispanic White men in only 2 years,
being 1974–1976 and 1980–1981 when comparing Hispanics and non-Hispanic 1997 and 1998. Young Black women had
when the difference between the Whites. higher rates than Black men in 1996 and
14
Data for Asian attainment is only avail- 1997. Young Hispanic women had higher
sexes for Hispanics was briefly able from 1988 forward. rates than Hispanic men only in 1998–2000.

Figure 8.
Percentage Point Difference Between Bachelor's Degree Attainment of Women
and Men Aged 25 to 34, by Race and Hispanic Origin: 1976 to 2015
(3-Year Running Average)
Percentage point difference favoring women
10

8 Women more likely than

men to have degree

Black
-2

Hispanic
-4

-6
Women less likely than
men to have degree
-8 Asian

White, non-Hispanic
-10
1975 1980 1985 1990 1995 2000 2005 2010 2015

Source: U.S. Census Bureau, 2015 Current Population Survey.

10 U.S. Census Bureau

women were significantly ahead of the 50 states and the District of other factors. As a result, appar-
their male counterparts, with the Columbia. The data do not repre- ent differences between the
exception of four instances where sent residents of Puerto Rico and estimates for two or more groups
Black women and men were not U.S. Island Areas. may not be statistically significant.
statistically different (2002, 2003, All comparative statements have
The data are based on a sample of
2007, and 2008). Asian women did undergone statistical testing and
about 100,000 addresses. The esti-
not pull significantly ahead of Asian are significant at the 90 percent
mates in this report are controlled
men until 2010. In terms of differ- confidence level, unless otherwise
to independent national population
ences between race and Hispanic noted. In this report, the variances
estimates by age, sex, race, and
origin groups, there was a switch of estimates were calculated using
Hispanic origin for March 2015.
from the period before 1990. In both the Successive Difference
Beginning with 2010, estimates are
the earlier period, it was among Replication (SDR) method and the
based on 2010 Census population
non-Hispanic Whites that women Generalized Variance Function
counts and are updated annually
had the lowest percentage with a (GVF) approach. Further informa-
taking into account births, deaths,
bachelor’s degree relative to men of tion about the source and accuracy
emigration, and immigration. The
the same group (with the possible of the estimates is available at
CPS is a household survey primarily
exception of Asian women, about <[Link]
used to collect employment data.
whom we don’t have sufficient /publications/2015/demo
The sample universe for the basic
data). In the period after 2000, /[Link]>.
CPS consists of the resident civilian
young non-Hispanic White women
noninstitutionalized population of MORE INFORMATION
often had the highest percentage
the United States. People in institu-
with a bachelor’s degree relative Detailed tabulations, related
tions, such as prisons, long-term
to men of the same group, com- information, and historic data are
care hospitals, and nursing homes,
pared to Black, Asian, or Hispanic available on the Internet at the
are not eligible to be interviewed in
women.16 Educational Attainment page on the
the CPS. Students living in dormi-
tories are included in the estimates Census Bureau’s Web site at
SOURCE OF ESTIMATES
only if information about them is <[Link]/hhes/socdemo
The data in this report are from the reported in an interview at their /education/[Link]>.
2015 Current Population Survey parents’ home. Since the CPS is a
(CPS) Annual Social and Economic household survey, people who are CONTACT
Supplement (ASEC) and earlier homeless and not living in shelters U.S. Census Bureau Customer
years of the supplement (prior to are not included in the sample. The Services Center
2003, known as the Annual sample universe for the CPS ASEC
Demographic Supplement or the 1-800-923-8282 (toll free)
is slightly larger than that of the
March Supplement). Data from the basic CPS since it includes military <[Link]
2015 CPS ASEC were collected in personnel who live in a household
with at least one other civilian
16
The female-male gap was larger for
non-Hispanic Whites than Blacks in 2002,
adult, regardless of whether they
2007, 2008, and 2014, larger for live off post or on post. All other
non-Hispanic Whites than for Asians in 2001–
2002, 2004–2006, 2008–2010, 2012–2013,
Armed Forces are excluded. For fur-
and 2015 and larger for non-Hispanic Whites ther documentation about the CPS
than for Hispanics in 2004–2006, 2008–2010,
ASEC, see <[Link]
2012–2013, and 2015. Some sources report
that Black women are still at least tied for /programs-surveys/cps
the lead relative to men (compared with /techdocs/[Link]>.
non-Hispanic White, Asian, and Hispanic
women) because Black women form a high
proportion of total Black graduates (see, for Most of the data from the CPS
example, Antoine M. Garibaldi, “Four Decades ASEC were collected in March, with
of Progress...and Decline: An Assessment of
African American Educational Attainment,” some data collected in February
The Journal of Negro Education, Vol. 66, No. and April. The estimates in this
2, Spring 1997, pp. 105–120. This appar-
ent contradiction is resolved by the differ-
report are based on responses
ence between “percentage point gap” and from a sample of the population
“proportion of total” as measures. The same
percentage point difference will represent a
and may differ from actual values
larger share when the overall proportion in because of sampling variability or
the group is smaller.

U.S. Census Bureau 11