Internet Use Report

“President Obama believes that all Americans should have access to broadband and
the transformative opportunities it affords. Broadband services allow individuals to
access new career and educational opportunities. They help businesses reach new
markets and improve efficiency. They support struggling communities that seek to
attract new industries. And they enhance the government’s capacity to deliver
critical services.”
From: Recovery Act Investments in Broadband
National Economic Council, Executive Office of the President
December 2009

The release of this report, Digital Nation: 21st Century America’s Progress Toward Universal
Broadband Internet Access, by the U.S. Department of Commerce occurs at a critical juncture in
the nation’s quest for universal broadband Internet access. The report confirms that at the end of
the first decade of the 21st Century, too many Americans still rely on slow, narrowband Internet
access or do not use the Internet at all. This fact and others revealed in the report underscore the
importance of the Administration’s policy objective to ensure that all Americans have affordable
access to broadband Internet services.
The report’s findings are based on data collected in October 2009 through a special Internet Use
Supplement, sponsored by NTIA, to the U.S. Census Bureau’s Current Population Survey. With
a sample size of approximately 54,000 households and 129,000 citizens, the survey data provides
compelling information on the state of Internet use and broadband access across America. In
combination with other data collection efforts currently taking place at the Federal
Communications Commission, NTIA and other federal, state and private entities, this report will
greatly enrich our knowledge across numerous dimensions on the status of Internet connectivity.
The Internet has not only transformed the way we communicate, but also how we live, work, and
learn. Although life without high speed Internet service seems unimaginable for many
Americans, for too many others, broadband is still unattainable. As the world leader in
technology innovation and the place where the Internet was pioneered, we can and must do
better. This report will help identify both the gaps in Internet access and the reasons people that
have such access are choosing not to use it. We hope that, armed with this new data,
policymakers can support our nation’s continued competitiveness in the 21st Century information
economy.
Lawrence E. Strickling
Assistant Secretary for Communications and Information
U.S. Department of Commerce

NATIONAL TELECOMMUNICATIONS AND INFORMATION ADMINISTRATION
Lawrence E. Strickling, Assistant
Secretary for Communications and
Information

Anna Gomez, Deputy Assistant Secretary
for Communications and Information

U.S. Commerce Department Project Team
NTIA:

U.S. Bureau of the Census:

Office of Policy Analysis & Development:
Daniel J. Weitzner, Associate
Administrator for Policy Analysis
and Development
James McConnaughey, Chief Economist
Dennis Amari, Senior Policy Analyst

Demographic Surveys Division
Lisa Clement
Gregory Weyland
Karen G. Wms. Woods

Office of the Assistant Secretary:
Deena Shetler, Senior Policy Advisor

Technologies Management Office
Agatha Heesock Jung
Demographic Statistical Methods Division
Thomas F. Moore
Kimball T. Jonas
Ariel I. Teichman
Housing and Household Economic
Statistics Division: Population Division
Kurt J. Bauman

Acknowledgments
The Project Team would like to thank Tom Power, Jessica Schafer, Bart Forbes, Charles Franz,
and Josephine Arnold of NTIA; Mark Doms of the Economics and Statistics Administration;
David Johnson of the Census Bureau; and Marc Berejka and Patricia Buckley of the Office of
the Secretary for their contributions to this report.

EXECUTIVE SUMMARY
During the first decade of the 21st Century, U.S. broadband Internet connectivity by households
has increased dramatically as its importance to our economy and way of life has grown. Based
on a survey of over 50,000 households commissioned by the National Telecommunications and
Information Administration (NTIA) and conducted by the United States Census Bureau, virtually
all demographic groups have increased their adoption of broadband services at home over time.
The data also reveal that demographic disparities among groups have persisted over time.
Persons with high incomes, those who are younger, Asians and Whites, the more highlyeducated, married couples, and the employed tend to have higher rates of broadband use at home.
Conversely, persons with low incomes, seniors, minorities, the less-educated, non-family
households, and the non-employed tend to lag behind other groups in home broadband use.
Survey results demonstrate that persons in rural areas are less likely to use the Internet. For
example, Blacks and Hispanics in rural areas exhibit a lesser propensity to use broadband than
their counterparts in urban areas. A substantial difference in home broadband penetration
remains between urban and rural areas. Although the gap has declined since 2007, it still is
significant.
Despite the growing importance of the Internet in American life, over 30 percent of households
and 35 percent of persons do not use the Internet at home, and 30 percent of all persons do not
use the Internet anywhere. Those with no broadband access at home amount to more than 35
percent of all households and approximately 40 percent of all persons, with a larger proportion in
rural areas in both categories. Overall, the two most important reasons given by survey
respondents for not having broadband access at home are “don’t need” and “too expensive.”1
Inadequate or no computer is also a major reason given for no home broadband adoption. In rural
America, lack of availability is a much more important reason for non-adoption than in urban
areas.
The U.S. Department of Commerce will undertake a more detailed analysis later this year when
the full data base becomes available, and anticipates sponsoring new collections of Census data
and conducting analyses of these data bases. We also will look forward to the findings that the
broader research community will provide based on this data.

INTRODUCTION
Universal access to and adoption of 21st Century broadband for all citizens is a top priority for
the Obama Administration. Widespread access is critical to America’s future as the world’s
economic leader because of its impact on increasing our productivity, global competitiveness,
and improving Americans’ quality of life – through economic growth and development, job
creation, national security, telemedicine, distance learning, public safety, civic engagement, and
telework. As the President stated:
One key to strengthening education, entrepreneurship, and innovation in communities… is
to harness the full power of the Internet, and that means faster and more widely available
broadband.2

3

In the analysis below, the U.S. Department of Commerce’s NTIA probes the data collected by its
sister agency, U.S. Bureau of the Census, as part of the Current Population Survey (CPS). This
special Internet Use Supplement periodically surveys approximately 54,000 households and
gathers information on some 129,000 persons.3 The Census Bureau conducted this survey in
October 2009, the eighth such Internet survey sponsored by NTIA since the early 1990s.4
Below, the report documents the rapid growth overall of both broadband and the Internet in
general, and the disparate increases in adoption experienced by demographic groups and
geographic areas.5 Finally, we examine the major reasons why some Americans do not access
broadband Internet at home.6 The raw data on which this report is based are posted at
http://www.ntia.doc.gov/data/index.html and can be found through dataset pointers at
http://www.data.gov.7

ACCESS AND USE
During the first decade of the 21st Century, U.S. broadband Internet connectivity by households
has grown dramatically as its importance to our economy and way of life has grown.8
Figure 1: Percent of Households with Computers and Internet Connections,
Selected Years, 1997-2009*
80
68.7

Computer
61.7

Percent of U.S. Households

61.8
56.2

60

63.5

51.0
42.1

50.3

54.6

50.8

40
41.5

Internet

36.6

Broadband Internet
26.2

20
19.9

18.6
4.4

9.1

0
|

|

Oct 97

Dec 98

|

|

Aug 00 Sep 01

|

|

|

Oct 03

Oct 07

Oct 09

* Note: 2001, 2003, 2007, and 2009 Census-based weights and earlier years use 1990 Census-based weights.

In October 2009, according to the Census Current Population Survey data, 63.5 percent (75.8
million) of U.S. households used a high-speed Internet – “broadband” -- service (i.e.,
technologies that are faster than dial-up, such as DSL, cable modem, fiber optics, satellite, and
wireless). This represented a 25 percent increase from just two years earlier (50.8 percent in
October 2007). From the initial CPS study results in August 2000 (4.4 percent), broadband
adoption exhibited robust gains each time new data results were developed. (See Figure 1.)
During this time, the incidence of dial-up use leveled off, then plummeted. In 2000, dial-up
already was a thriving service, with 37.0 percent (39.0 million) of households having such

4

capability by August. In October 2003, the incidence of dial-up use had slipped to 34.3 percent
(38.6 million) of all households, only to decline more than sevenfold by October 2009 to 4.7
percent (5.6 million). Virtually all demographic groups have experienced rising broadband use at
home over time. Demographic groups categorized by family income levels, age, race,
educational attainment, employment status, household types, and gender all have enjoyed a
higher incidence of broadband connectivity since 2007. (See Figures 2-7 below.)
The data also have revealed that the basic demographic characteristics of broadband use at home
have tended not to change.
Usage by income. In 2009, when viewed by income, the users of broadband at home ranged
from the highest percentages by those persons who are most affluent (with annual family income
of $150,000 or greater) to the lowest percentages by those with $15,000 annual family income or
less. These 2009 results mirror the 2007 survey which also found that the highest broadband use
at home was by those who were most affluent and the lowest use was by those persons living in
households with $15,000 annual family income or less.
Figure 2: Persons Using Broadband in the Home by Family Income, 2007-2009
100
Oct. 2007
Oct. 2009

78.7

80

Percent of All Persons Ages 3+

88.7

84.9
82.4

70.0
75.5
58.7

60

56.8

45.0
35.2

40

68.1

42.6

29.2
33.2
20

24.2
18.6

0

Family Income

Usage by age. By age brackets, persons 18-24 years old exhibited the greatest broadband use at
home, while seniors (55 years or older) used broadband at home the least. Those 25-54 years old
proved to be middle range in adoption as well as in the array of age brackets. This is consistent
with our findings in 2007.

5

Figure 3: Persons Using Broadband in the Home by Age, 2007-2009
90

80.8

Oct. 2007
Oct. 2009

Percent of All Persons Ages 3+

75

69.3

68.2
58.5

60

57.8

55.9
45

65.8

69.3

46.1

53.5

47.2

30

33.7

15

0
5-17 years

18-24 years

25-34 years

35-44 years

45-54 years

55+ years

Age

Usage by race and ethnicity. Gauging broadband use at home by race and ethnicity, Asian nonHispanics led all other groups in 2009, with White non-Hispanics second in usage, followed by a
grouping of Black non-Hispanics, Native Americans (American Indians/Alaskan Natives), and
Hispanics. In 2007, the pattern was not dissimilar; although Hispanics and Native Americans
switched places, the small differential may not be statistically significant.9
Figure 4: Persons Using Broadband in the Home by Race, 2007-2009
80
Oct. 2007

67.3

Percent of All Persons Ages 3+

65.7

Oct. 2009

60

59.8
45.9

53.6

42.6

39.7

40

33.5
29.2

26.6

20

0
White Non-Hispanic

Black Non-Hispanic

AI/AN Non-Hispanic

Asian Non-Hispanic

Hispanic

Race

Usage by education level. For households with householders 25 years and older in 2009, 84
percent of those with college degrees had broadband access at home. In contrast, only 28 percent
of those householders with less than a high school diploma had such access. The conclusion that

6

persons with the highest levels of education exhibit the highest broadband use and those with the
least education experience the lowest adoption rate is consistent with past survey results on the
subject.10
Usage by employment status. Employment status revealed a pattern with broadband use at home
that featured the highest percentage use by those persons who were employed, with the
unemployed and particularly those persons not in the labor force lagging behind in usage. This is
similar to our findings in 2007.
Figure 5: Persons Using Broadband in the Home by Employment, 2007-2009
80
Oct. 2007
Oct. 2009

Percent of All Persons Ages 16+

70.5

58.4

60

57.4

47.4
39.0

40

34.8
20

0
Employed

Unemployed

Not in Labor Force

Employment

Usage by household type. Classifying by household type, married couples with children younger
than 18 years old surpassed all other groups in broadband use at home in 2009 and 2007,
followed by family households (i.e., two-parent or single parent) without younger children.
Non-family (i.e., where the householder is unrelated to other members of the household) rated
lowest in both years, followed by female householders (heads of house) with young children.

7

Figure 6: Households Using Broadband in the Home by Household Type,
2007-2009
90
Oct. 2007

79.8

Oct. 2009

Percent of Householders

75

67.7
60.0

67.0

60

56.9
50.8
53.4

45

47.6
43.4

39.6

30

15

0
Mar. Couple
w/Child<18

Male Hhldr w/Child<18 Female Hhldr w/Child
<18

Family Hhld w/o
Child<18

Non-family
Households

Household Type

Usage by gender. With respect to gender, males registered only slightly higher broadband use at
home than females in 2009 (59.3 percent v. 59.0 percent). This is consistent with our findings in
2007 (48.3 percent v. 47.0 percent).
Figure 7: Persons Using Broadband in the Home by Gender, 2007-2009
75
Oct. 2007
Oct. 2009

59.3

Percent of All Persons Ages 3+

60

45

59.0

48.3

47.0

30

15

0
Males

Gender

Females

Usage by location. Americans in rural areas tend to have lower broadband adoption rates than
their demographic counterparts in urban areas. For example, in 2009 Blacks (28.8 percent) and
Hispanics (33.8 percent) in rural areas exhibited much lower levels of broadband use at home

8

than their counterparts in urban areas (47.8 percent and 40.1 percent, respectively). Similarly,
both employed (61.7 percent) and unemployed persons (49.6 percent) in rural America had
significantly lower broadband use at home than their counterparts in urban areas (72.8 percent
and 60.2 percent, respectively). This is consistent with our findings in 2007, albeit at lower levels
of participation.

Blacks Non-Hispanics

Figure 8: Use of Broadband at Home by Black Non-Hispanics and Hispanics in
Urban and Rural Areas, 2007-2009
47.8
2009

28.8

35.4

Urban
Rural

2007

16.6

40.1
2009
Hispanics

33.8

29.4
2007

26.8
0

10

20

30

40

50

60

Percent of Persons Ages 3+

Figure 9: Use of Broadband at Home by Employed and Unemployed in Urban
and Rural Areas, 2007-2009
72.8
2009
Employed

61.7
60.6

Urban
Rural

2007

44.4

60.2
Unemployed

2009

49.6
47.2
2007

39.5
0

10

20

30

40

50

60

70

80

Percent of Persons Ages 3+

9

There remains a substantial difference in overall broadband use at home between urban and rural
areas. The gap has declined since 2007 but still exists. In 2009, 65.9 percent of urban households
and 54.1 percent of rural households accessed broadband service. In contrast, 8.9 percent of rural
households and only 3.7 percent of urban households used dial-up. In 2007, 53.8 percent of
households in urban areas and 38.8 percent of households in rural America were broadband
users. Again, rural homes relied more heavily on dial-up (19.3 percent) than urban did (8.5
percent) that year. Broadband use at home also varies by regions, with the West (68.0 percent of
households) and Northeast (67.0 percent) leading, followed by the Midwest (62.2 percent), and
the South (60.0 percent) in 2009.
Internet usage anywhere. In 2009, the incidence of Internet use anywhere (i.e., inside or outside
the home) by Americans totaled 68.4 percent (197.9 million persons, ages three and older).11
This represents an increase from 62.4 percent (177.9 million) in 2007. Similar to the broadband
pattern, all demographic categories with respect to Internet use anywhere experienced rising
adoption over time but historical demographic differences in use have continued. Interestingly,
the urban-rural gap in Internet use anywhere in 2009 registered only 4.4 percentage points (69.3
v. 64.9 percent), which was decidedly less than the broadband use at home differential. In 2007,
however, the urban and rural divide relating to Internet use in all areas equaled only 2.2
percentage points (62.8 v. 60.6 percent).

NON-USE
Non-usage at home. Collectively, the proportion of those Americans who do not use the Internet
at home declined between 2007 and 2009 but persists at a level higher than 30 percent today.
Utilizing a household measurement, the percentage dropped by seven percentage points during
the span, registering 31.3 percent in 2009. Measured in persons (three years and older), such
non-users numbered 36.8 percent in 2009 and had dropped almost seven percentage points in
two years.
Those persons who do not use broadband at home total more than 35 percent of all households
and approximately 40 percent of all persons, with a larger proportion in rural areas. More
specifically, 36.5 percent of households and 40.8 percent of persons did not use such high-speed
Internet at home in 2009. The urban-rural gap was found to exist, regardless of the unit of
measure. In rural areas, 45.9 percent of households and 48.4 percent of persons had no home
broadband access, while the corresponding numbers for urban areas are 34.1 percent for
households and 38.9 percent for persons. In 2007, the overall figures and those for rural and
urban respectively registered more than ten percentage points higher.

10

Figure 10: Persons and Households with No Internet Use at Home or Anywhere,
and No Broadband Use, 2007-2009
36.8

No Internet Use at
Home/Persons

43.5
Oct. 2009

31.3

No Internet Use at
Home/Hhld

Oct. 2007

38.3
31.6

No Internet Use
Anywhere/Persons

37.6
40.8

No Broadband Use at
Home/Persons

52.4
36.5

No Broadband Use at
Home/Hhld

49.2
0

10

20

30

40

50

60

Percent of All Person Ages 3+

Non-usage anywhere. Similarly, the category of not using the Internet inside or outside the home
remains at more than 30 percent of all persons. Those persons (ages three and older) who do not
use the Internet at all numbered 31.6 percent in 2009 and 37.6 percent in 2007.
Figure 11: Persons and Households with No Broadband in Rural and Urban
Areas, 2007-2009
38.9
Urban/Persons

49.8
Oct. 2009
Oct. 2007

48.4
Rural/Persons

62.4

34.1
Urban/Households

46.2

45.9
Rural/Households

61.2
0

10

20

30

40

50

60

70

Percent of Persons Ages 3+

11

Main reasons for non-use of broadband. Overall, the two most important reasons for no
broadband access at home are “don’t need” and “too expensive.” Households without high-speed
Internet access at home stated that “don’t need” (a value proposition) is more important than cost
(affordability). The next most important reason is “no computer or inadequate computer,”
followed by “can use somewhere else,” “not available,” and “lack of skill.”
Figure 12: Main Reason for No High Speed Internet Use at Home, 2009
Don't Need/
Not Interested
37.8%

Can Use
Somewhere Else
4.4%

Not Available
3.6%

No Computer or
Computer
Inadequate
18.3%

Other
6.6%

Too Expensive
26.3%

Lack of Skill
3.0%

In rural America, “not available” is a much more important reason for non-adoption of home
broadband than in urban areas. “Don’t need” and cost ranked highest in both categories, and the
“no computer or computer inadequate” issue ranked third. However, “not available” accounts for
more than 10 percent of the main reasons for non-use in rural areas but accounts for only about a
one percent factor in urban areas.

12

Figure 13: Main Reason for No High-Speed Internet Use at Home, Rural/Urban, 2009
Rural: 10.9 Million Households

Can Use
Somewhere
Else
3.6%

Urban: 32.1 Million Households

No
Computer or
Computer
Inadequate
16.3%

Not
Available
11.1%

Lack of Skill
2.3%

Can Use
Somewhere
Else
4.7%

Not Available
1.1%
No Computer
or Computer
Inadequate
19.0%

Don't
Need/Not
Interested
37.7%

Lack of Skill
3.2%

Too
Expensive
22.3%

Don't
Need/Not
Interested
38.1%
Other
6.4%

Too
Expensive
27.6%

Other
6.2%

When other types of non-use are examined, however, the rankings can and do change. For
example, respondents who do not use the Internet anywhere ranked the value proposition
significantly higher than affordability.
Figure 14: Main Reason Given for No Internet Use at Any Location, 2009
Lack of Skill
4.3%

Too Expensive
18.6%
Other
5.5%

No Computer or
Computer
Inadequate
22.3%

Not Available
0.7%

Can Use
Somewhere Else
1.4%

Don’t Need/
Not Interested
47.2%

This contrasts with the category of households that do not access the Internet at home, which
rated cost as the clear-cut top concern.

13

Figure 15: Main Reason Given for No Internet Connection at Home, 2009
Too Expensive
38.9%

Other
8.9%

Lack of Skill
0.4%

Don’t Need/
Not Interested
16.7%

No Computer or
Computer
Inadequate
16.8%
Not Available
2.7%

Can Use
Somewhere Else
14.8%

Similarly, those in households with dial-up service identified cost as the most important reason
for not having broadband connectivity at home.
Figure 16: Main Reason Given for Lack of Broadband in Households with
Dial-up Internet Access, 2009
Too Expensive
41.3%

Lack of Skill
0.8%

No Computer or
Computer
Inadequate
1.1%
Not Available
19.9%

Other
8.1%

Don’t Need/
Not Interested
27.3%

Can Use
Somewhere Else
1.6%

The answer category “no computer or computer inadequate” is highly ranked by each of the
above groups (Figures 13-16) except the dial-up category. The latter cites lack of availability as
much more important, especially in rural areas where it is cited as the most important reason
(42.6 percent), followed by cost (32.5 percent).

14

CONCLUSION
The Internet is integral to the U.S. economy and our standard of living. The nation’s broadband
Internet access adoption rate is at an all-time high, but a number of Americans still do not use the
technology. This initial analysis of the 2009 survey results begins the process of developing a
factual basis for sound policymaking to expand the adoption of and access to Internet
technology, particularly broadband.
This preview report is based on summary Current Population Survey (CPS) data provided to
NTIA by the Census Bureau. Later this year, the U.S. Department of Commerce will make the
complete data set publicly available, enabling a much more comprehensive analysis of Internet
usage patterns around the country. We anticipate that a more detailed examination of the CPS
data base would offer additional insights for public policymakers. The Commerce Department’s
National Telecommunications & Information Administration and Economics and Statistics
Administration will undertake such an analysis later this year when the full additional CPS data
base becomes available. We will also look forward to the findings that the broader research
community will provide based on this data.
While many Americans have come to grasp the importance of broadband as evidenced by
increasing use of the technology across virtually all demographic groups and geographic areas,
there exist differences in adoption rates across groups and areas that have tended to persist. Nonadoption rates for the Internet and broadband total 30 percent or more. Survey results provide
insights as to why this non-use occurs. Lack of perceived value (“don’t need”) in using the
technology ranks number one among the major reasons for non-use with respect to broadband at
home and Internet anywhere, and for rural and urban areas. Affordability (“too expensive”),
however, rates highest among the major reasons for eschewing broadband at home among those
with either no Internet at home or only dial-up service. Lack of an adequate or any computer
ranks high generally as a rationale for not having broadband at home, as does lack of availability
in rural areas. Further probing finds that the most important reasons that those with dial-up
service in rural areas do not subscribe to broadband include the latter’s unavailability and
affordability.
These findings provide a better understanding of the scope and nature of broadband adoption in
our nation. Knowing which demographic groups or geographic areas are either leading or
lagging – and why -- in their use of this high-speed Internet can sharpen the focus of public
decision-making that can help bridge this technological divide, boost economic growth and
create jobs. The Obama Administration's Open Government initiative is pursuing a wide range of
programs making government information and services available online. These efforts, along
with innovative commercial and non-commercial activities, can bolster demand for broadband
Internet access in market segments identified by this report as experiencing weak demand and
high access barriers. This Administration’s broadband deployment and adoption initiative is a
high-priority effort to help ensure that Americans can truly enjoy and participate in the Internet
economy.

15

1

By “most important,” we mean the most frequently ascribed major reason given by respondents in the survey.
President Barack Obama, September 21, 2009, in Troy New York. http://www.whitehouse.gov/the_press_office/Remarks-bythe-President-on-Innovation-and-Sustainable-Growth-at-Hudson-Valley-Community-College/
3
For household-level estimates based on the total sample, the error attributable to sampling and other random effects at the 90
percent confidence level is no more than plus or minus 0.35 percentage points based on a standard error (SE) of 0.21 percentage
points. For results based on Internet households, the margin of sampling error is no more than plus or minus 0.43 percentage
points, based on a SE of 0.26 percentage points.
4
More specifically, NTIA wholly or partially funded (with Commerce’s Economics and Statistics Administration) and designed
survey instruments for CPS Computer and Internet Use Supplements in 1994, 1997, 1998, 2000, 2001, and 2003, and Internet
Use Supplements in 2007 and 2009. The data became the basis for the Commerce reports “Falling Through the Net” (1995,
1998, 1999, 2000) and “A Nation Online” (2002, 2004), and provided input into the NTIA report “Networked Nation: Broadband
in America 2007.”
5
In a subsequent report, NTIA and ESA will examine in more detail the demographic details available in the October 2009 CPS
data base.
6
In this report, we examine broadband from the demand side based on the Census CPS survey of households. Thus, terms such
as “use,” “adopt,” “access,” and “connect” refer to the perspective of a household or person. This is to be distinguished from
supply-side (provider) considerations such as “deployment” that, in turn, can result in demand perceptions such as “lack of
availability.”
7
Data from the October 2009 CPS Internet Use survey can be retrieved at www.ntia.doc.gov/data/index.html. For historical CPS
data used in this report, see also "Internet and Computer Use Supplements" at http://www.bls.census.gov/cps_ftp.html#cpssupps.
8
As shown in Figure 1, household adoption of both Internet access overall and computers has been rising significantly since
1997 albeit collection of the latter data by Census ceased in 2003.
9
Figure 4 provides percentages for Native Americans in the column, “AI/AN Non-Hispanic,” which is an abbreviation for
“American Indians/ Alaskan Natives Non-Hispanic.”)
10
Previous years' data on Internet use have shown consistently that people with higher education levels are more likely to use the
Internet. See Jennifer Cheeseman Day, Alex Janus and Jessica Davis "Computer and Internet Use in the United States: 2003"
Series P23-208, Washington, DC: U.S. Census Bureau. Using a basis of persons who are 16 years and older in the October 2009
CPS data collection, the direct relationship between educational level and broadband adoption generally holds. This sample
would include current high school students as well as adults who never finished high school, thereby introducing distortions into
the analysis. The Department of Commerce’s more detailed look at educational attainment and broadband use in the future will
focus on the more meaningful CPS data set that includes only those persons 25 years of age and above.
11
Because a household measure is less useful than a person basis in gauging Internet or broadband use outside the home, it is not
used in this context in the report.
2

16