US v. Aaron Graham

Filing

OPINION ATTACHMENTS filed. Copies to all parties and the district court/agency. [12-4659, 12-4825]

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 1 of 57

NUMBERS, FAC AND TRENDS SHAPING THE W
CTS
S
WORLD

,
NOVEMBER, 12, 2014

Pu c Pe epti ns of
ubli erce ion f
Pr cy and Sec rity in
rivac a d cur
the Po -Sno den Er
e ost- owd n ra
5
1

Few see core comm icatio/20
w
muni /0ons
3
cha
annels as “very secu 08 p
s “ y weure” places to
s
d
v
sha pr
are rivate info ie tion
e 59,orma
.
No

-46
12

FOR FUR
RTHER INFORMA
ATION
ON THIS REPORT:
Mary Mad
dden, Senior Researcher, Internet
Project
202.419 .4372
www.pew
wresearch.org

RECOMMEND CITATION: Pew Research Center, November, 20 “Public Perce
DED
014
eptions of Privacy and Security in th Post-Snowden Era”
he
http://www.p
pewinternet.org/2014/11/12/public-privacy-percep
ptions/

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 2 of 57

1
PEW RESEARCH CENTER

About this Report
This report is the first in a series of studies that examines Americans’ privacy perceptions and
behaviors following the revelations about U.S. government surveillance programs by government
contractor Edward Snowden that began in June of 2013. To examine this topic in depth and over
an extended period of time, the Pew Research Center’s Internet Project commissioned a
representative online panel of 607 adults who are members of the GfK Knowledge Panel. These
panelists have agreed to respond to four surveys over the course of one year. The findings in this
report are based on the first survey, which was conducted in English and fielded online January 1128, 2014. In addition, a total of 26 panelists also participated in one of three online focus groups as
part of this study during August 2013 and March 2014.
This report is a collaborative effort based on the input and analysis of the following individuals:
Mary Madden, Senior Researcher, Internet Project
Lee Rainie, Director, Internet, Science and Technology Research
Kathryn Zickuhr, Research Associate, Internet Project
Maeve Duggan, Research Analyst, Internet Project
Aaron Smith, Senior Researcher, Internet Project

d
we

015
3/2
8/0

0

vie
, Internet Project on the topic of privacy and security
Other reports from the Pew Research Center
659
online can be found at: http://www.pewinternet.org/topics/privacy-and-safety/pages/2/
4
12.
NoResearch Center
About the Pew
Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes
and trends shaping America and the world. It does not take policy positions. It conducts public
opinion polling, demographic research, media content analysis and other empirical social science
research. The center studies U.S. politics and policy views; media and journalism; internet and
technology; religion and public life; Hispanic trends; global attitudes and U.S. social and demographic trends. All of the center’s reports are available at www.pewresearch.org. Pew Research
Center is a subsidiary of The Pew Charitable Trusts.
© Pew Research Center 2015

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 3 of 57

2
PEW RE
ESEARCH CENT
TER

Summ
mary of Findings
F
s
Privacy ev
vokes a const
tellation of concepts for Americans—
c
—some of th em tied to tr
raditional no
otions
of civil liberties and so
ome of them driven by concerns abo the surve
m
out
eillance of di
igital
communic
cations and the coming era of “big data.” While A
t
e
d
Americans’ associations with the top of
s
pic
privacy ar varied, the majority of adults in a new survey b the Pew R
re
e
f
by
Research Ce
enter feel tha
at
their priva is being challenged along such co dimensio as the se
acy
c
a
ore
ons
ecurity of their personal
l
informatio and their ability to re
on
r
etain confide
entiality.

d
we

e

, vi
659

015
3/2
8/0

0

4
12.
When Am
mericans are asked what comes to mind when the hear the w
a
c
ey
word “privac there are
cy,”
e
No

patterns to their answ
o
wers. As the above word cloud illustr ates, they gi importan weight to t
a
c
ive
nt
the
idea that privacy appli to person material—
p
ies
nal
—their space their “stuf their soli
e,
ff,”
itude, and,
important their “rig
tly,
ghts.” Beyon the freque
nd
ency of indiv
vidual words when resp
s,
ponses are
grouped in themes, the largest block of answ
nto
b
wers ties to c
concepts of s
security, safe and
fety,
protection For many others, notio of secrec and keepi things “h
n.
ons
cy
ing
hidden” are t of mind when
top
thinking about privacy
a
y.

Most are aware of government efforts to monitor com
m
mmunicatio
ons
More than a year after contractor Edward Sno
n
r
owden leake document about widespread
ed
ts
governme surveillan by the NSA, the casc
ent
nce
N
cade of news stories abou the revela
s
ut
ations contin to
nue
register widely among the public. Some 43% of adults hav heard “a l
w
g
o
ve
lot” about “th governme
he
ent
collecting information about telep
n
phone calls, emails, and other online communications as part of
e
e
efforts to monitor terr
m
rorist activity and anot
y,”
ther 44% hav heard “a l
ve
little.” Just 5 of adults in
5%
our panel said they ha heard “nothing at all about thes programs.
ave
l”
se
.

www.p
pewresearch.or
rg

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 4 of 57

3
PEW RESEARCH CENTER

Widespread concern about surveillance by government and businesses
Perhaps most striking is Americans’ lack of confidence that they have control over their personal
information. That pervasive concern applies to everyday communications channels and to the
collectors of their information—both in the government and in corporations. For example:





91% of adults in the survey “agree” or “strongly agree” that consumers have lost control
over how personal information is collected and used by companies.
88% of adults “agree” or “strongly agree” that it would be very difficult to remove
inaccurate information about them online.
80% of those who use social networking sites say they are concerned about third parties
like advertisers or businesses accessing the data they share on these sites.
70% of social networking site users say that they are at least somewhat concerned about
the government accessing some of the information they share on social networking sites
without their knowledge.

015
3/2
8/0

Yet, even as Americans express concern about government access to their data, they feel as though
government could do more to regulate what advertisers do with their personal information:





0
ed should be concerned about the
80% of adults “agree” or “strongly agree” that Americans
w
government’s monitoring of phone callsiand internet communications. Just 18% “disagree”
ve
or “strongly disagree” with that notion.
59,
-46
64% believe the government should do more to regulate advertisers, compared with 34%
. 12
who think the government should not get more involved.
No
Only 36% “agree” or “strongly agree” with the statement: “It is a good thing for society if
people believe that someone is keeping an eye on the things that they do online.”

In the commercial context, consumers are skeptical about some of the benefits of personal data
sharing, but are willing to make tradeoffs in certain circumstances when their sharing of
information provides access to free services.



61% of adults “disagree” or “strongly disagree” with the statement: “I appreciate that online
services are more efficient because of the increased access they have to my personal data.”
At the same time, 55% “agree” or “strongly agree” with the statement: “I am willing to
share some information about myself with companies in order to use online services for
free.”

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 5 of 57

4
PEW RESEARCH CENTER

There is little confidence in the security of common communications channels, and those
who have heard about government surveillance programs are the least confident
Across the board, there is a universal lack of confidence among adults in the security of everyday
communications channels—particularly when it comes to the use of online tools. Across six
different methods of mediated communication, there is not one mode through which a majority of
the American public feels “very secure” when sharing private information with another trusted
person or organization:






81% feel “not very” or
The public feels most secure using landline phones,
“not at all secure” using
least secure on social media
social media sites when
% of adults who feel varying degrees of security when sharing private info
they want to share
with another trusted person or organization
private information with
another trusted person
or organization.
68% feel insecure using
chat or instant messages
to share private
information.
58% feel insecure
sending private info via
text messages.
57% feel insecure
sending private
information via email.
46% feel “not very” or
“not at all secure” calling
on their cell phone when
Source: Pew Research Privacy Panel Survey, January 2014. N=607 adults, ages 18 and
they want to share
older.
private information.
PEW RESEARCH CENTER
31% feel “not very” or
“not at all secure” using a landline phone when they want to share private information.

d
we

e

, vi
659

015
3/2
8/0

0

4







12.
No

Americans’ lack of confidence in core communications channels tracks closely with how much they
have heard about government surveillance programs. For five out of the six communications
channels we asked about, those who have heard “a lot” about government surveillance are
significantly more likely than those who have heard just “a little” or “nothing at all” to consider the

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 6 of 57

5
PEW RESEARCH CENTER

method to be “not at all secure” for sharing private information with another trusted person or
organization.

Most say they want to do more to protect their privacy, but many believe it is not possible
to be anonymous online
When it comes to their own role in managing the personal information they feel is sensitive, most
adults express a desire to take additional steps to protect their data online: When asked if they feel
as though their own efforts to protect the privacy of their personal information online are
sufficient, 61% say they feel they “would like to do more,” while 37% say they "already do enough."
When they want to have anonymity online, few feel that is easy to achieve. Just 24% of adults
“agree” or “strongly agree” with the statement: “It is easy for me to be anonymous when I am
online.”

015
3/2
Some people are more anxious than others to keep track of their /0
08 online reputation. Adults under
the age of 50 are far more likely to be “self-searchers” than those ages 50 and older, and adults
d
we
with higher levels of household income and education stand out as especially likely to check up on
e
their own digital footprints.
, vi
659
4
 62% adults have ever used a search engine to look up their own name or see what
12.
information No them is on the internet.
about
 47% say they generally assume that people they meet will search for information about
Not everyone monitors their online reputation very vigilantly, even though many assume
others will check up on their digital footprints



them on the internet, while 50% do not.
However, just 6% of adults have set up some sort of automatic alert to notify them when
their name is mentioned in a news story, blog, or elsewhere online.

Context matters as people decide whether to disclose information or not
One of the ways that people cope with the challenges to their privacy online is to employ multiple
strategies for managing identity and reputation across different networks and transactions. As
previous findings from the Pew Research Center have suggested, users bounce back and forth
between different levels of disclosure depending on the context. This survey also finds that when
people post comments, questions or other information, they do so using a range of identifiers—
using a screen name, their actual name, or posting anonymously.
Among all adults:

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 7 of 57

6
PEW RESEARCH CENTER





59% have posted comments, questions or other information online using a user name or
screen name that people associate with them.
55% have done so using their real name.
42% have done so anonymously.

In some cases, the choices people make about disclosure may be tied to work-related policies.
Among employed adults:



24% of employed adults say that their employer has rules or guidelines about how they are
allowed to present themselves online.
11% say that their job requires them to promote themselves through social media or other
online tools.

Different types of information elicit different levels of sensitivity among Americans

015
3/2
8/0

Social security numbers are universally considered to be the most sensitive piece of personal
information, while media tastes and purchasing habits are among the least sensitive categories of
data.

d
we

e

, vi
659

4

12.
No

www.pewresearch.org

0

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 8 of 57

7
PEW RESEARCH CENTER

Social security numbers, health info and phone conversations among the most
sensitive data
% of adults who report varying levels of sensitivity about the following kinds of info

d
we

e

, vi
659

015
3/2
8/0

0

4

12.
No

Source: Pew Research Privacy Panel Survey, January 2014. N=607 adults, ages 18 and older.
PEW RESEARCH CENTER

At the same time that Americans express these broad sensitivities toward various kinds of
information, they are actively engaged in negotiating the benefits and risks of sharing this data in
their daily interactions with friends, family, co-workers, businesses and government. And even as
they feel concerned about the possibility of misinformation circulating online, relatively few report
negative experiences tied to their digital footprints.


11% of adults say they have had any bad experiences because embarrassing or inaccurate
information was posted about them online.



16% say they have asked someone to remove or correct information about them that was
posted online.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 9 of 57

8
PEW RESEARCH CENTER

About this survey
The analysis in this report is based on a survey conducted January 10-27, 2014 among a sample of
607 adults, 18 years of age or older. The survey was conducted by the GfK Group using
KnowledgePanel, its nationally representative online research panel. GfK selected a representative
sample of 1,537 English-speaking panelists to invite to join the subpanel and take the first survey.
Of the 935 panelists who responded to the invitation (60.8%), 607 agreed to join the subpanel and
subsequently completed the first survey (64.9%). This group has agreed to take four online surveys
about “current issues, some of which relate to technology” over the course of a year and possibly
participate in one or more 45-60-minute online focus group chat sessions. A random subset of the
subpanel receive occasional invitations to participate in these online focus groups. For this report,
a total of 26 panelists participated in one of three online focus groups conducted during August
2013 and March 2014. Sampling error for the total sample of 607 respondents is plus or minus 4.6
percentage points at the 95% level of confidence.1

015
3/2
8/0

For more information on the GfK Privacy Panel, please see the Methods section at the end of this
report.

d
we

e

, vi
659

0

4

12.
No

The original margin of error provided by the vendor and published here (3.98) used a less conservative
design effect. However, the analysis in the report was based on the more conservative design effect and the
significance of the findings is not affected by this change.

1

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 10 of 57

9
PEW RESEARCH CENTER

Table of Contents
About this Report

1

About the Pew Research Center

1

Summary of Findings

2

Acknowledgements

10

Introduction

11

Americans’ Perceptions of Privacy are Varied

14

Public Confidence in the Security of Core Communications Channels is Low

23

Few Feel that the Government or Advertisers can be Trusted

28

015
Most Would Like to do More to Protect their Personal Information Online
3/2
0
Appendix—Detailed Demographic Tables
08/
d
Methods
we
e
, vi
659
4
12.
No
Americans Consider Certain Kinds of Data to be More Sensitive than Others

www.pewresearch.org

31
40
46
55

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 11 of 57

10
PEW RESEARCH CENTER

Acknowledgements
The authors would like to acknowledge the generous contributions of the various outside reviewers
who offered their insights at various stages of this project. In particular, we would like to thank:
Tiffany Barrett, danah boyd, Mary Culnan and all of the attendees of the Future of Privacy Forum
Research Seminar Series, Urs Gasser, Chris Hoofnagle, Michael Kaiser, Kirsten Martin and Katie
Shilton. In addition, the authors are grateful for the ongoing editorial, methodological and
production-related support provided by the staff of the Pew Research Center.
While we greatly appreciate all of these contributions, the authors alone bear responsibility for the
presentation of these findings, as well as any omissions or errors.

d
we

e

, vi
659

4

12.
No

www.pewresearch.org

0

015
3/2
8/0

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 12 of 57

11
PEW RESEARCH CENTER

Introduction
In the wake of Edward Snowden’s 2013 revelations of widespread government surveillance of
Americans’ phone and email records, various news organizations have explored the idea that
“privacy is dead.” At the same time, others have suggested that renewed public attention to the
topic has meant that privacy is, in fact, “thriving online,” or at the very least, “not dead yet.” Some
of this disconnect is attributable to the longstanding tendency of privacy debates to frame the issue
in binary terms; either people have privacy or they don’t. But in practice, some scholars and
analysts have argued that privacy is not something that one simply has, but “something that one
seeks to achieve,” through ongoing negotiation of new contexts and changes in the way
information flows.
What is perhaps a less disputed notion is that privacy is being discussed with new urgency in
recent years in America. Between June 2013 and June 2014, there were nearly 1,000 English
language news articles that included the word “privacy,”2 and 395 current pieces of legislation in
the 113th Congress mentioned the term.3 Internet users also took to social media to talk about the
topic; during the same one-year period, the word “privacy” was included in 3,783,091 tweets.4

d
we

015
3/2
8/0

0

The privacy-focused media coverage, policy debates and the social media discussion has been
wide-ranging. While the revelations by Edward Snowden about the National Security Agency
(NSA) surveillance programs have drawn considerable attention, there have also been multiple
large-scale consumer data breaches, White House-commissioned reports on “big data and
privacy,” and an ongoing controversy about the privacy of students’ educational data to drive
public interest in and discussion of privacy.

e

, vi
659

4

12.
No

The urgency of these privacy-related discussions has increased as policymakers have proposed a
number of measures to address government surveillance and commercial data collection. Previous
research by the Pew Research Center’s Internet Project has found that public concern about the
amount of personal information available online is growing and the majority of internet users
think that current privacy laws “are not good enough” in protecting people’s privacy online.
At the same time, however, what exactly “protecting privacy online” means in practice may differ
quite substantially from person to person, in different online contexts and transactions, and in
response to current events. For instance, in the specific context of national security, Americans’
views about the government’s collection of telephone and internet data as part of anti-terrorism
Search conducted via the “English language news” filter on LexisNexis for the period of January 1-June 18, 2014.
Search conducted via www.congress.gov on June 27, 2014. A separate search of Westlaw archives conducted in October 2014 indicated
that 6,932 federal cases had mentioned the word “privacy” in their decisions since June 2013.
4 Analysis conducted using Crimson Hexagon (date range from January 1, 2014-June 18, 2014).
2
3

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 13 of 57

12
PEW RESEARCH CENTER

efforts are divided. Yet there is a clear trend, confirmed in this survey, that Americans’ opinions
have shifted from relatively clear support at the time the Snowden revelations came to light to
relative disapproval.5
Americans’ associations with the topic of privacy are also complicated and changing, particularly
as younger adults approach networked environments with a different social calculus for assessing
the perceived benefits and risks of these spaces.6 And for older adults, the widespread integration
of digital communications technology into nearly every facet of daily life has meant that even those
who are not connected to these networks are still affected by the data that is collected and courses
through the internet.
In an effort to explore a range of questions about Americans’ privacy behaviors and attitudes, the
Pew Research Center’s Internet Project created a custom online panel of adults who agreed to
respond to four surveys and participate in occasional focus groups over the course of a year. Using
a random subsample of the full GfK Knowledge Panel allows researchers to draw connections
between survey responses and focus group discussions as well as the ability to track changes in key
privacy measures over time.

d
we

015
3/2
8/0

0

The first set of findings from this research panel suggests that Americans’ perceptions of privacy
and their sensitivities about different kinds of personal information are varied, but their lack of
confidence in the security of digital communications channels is universal. Among the general
public, there is not a high level confidence in the security of everyday communications channels—
particularly when it comes to the use of online tools. Across six different methods of mediated
communication, there is not one mode through which a majority of the American public feels “very
secure” sharing private information with another trusted person or organization.

e

, vi
659

4

12.
No

When they have to make a choice, the public feels most secure communicating private information
via calls placed on a landline telephone or cell phone. But text messages and email are not as
widely trusted. And social media sites, chat and instant messenger applications are rarely
considered “very secure” means of communicating sensitive information to another trusted person
or organization.
Few members of our online panel expressed high levels of trust in the government or advertisers,
and most panelists are at least somewhat concerned about those entities accessing their
information on social networking sites—particularly advertisers. Most adults have heard about the
See: http://www.people-press.org/2014/01/20/obamas-nsa-speech-has-little-impact-on-skeptical-public/
See: danah boyd, It’s Complicated and Madden et al., “Teens, Social Media and Privacy”: http://www.pewinternet.org/2013/05/21/teenssocial-media-and-privacy/
5
6

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 14 of 57

13
PEW RESEARCH CENTER

disclosures of government surveillance of communications, and a majority believes that online
surveillance is not good for society. Those who have heard the most about the government
disclosures and those who have checked up on their own digital footprints online are more privacy
sensitive across an array of measures in the survey.

d
we

e

, vi
659

4

12.
No

www.pewresearch.org

0

015
3/2
8/0

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 15 of 57

14
PEW RE
ESEARCH CENT
TER

Americ
cans’ Pe
erceptio of Privacy a Varied
ons P
are
The term “privacy” evokes a constellation of concept in the min of the A
e
ts
nds
American pu
ublic
To better understand how the pub thinks ab
u
blic
bout privacy a represen
y,
ntative sample of 607 adu
ults
were asked an open-en
nded questio in an onli survey: “
on
ine
“When you h
hear the wor “privacy,” what
rd
”
7 The resp
comes to mind for you
m
u?”
ponses that followed wer striking in their varia
f
re
n
ance, ranging from
g
one-word entries to le
engthier desc
criptions tha touched o n multiple c
at
concepts.

d
we

e

, vi
659

015
3/2
8/0

0

4
12.
No coded, a set of key words and themes emerg as the mo frequenttly
Once the responses were
r
w
ged
ost
referenced and top-of
d
f-mind for th general pu
he
ublic. Each o the top ten themes wa referenced in at
of
n
as
d
least 5% of the total re
o
esponses. However, a fu 22% of th responses referenced s
H
ull
he
some other t
theme
that was mentioned on a handfu of times or was entirel unique.
m
nly
ul
r
ly

7

A full discuss
sion of the sample is available at th end of this report.
e
he

www.p
pewresearch.or
rg

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 16 of 57

15
PEW RESEARCH CENTER

“Privacy” in the words of the public
% that referenced each theme in response to open-ended
question: “When you hear the word “privacy,” what comes
to mind for you? Tell us the first few words that pop into
your head…”

Security/safety/protection
Personal
Secret/hidden
Rights/Let Alone/4th Amendment
My business/stay out
Don't have/doesn't exist
Information
Internet/technology
Keep to self/no one else
NSA/Snowden
Private/Privacy of something
Confidential
Surveillance, spying, tracking
Freedom/liberty
Home/own space
Not sharing
Invasion/violation/intrusion
Controlling/limiting
Not for public
Government
Trust
Other*

d
we

e

, vi
659

4

12.
No

14%
12%
11%
10%
9%
9%
8%
7%
7%
5%
5%
4%
4%
3%
3%
3%
3%
2%
2%
2%
2%
22%

0

015
3/2
8/0

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. Table displays responses to the question:
“When you hear the word, ‘privacy,’ what comes to mind for you?” A
portion of the sample did not respond, leaving 589 responses that were
coded by themes. Results do not add up to 100% because many
responses referenced multiple themes.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 17 of 57

16
PEW RESEARCH CENTER

A large segment of the responses associated privacy with concepts of security, safety and
protection
For many Americans, privacy is closely associated with references to security.8 Even as “privacy”
and “security” signal distinct sectors of technological development and legal protections, these
concepts are often blurred and overlapping for the general public. Among all of the themes
referenced in the open-ended responses to the online survey, security, safety and protection was
the most frequently-referenced category; 14% of the responses used these phrases in some form.
Respondents associated privacy with the “security of personal information” or as something that
“must be protected.” And among the most common one-word responses were simply the words
“secure” and “security.”
In online focus groups, smaller groups of respondents from the survey were asked specifically
about the way they think about privacy versus security online. In many cases, respondents viewed
the terms as interchangeable:

015
3/2
8/0

Q: Is there any difference in the way you think about privacy and security
online?

e

, vi
659

“I think it's pretty much the same.”

d
we

0

4
12-much the same thing.”
.
“Not to me, that is pretty
No
“I see them as the same.”

“Pretty much go hand [in] hand.”
However, some participants viewed the concepts as more distinct, with security signaling issues
around personal safety, financial matters and protection from external threats online:
“Privacy is keeping something from someone, security is having the confidence that
things or you [are] going to be ok.”
“In my mind, privacy deals more on the side of personal issues while security deals with
financial issues.”
This association is often communicated in various privacy policies directed at consumers with regard to data security. However, it is also
worth noting here that a different concept of security may be evoked by the language of the Fourth Amendment, which emphasizes the “right
of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” Specific references to
the Fourth Amendment were coded separately, but some references to being “secure” could be overlapping in some cases.
8

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 18 of 57

17
PEW RESEARCH CENTER

“Security to me means a firewall, a secure sight and a good filter on your computer.
Privacy is more like photos, and personal info.”

Privacy also signals a range of things that are considered to be personal
As with the focus group discussions, a slightly smaller portion of the survey responses (12%) used
some variation of the word, “personal.” While online survey respondents most often used the term
in the context of “personal information,” they also described privacy as personal in many other
combinations, such as: “my personal business,” “personal life,” personal space,” “personal stuff”
“my personal solitude,” and a “personal right.”

Many respondents associated privacy with the ability to keep some things secret or hidden
About one in ten (11%) responses included the word “secret” or some variation of things that are
hidden. Respondents described privacy as: “keeping secret,” “secret, private, for your eyes only
type of thing,” or as things that are “protected, secret, concealed.” Other responses suggested
privacy as connected to having a “Hidden agenda” or things that are “secret, undercover.”

015
/2
3were clustered around
0
Other common themes that emerged from the open-ended responses
08/
privacy as:
d
we
e
 A set of rights, such as the “right to be let alone” (10%).
, vi
9
 Others “staying out of my 65
business” (9%).
4
 Something people “don’t have” or “doesn’t exist” (9%).
12.
 Associated with information and the ability to control and limit access to it (8%).
No and technology (7%).
 Tied to the internet



Things people want to keep to themselves and no one else (7%).
Associated with references to the National Security Agency (NSA) and Edward
Snowden (5%).

Most have heard at least a little about government surveillance
Beyond specific references to government surveillance programs in the adults’ associations with
the word “privacy,” almost all of the participants in our online panel said that they have heard at
least something about “the government collecting information about telephone calls, emails, and
other online communications as part of efforts to monitor terrorist activity.” And those who have

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 19 of 57

18
PEW RESEARCH CENTER

heard the most about the government disclosures are more privacy sensitive across an array of
measures in the survey.9
Some 43% of adults have heard “a lot” about this government
surveillance, and another 44% had heard “a little.” Just 5% of
adults in our panel said they had heard “nothing at all” about
these programs.

Most have heard at least
a little about government
surveillance
Among all adults, the % who say they
have heard “a lot,” “a little” or
“nothing at all” about gov’t collecting
info

A lot

43

A little

44

015
2

Nothing at all

9, v
65

d

e
ew
i

/
/03
08

5

Source: Pew Research Privacy Panel Survey,
January 2014. N=607 adults, ages 18 and
older.
PEW RESEARCH CENTER

4

12.
No

Other recent surveys have found correlations between privacy-related awareness and concern. See, Chris Jay Hoofnagle and Jennifer M.
Urban’s discussion in “Alan Westin’s Privacy Homo Economicus,” available at:
http://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?article=3399&context=facpubs

9

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 20 of 57

19
PEW RESEARCH CENTER

Looking at demographics, we find that men were
Older adults among most likely to have
much more likely than women to say they have
heard “a lot” about gov’t surveillance
heard a lot about the NSA revelations (50% vs.
36%), and those ages 65 and older were more
likely than younger age groups to have heard a
lot (57% vs. 37% of those under age 50).10 Adults
with higher levels of education and household
income were also more likely to report hearing a
lot compared with those who have lower levels of
education.

d
we

e

, vi
659

015
3/2
8/0

0

4

12.
No

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

Adults ages 65 and older are also more likely to keep up with news in general: http://www.people-press.org/2012/09/27/section-1watching-reading-and-listening-to-the-news-3/
10

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 21 of 57

20
PEW RESEARCH CENTER

Those who have heard “a lot” about government surveillance programs are also more
aware of their own digital footprints
A majority of adults say that
Surveillance awareness and self-searching
they keep track of their digital
% of adults who use online search engines to look up their own name
footprints, but those who have
a high level of awareness about
government surveillance are
Total
62
more likely to say they search
for information about
Heard a lot
themselves online. Overall, six
about gov't
71
surveillance
in ten (62%) of those who
Heard a little or
participated in our online panel
nothing about
57
have ever used a search engine
gov't surveillance
to look up their own name or
see what information about
Source: Pew Research Privacy Panel Survey, January 2014. N=607 adults, ages 18 and
older.
11 Those
them is on the internet.
PEW RESEARCH CENTER
who have heard a lot about
government surveillance of communications are more likely to be self-searchers than those who
have heard a little or nothing about it (71% vs 57%).12

d
we

e

, vi
659

015
3/2
8/0

0

4

12.
No

11 In May 2013, an RDD telephone survey of adults found that 56% of internet users had used a search engine to look up their own name and
see what information is available about them online: http://www.pewresearch.org/fact-tank/2013/09/27/majority-of-online-americansgoogle-themselves/
12 While self-searching activity is associated with several measures of increased privacy-related sensitivity throughout the survey, it is also
worth noting that self-searching can be one way to link IP addresses to individual users.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 22 of 57

21
PEW RESEARCH CENTER

Self-searching activity varies greatly across
different groups, particularly by age, income,
and household education. Adults under the age
of 50 are far more likely to be “self-searchers”
than those ages 50 and older, and adults with
higher levels of household income and education
stand out as especially likely to check up on their
own digital footprints.

A majority of adults have searched for
information about themselves online
Total
Age
18-29

54

65+

49

Education
No college

44

Some college

70

College graduate

79

Household income
Less than $30,000/year
$30,000-$49,999

/
/03
8

$50,000-$74,999
At least $75,000

d
we

e

4

71

50-64

A majority of adults (62%) disagree with the
statement “It is a good thing for society if people
believe that someone is keeping an eye on the
things that they do online,” including 20% who
“strongly disagree.” Another 36% do agree that
online surveillance is good for society, including
the 7% who say they “strongly agree.”

12.
No

69

30-49

Few feel it’s a “good thing” for society if
people believe they are being watched online

, vi
659

62%

0

015
2

44
60
61
73

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 23 of 57

22
PEW RESEARCH CENTER

Attitudes about online surveillance vary greatly
among different groups, particularly by age and
education. For instance, adults ages 50 and
older are generally less likely than younger
adults to see online surveillance as beneficial.
Those with lower levels of education are also
more likely to be in favor of online surveillance,
with 45% of those who have not attended
college agreeing overall—compared with 33% of
those with some college experience and 26% of
college graduates.

Most do not think it’s a good thing for
society if people believe they are being
watched online
Among all adults, the % who agree it is a good thing for
society if people believe that someone is keeping an eye
on the things that they do online

Strongly agree

7

Agree

29

Disagree
Strongly disagree

42
20

Finally, adults who have heard more about
government surveillance are more likely to
Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.
think such oversight could have drawbacks:
PEW RESEARCH CENTER
Just 23% of adults who have heard “a lot” about
the NSA revelations think online surveillance is
good for society, compared with 46% of those who have heard less about the NSA revelations.

d
we

e

, vi
659

015
3/2
8/0

0

Most Americans agree that citizens “should be concerned” about the government’s
monitoring programs

4

12.
No

Close to eight in ten (80%) American adults “agree” or “strongly agree” that Americans should be
concerned about the government’s monitoring of phone calls and internet communications. Just
18% “disagree” or “strongly disagree” with that notion.
Overall, 40% “strongly agree” that American citizens should be concerned, while 39% “agree.” Men
are more likely than women to “strongly agree” that the monitoring programs are cause for
concern (46% vs. 35%). However, there are no significant variations by age, income or education
levels.
Those who have heard “a lot” about government surveillance programs are considerably more
likely to hold strong views; 53% “strongly agree” that citizens should be concerned, compared with
33% of those who have heard only a little or nothing about the programs.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 24 of 57

23
PEW RESEARCH CENTER

Public Confidence in the Security of Core Communications
Channels is Low
Across the board, there is a universal lack of confidence in the security of everyday
communications channels—particularly when it comes to the use of online tools. Across six
different methods of mediated communication, there is not one mode through which a majority of
the American public feels “very secure” sharing private information with another trusted person or
organization.
Americans’ lack of confidence in core communications channels tracks closely with how much they
have heard about government surveillance programs. For five out of the six communications
channels we asked about, those who have heard “a lot” about government surveillance are
significantly more likely than those who have heard just “a little” or “nothing at all” to consider the
method to be “not at all secure” for sharing private information with another trusted person or
organization.

Landline phones

d
we

015
3/2
8/0

0

While the use of landline phones has generally been declining and the number of cell-only
households has been increasing, Americans express the greatest sense of security using landline
phones when sharing private information with another trusted entity. More than any other device
or communications tool, landline phones inspire the most confidence for these kinds of exchanges.

e

, vi
659

4
12-is still quite low. Just 16% of adults say they feel “very secure”
.
However, the level of confidence
No via the landline phone, while 51% say they feel “somewhat secure.”
sharing private information

About one in five (19%) say they feel “not very secure,” while 12% say they feel “not at all secure”
calling on a landline phone.
Women and men express the same low level of confidence in the security of landline phones.
Those in the highest income group—adults living in households earning $75,000 or more per
year—are more likely to report that they feel "very secure" using a landline phone when compared
with those living in households earning less than $30,000 annually (20% vs. 11%).
Yet, there are no significant differences by age—people across all generations are equally likely to
report low levels of confidence in using their landline when they want to share private information
with another trusted person or organization.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 25 of 57

24
PEW RESEARCH CENTER

Those who have heard a lot
about government surveillance
programs are twice as likely as
those who have heard a little or
nothing to say that landlines are
“not at all secure” (16% vs. 8%).

The public feels most secure using landline phones,
least secure on social media
% of adults who feel varying degrees of security when sharing private info
with another trusted person or organization

Cell phone calls
Calling on a cell phone ranks
below landline calls, but
considerably higher than online
communications. Among all
adults, less than one in ten (9%)
say they feel “very secure”
making a call on their mobile
phone when they have private
information to share with
another trusted person or
organization. Another 43% say
they feel “somewhat secure,”
while 29% say they feel “not very
secure.” Just over one in ten
(17%) say they feel “not at all
secure.”

d
we

e

, vi
659

015
3/2
8/0

0

Source: Pew Research Privacy Panel Survey, January 2014. N=607 adults, ages 18 and
older.

4

12.
No

PEW RESEARCH CENTER

Those with lower levels of education are among the least confident in the security of cell phone
communications. Among those who have not attended college, 21% consider calling on their cell
phone to be “not at all secure,” compared with 12% of those with college degrees.
Cell phone users across all age groups and both genders report similar levels of confidence in the
security of calls placed on their mobile phones.
Among adults who have heard a lot about government surveillance programs, 24% feel as though
cell phones are “not at all secure” when there’s a need to share private information with another
trusted person or organization. Half as many (12%) who have heard just a little or nothing about
government surveillance programs feel this strongly.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 26 of 57

25
PEW RESEARCH CENTER

Text messages
Relative to calls placed on a cell phone, text messages are viewed with more uncertainty; 7% of
adults say they feel “very secure” sharing private information with another trusted person or
organization via text message, while 32% feel “somewhat secure.” Another 37% feel as though text
messages are “not very secure,” while another (22%) feel as though texting is “not at all secure.”
Women and men express the same lack of confidence in the security of text messages to share
private information.
Adults ages 50 and older are significantly more likely than young adults 18-29 to rate text
messages as “not at all secure.” One in four (26%) adults over age 50 say this compared with 13%
of those ages 18-29. While there are some minor variations in responses across socioeconomic
groups, they are not consistent enough to suggest a clear pattern.

015
3/2
8/0

However, those who have heard a lot about government surveillance are twice as likely as those
who have heard only a little or nothing at all to say that they feel “not at all secure” sending private
information via text messages (30% vs. 15%).

d
we

e

Email

, vi
659

0

The public’s confidence in the security of email is on par with the way they feel about texting; only
5% of adults say they feel “very secure” sending private information via email, while 35% say they
feel “somewhat secure.” Another 36% feel that email is “not very secure” when they have private
information to share, and 21% say they feel “not at all secure” when using email.

4

12.
No

Men and women report similar views of email, and adults of all ages generally express the same
levels of confidence in the security of email. While there are no notable variations by education
levels, adults living in the lowest income households are among the most wary of email security. A
full 30% of those living in households earning less than $30,000 annually say that email is “not at
all secure,” while 18% of those in households earning $75,000 or more per year reported the same.
Among adults who have heard a lot about government surveillance programs, 26% feel as though
email is “not at all secure” when they want to share private information with another trusted
person or organization, while 16% who have heard a little or nothing about government
surveillance programs feel this way.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 27 of 57

26
PEW RESEARCH CENTER

Those who are more aware of gov’t surveillance more likely to see communications
channels as “not at all secure”
Among those who have heard “a lot” vs. “a little or nothing” about government surveillance programs, the % who feel
“not at all secure” sharing private info with another trusted person or organization via the following methods…
Heard a lot about surveillance

Heard a little/nothing about surveillance

61

50

36
30

30
26
16

15

. 12
No

0
ed

w

e
, vi
9

5
-46

Using social media Using chat or
to communicate instant messenger

Sending texts

015
3/2
8/0
24

Sending email

16

12

Calling on cell
phone

8

Using landline
phone

Source: Pew Research Privacy Panel Survey, January 2014. N=607 adults, ages 18 and older.
PEW RESEARCH CENTER

Chat or Instant Messenger
Few Americans view chat or instant messenger as a secure means to share private information;
only 4% feel “very secure” using chat and IM, and 25% feel these channels are “somewhat secure.”
Another 36% feel “not very secure” and 32% say they feel “not at all secure” using chat or instant
messenger to share private information.
The most notable variations across various demographic groups for this question are by age;
internet users over the age of 50 (38%) are more likely to see chat and IM as “not at all secure”
when compared with young adults ages 18-29 (28%).

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 28 of 57

27
PEW RESEARCH CENTER

Unlike other communications channels, the proportion of Americans who view chat and instant
messenger as insecure places to share private information does not vary according to their level of
awareness of government surveillance programs.

Social Media
American adults view social media sites as the least secure channel to communicate private
information to another trusted person or organization; just 2% view them as “very secure,” while
14% feel “somewhat secure” sharing sensitive information on social media. Just over one in four
(28%) feel “not very secure” sharing private information on social media and 53% say they feel
“not at all secure.”
Men and women are equally wary of using social media sites to share private information, but
those over the age of 30 are more likely than young adults to feel as though the sites are “not at all
secure.” Over half of adults ages 30 and older (57%) report this lack of confidence, compared with
37% of those ages 18-29.

015
3/2
In terms of education-related differences, college grads are generally less confident in the use of
/0
social media for sharing private information; 64% of adults 08 have a college degree feel “not at
who
d
all secure” sharing sensitive information over social media channels, compared with 47% of those
we
e
who have not attended college.
, vi
659 surveillance are more likely to view social media
- government
Those who have heard a lot about4
12compared with those who have lower levels of awareness (61% vs.
sites as “not at all secure”.when
No
50%).

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 29 of 57

28
PEW RESEARCH CENTER

Few Feel that the Government or Advertisers can be Trusted
Americans’ lack of confidence in communications security is accompanied by low levels of
trust in government and advertisers
The backdrop for this lack of confidence in communications security is low levels of trust in key
sectors that Americans associate with data collection and monitoring. While the surveillance
practices of government agencies have been the focus of many public discussions and debates
post-Snowden, previous research from the Pew Research Center’s Internet Project has suggested
that Americans are also concerned about data collection by advertisers. Additionally, public
concern over the amount of personal information businesses are collecting has been growing.
Since around 2010, levels of trust in the federal government have been at historic lows. Previous
Pew Research data shows that trust in the government has been declining since the early 2000s
(immediately after 9-11) to its current post-recession level; trust in government has not been this
low since the summer of 1994.13

015
Currently, just 18% of adults believe that the government in Washington can be trusted to do what
3/2
0
is right either all (2%) or most (16%) of the time, while another 67% say that the government can
08/
d
be trusted only some of the time and 14% say it can never be trusted. As has been the case in other
we as adults who did not attend college
e
surveys, there are significant differences by education level,
, vi be trusted (21%) compared with those with at
are more likely to say the government 59
6can “never”
4
least some college experience (9%).
12.
No

For more information, see “Trust in Government Nears Record Low, But Most Federal Agencies Are Viewed Favorably” http://www.peoplepress.org/2013/10/18/trust-in-government-nears-record-low-but-most-federal-agencies-are-viewed-favorably/ and its accompanying
interactive chart: http://www.people-press.org/2013/10/18/trust-in-government-interactive/
13

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 30 of 57

29
PEW RESEARCH CENTER

However, Americans do not trust advertisers any
more than they do the federal government. Few
respondents believe that advertisers can be
trusted to do what is right just about all (1%) or
most (11%) of the time; 71% say that advertisers
can be trusted only some of the time, and 16%
say they can never be trusted. There are few
differences in adults’ trust levels toward
advertisers by demographic group, with the
exception of household income: 27% of those
living in lower-income households (earning less
than $30,000 per year) say that advertisers can
“never” be trusted, making them more likely to
say this than those living in higher income
households.

Low levels of trust in government and
advertisers
Among adults ages 18 and older
How often do you trust ___ to do the right thing?

Just about always

Government
2%

Advertisers
1%

Most of the time

16

11

Only some of the time

67

71

Never

14

16

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.
PEW RESEARCH CENTER

015
Among the 64% of adults who use social
3/2
Broad concern about government and
0
networking sites, we find that seven in ten say
third-party access to info on social
08/
d
(70%) say that they are at least somewhat
networking sites
we
concerned about the government accessing some
e
, vi
of the information they share on those sites
659
without their knowledge; this includes 37% who
4
12say they are “very concerned,” and 34% say they
.
are “somewhat concerned.” Another 25% say
No
% Among adults ages 18 and older who use social
networking sites

How concerned are you that some of the info you share on
social networking sites might be accessed by ___ without
your knowledge?

they are “not too concerned” about the
government accessing their information, and just
4% say they are “not at all concerned.”

The
government

Third parties like
advertisers or
businesses

Very concerned

37

35

Somewhat concerned

34

45

Similarly, 80% of social networking site users on
Not too concerned
25
17
the panel say that they are at least somewhat
Not at all concerned
4
2
concerned about third parties such as advertisers
Source: Pew Research Privacy Panel Survey, January 2014. N=607
or businesses accessing some of the information adults, ages 18 and older.
PEW RESEARCH CENTER
they share on those sites without their
knowledge; this includes 35% who say they are
“very concerned” and 45% who say they are “somewhat concerned.” Another 17% say they are “not
too concerned,” and just 2% say they are “not at all concerned.” In contrast to many other
questions in the survey, there were only minor variations by age, education, or household income.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 31 of 57

30
PEW RESEARCH CENTER

An overwhelming majority of the American public senses a loss of control over how their
personal information is collected and used by companies
Beyond social networking sites, Americans express a broader loss of control over the way their
personal data is managed by companies. Fully 91% of adults “agree” or “strongly agree” that
“consumers have lost control over how personal information is collected and used by companies.”
This includes 45% who “strongly agree” and 46% who “agree” that consumers have lost control.
Another 6% “disagree,” while only 1% “strongly disagree” with this sentiment.
Those with a college education are more likely than those who have not attended college to
“strongly agree” that consumers have lost control, 51% vs. 40%.
Respondents who are more aware of government surveillance programs also express a greater loss
of control over how their personal information is collected and used by companies. Those who said
they had heard a lot were more likely to “strongly agree” with a loss of control over their personal
information compared with those who had heard “a little” or “nothing” about surveillance
programs (58% vs. 37%).

d
we

015
3/2
8/0

0

Most Americans support greater regulation of advertisers and the way they handle
personal information

e

, vi
659

Even as Americans express concern about government access to their data, they feel as though
government could do more to regulate what advertisers do with their personal information; 64%
believe the government should do more to regulate advertisers, compared with 34% who think the
government should not get more involved.

4

12.
No

Support for more regulation of advertisers is consistent across an array of demographic groups.
However, those who have a college education are more likely than those who have not attended
college to support more government intervention (69% vs. 58%).

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 32 of 57

31
PEW RESEARCH CENTER

Americans Consider Certain Kinds of Data to be More
Sensitive than Others
Different kinds of information evoke varying levels of sensitivity among Americans. Social security
numbers are universally considered to be the most sensitive piece of personal information, while
media tastes and purchasing habits are among the least sensitive categories of data. In general,
about half of respondents view the content of phone conversations, email messages and text
messages as “very sensitive,” and one in four see that data as “somewhat sensitive.”
There are various demographic patterns that are echoed across many of the responses:



Those who have higher levels of income and education report greater sensitivity for nearly
every kind of data included in the survey.
Those who have heard a lot about government surveillance programs and those who have
searched for information connected to their names online report higher sensitivity levels
for most categories of information. These differences are especially notable when looking at
the sensitivity of phone conversations and email messages.

d
we

015
3/2
8/0

0

Social security numbers are considered to be the most sensitive piece of personal
information—by far

e

, vi
659

A full 90% of adults feel as though their social security number is a “very sensitive” piece of
information, and this view is broadly held across all demographic groups. Another 5% consider
their social security number to be “somewhat sensitive,” while 2% say it is “not too sensitive.” Only
1% of respondents say their social security number is “not at all sensitive.”

4

12.
No

Those who have some college education are somewhat more likely than those who have not
attended college to say that their social security number is “very sensitive” (94% vs. 85%).
Similarly, those living in the highest income households also report higher sensitivity levels; for
instance, 95% of those living in households earning $75,000 or more per year view their social
security number as “very sensitive,” compared with 83% of those living in households earning less
than $50,000 annually.
Among those who have heard a lot about government surveillance programs, 95% see their social
security number as very sensitive, compared with 89% of those who have heard only a little or
nothing at all about the programs. Those who have searched for information about themselves
online also express greater sensitivity about their social security number relative to those who
haven’t checked up on their own digital footprints (94% vs. 85%).

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 33 of 57

32
PEW RESEARCH CENTER

Social security numbers, health info and phone conversations among the most
sensitive data
% of adults who report varying levels of sensitivity about the following kinds of info

d
we

e

, vi
659

015
3/2
8/0

0

4

12.
No

Source: Pew Research Privacy Panel Survey, January 2014. N=607 adults, ages 18 and older.
PEW RESEARCH CENTER

A majority also consider health information to be very sensitive
More than half (55%) of adults consider the state of their health and the medicines they take to be
“very sensitive” information while 26% view that kind of information to be “somewhat sensitive.”
Another 12% say that health information is “not too sensitive,” and 5% say it is “not at all
sensitive.”
Men and women rank the sensitivity of health information equally. And adults of all ages are just
as likely to say that their health data is “very sensitive.” In general, those who have higher levels of
education and income report a greater level of sensitivity with regard to the details of their health
information.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 34 of 57

33
PEW RESEARCH CENTER

Those who have heard a lot about government surveillance programs are especially likely to say
their health information is “very sensitive” (65%), along with those who check up on their digital
footprints online (61%)

The content of phone conversations is just as sensitive as health info
Americans generally consider the content of their phone conversations to be just as sensitive as
information about the state of their health and the medicines they take. More than half (54%) say
that phone conversations are “very sensitive,” while 27% consider them to be “somewhat
sensitive.” Another 13% see that content as “not too sensitive” and just 4% consider phone
conversations to be “not at all sensitive.”14
However, unlike the consistent findings among men and women for health information, men are
more likely than women to say that the content of their phone conversations is “very sensitive”
(59% vs. 50%).

015
3/2
8/0

As with many other kinds of information, differences by income, government surveillance
awareness and self-searching activity are significant. Yet, there were no consistent variations by
education or age.

d
we

0

e
, vi
Americans’ sensitivities about the content of their email messages rival their feelings about health
659
4
information and the content of phone conversations. Among all adults, 52% consider the content
12.
of their email messages to be “very sensitive,” while 25% consider that information to be
No 13% say they consider their email content to be “not too sensitive,”
somewhat sensitive. Another
Men and young adults are more concerned about the content of their email messages

and 7% say their messages are “not at all sensitive.”15
However, men express a higher level of sensitivity about their email messages when compared
with women; 57% of men say their email content is “very sensitive,” while 47% of women report
this.
Young adults are more likely than seniors ages 65 and older to consider the content of their email
to be “very sensitive” information (59% vs. 42%). Those with a college education express greater
sensitivities relative to those who have not attended college (60% vs. 45%), and those in the

14 As noted in the previous chapter, just 16% say they feel “very secure” using a landline phone to share private information with another
trusted person or organization and 51% say they feel “somewhat secure” doing this.
15 By comparison, only 5% say they feel “very secure” and 35% feel “somewhat secure” sending email when they have private information to
share with another trusted person or organization.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 35 of 57

34
PEW RESEARCH CENTER

highest-income households are also more likely to consider their email content to be “very
sensitive” when compared with those in the lowest income group (61% vs. 43%).
Those who have heard a lot about government surveillance programs (62%), are especially likely to
say the content of their email messages is “very sensitive” information along with those who have
checked up on their digital footprints online (58%).

Physical location data is seen as more sensitive among the college educated
Half of adults (50%) feel as though the details of their physical location gathered over a period of
time from the GPS on a cell phone is “very sensitive” information. Another 32% consider this data
to be “somewhat sensitive.” Just 11% say that data is “not too sensitive,” and 5% consider location
data to be “not at all sensitive.”
There are no consistent variations by gender or age for this question. However, those with a
college degree are considerably more likely than those who have not attended college to say that
this data is “very sensitive” (55% vs. 44%). Similarly, those in higher income households consider
location data to be more sensitive.

015
3/2
8/0

0
ed awareness and self-searching
Variations by the respondents’ level of governmentw
surveillance
ie in the sample who are mobile internet
activity are also significant. However, the 63% of adults
9, v or other handheld device at least occasionally
users who access the internet on a cell 5
phone, tablet
-46 users to consider location data to be “very sensitive.”
are no more likely than non-mobile internet
. 12
No
Men and women are equally likely to consider text messages as sensitive
Just under half (49%) of adults said they consider the content of their text messages to be “very
sensitive.” About one in four (26%) consider this data to be “somewhat sensitive,” while 13% feel
as though text messages are “not too sensitive.” Only 8% consider the content of their text
messages to be “not at all sensitive.”16
When men report increased sensitivity with regard to the content of their email messages, men
and women have similar sensitivities about the content of their text messages.
However, young adults, who are among the most likely to be fervent users of text messaging,
express higher levels of sensitivity regarding text messages when compared with seniors; 55% of

16 As noted above, 7% say they feel “very secure” and 32% feel “somewhat secure” sending text messages when they have private
information to share with another trusted person or organization.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 36 of 57

35
PEW RESEARCH CENTER

those ages 18-29 say they consider the content of their text messages to be “very sensitive,”
compared with 38% of those ages 65 and older.
Those who have a college education and those who live in households earning at least $75,000 per
year consider text messages to be more sensitive when compared with lower education and income
groups.
Variations by the respondents’ level of government surveillance awareness and self-searching
activity are also significant. Mobile internet users are more likely than non-users to say the content
of their text messages is “very sensitive.”

Records of numbers called or texted are viewed as most sensitive by younger adults
Slightly less than half (45%) of adults said they feel as though the numbers they have called or
texted from their phone is “very sensitive” information. About one in three (30%) consider this
data to be “somewhat sensitive,” while 16% consider the numbers called or texted to be “not too
sensitive.” Only 6% consider a record of these numbers to be “not at all sensitive.”

d
we

015
3/2
8/0

0

Young adults express higher levels of sensitivity regarding this data when compared with seniors;
53% of those ages 18-29 say they view the record of who they have called or texted to be “very
sensitive,” data compared with 36% of those ages 65 and older.

e

, vi
659

4

12.
No

Men and women express the same levels of sensitivity about these numbers. Variations by
education are not significant, but those who live in households earning at least $75,000 per year
consider text messages to be more sensitive when compared with those living in households
earning less than $30,000 per year.
Views on the sensitivity of these numbers vary by the respondents’ level of government
surveillance awareness and self-searching activity. Mobile internet users are more likely than nonusers to say these numbers are is “very sensitive.”

About four in ten see their birth date as very sensitive, with little demographic variation
Adults’ views on the sensitivity of their birth date as a piece of information are most striking in
their consistency across various demographic groups when looking at those who view this piece of
data as “very sensitive.” Overall 41% see their birth date as very sensitive data, a share which
doesn’t vary much across demographic groups.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 37 of 57

36
PEW RESEARCH CENTER

The one variation that does stand out most notably is the respondent’s level of awareness about
government surveillance; 47% of those who have heard about surveillance say that their birth date
is a very sensitive piece of information, compared with 37% who have heard only a little or nothing
at all.
People’s sensitivities about this information do differ by age, income and education when we look
at the other extreme—responses among those who view this information as “not at all sensitive.”
Most notably, 25% of seniors ages 65 and older see their birth date as “not at all sensitive,”
compared with 12% of those ages 50-64 and 11% of those under age 50.

Those with higher levels of income and education are more concerned about details of
their relationship history
Four in ten adults holds the view that their relationship history is “very sensitive” information.
(The survey did not specify the kind of relationship—whether friends, romantic or professional.)
About one in three (31%) consider this information to be “somewhat sensitive,” while 14% consider
relationship history to be “not too sensitive.” Another 12% consider information about their
relationship history to be “not at all sensitive.”

015
3/2
8/0

0
ed groups. However, those with higher
There is little consistent variation across core demographic
w
levels of education and income are generally more likely than other groups to rank their
vie
,
relationship history as “very sensitive”59
information. For instance, 52% of college grads consider
-46
this to be highly sensitive data compared with just 29% of those who have not attended college.
. 12
Similarly, 48% of those in households earning $75,000 or more per year consider relationship
No
history to be “very sensitive” data, compared with only 32% of those living in households earning
less than $30,000 per year.
Views on the sensitivity of these numbers also vary by the respondents’ level of government
surveillance awareness and self-searching activity.

Media tastes and purchasing habits are among the least sensitive categories of
information
Seven of the items we asked about registered as considerably less sensitive when compared with
the range of information discussed in the preceding section. For each of these categories of
information, less than one-third of respondents said they considered the material to be “very
sensitive.” However, there were no items for which a majority registered the material to be “not at
all sensitive.”

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 38 of 57

37
PEW RESEARCH CENTER









27% consider the websites they have visited to be “very sensitive,” while 43% see that
information as “somewhat sensitive.”
24% consider the searches they have made using online search engines to be “very
sensitive,” while 41% see that data as “somewhat sensitive.”
22% consider their religious and spiritual views to be “very sensitive,” while 23% view that
information as “somewhat sensitive.”
22% consider information about who your friends are and what they are like to be “very
sensitive,” while 46% see that data as “somewhat sensitive.”
20% consider their political views and the candidates they support to be “very sensitive,”
while 31% feel it is “somewhat sensitive.”
9% consider information about their media tastes to be “very sensitive,” while 22% see that
information as “somewhat sensitive.”
8% consider data about their basic purchasing habits to be “very sensitive,” while 33% view
it as “somewhat sensitive.”

015
3/2
8/0

The responses for these questions generally follow a familiar pattern; for most items, those who
have higher education and income levels express greater sensitivity when compared with those
who have lower levels of education or income. Respondents’ concerns also differed significantly
according to their level of awareness of government surveillance.

d
we

e

, vi
659

0

About one in three adults say they value the greater efficiency of online services because
of the increased access they have to personal data

4

12.
No

Some companies rely on collecting certain data to provide a more personalized experience for each
individual by customizing advertisements and other features, while others rely on access to data to
support a wide range of security features and product improvements. About a third (36%) of
adults “agree” or “strongly agree” with the statement: “I appreciate that online services are more
efficient because of the increased access they have to my personal data.” However, only 4%
“strongly agree” with that statement. By comparison, almost two-thirds (61%) say they disagree,
including 15% who “strongly disagree.”
Men and women are equally likely to disagree with this statement, but younger adults are
somewhat more likely to value the increased efficiency of online services when compared with
those ages 50-64.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 39 of 57

38
PEW RESEARCH CENTER

Those who access the internet on a mobile
device like a smartphone or tablet were
significantly more likely than non-mobile users
to “agree” or “strongly agree” that they
appreciate the greater efficiency of online
services due to personal data collection.

Mobile internet users more likely to
value efficiency of services that rely on
personal data
Among mobile internet users vs. non-mobile internet
users, the % who agree vs. disagree that online services
are more efficient b/c of increased access to personal
data

Those who have heard a lot about recent
government surveillance are significantly more
likely to “disagree” or “strongly disagree” that
online services are more efficient due to
increased access to personal data. Some 72% of
those who have heard a lot reported this,
compared with 56% of those who have heard a
little or nothing.

Many are willing to share some information
about themselves with companies to use
online services for free

Strongly agree/agree

Strongly disagree/disagree
69

58
41

, vi
659

e

d
we

0

015
3/2
8/0

Mobile internet users

29

Non-mobile internet users

Source: Pew Research Privacy Panel Survey, January 2014. N=607
Even as Americans indicate widespread
adults, ages 18 and older.
concern about the loss of control over their
PEW RESEARCH CENTER
personal information and little enthusiasm for
the ways that access to their data may make
certain online services more efficient, a majority of respondents agree that they are willing to share
some personal data in exchange for access to free online services.17 Some 55% “agree” or “strongly
agree” that that they are willing to share some information about themselves with companies in
order to use online services for free. Another 43% “disagree” or “strongly disagree” with this
statement.

4

12.
No

Adults who access the internet on a mobile device like a smartphone or tablet were particularly
willing to trade some of their personal data for free services; 62% mobile internet users “agree” or
“strongly agree,” compared with 45% of non-mobile internet users.

17 These kinds of tensions between concerns and behaviors are common in privacy-related surveys, and are often referred to as the “privacy
paradox.” While some suggest that this is evidence that consumers’ concerns are invalid, other scholars have argued that this tension may
due to a knowledge gap among consumers and the high social costs of “exiting” various platforms. For a lengthier discussion, see:
https://medium.com/@hoofnagle/exit-voice-and-the-privacy-paradox-662a922ff7c8

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 40 of 57

39
PEW RESEARCH CENTER

Social media users were also particularly likely
to “agree” or “strongly agree” that they are
willing to exchange some of their personal data
for free online services. Some 60% said so,
compared with 46% of those who do not use
social media.

Social media users more comfortable
sharing personal info in exchange for
free services
Among social media users vs. non-social media users,
the % who agree vs. disagree that they are willing to
share some info about themselves in exchange for free
online services

Adults who have searched for themselves online
were also more willing to exchange information
for services; 61% “agree” or “strongly agree”
that they would be willing to share some
personal data for free online services, compared
with 45% of those who have not searched for
themselves.

Strongly agree/agree
60

52
46
39

d
we

e

.
No

, vi
659

4
12-

Strongly disagree/disagree

015
3/2
8/0

0

Social media users

Non-social media users

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 41 of 57

40
PEW RESEARCH CENTER

Most Would Like to Do More to Protect their Personal
Information Online
Six in ten adults feel as though they “would like to do more” to protect the privacy of their
personal information online
When asked if they feel as though their own efforts to protect the privacy of their personal
information online are sufficient, 61% say they feel as though they “would like to do more,” while
37% say they “already do enough.”
Men and women and adults of all ages are equally likely to say they would like to do more to
protect the privacy of their personal information online. However, those with the lowest levels of
education are more likely to express confidence in their efforts while those who are college
educated feel as though they could do more.

015
3/2
8/0

Mobile internet users are significantly more likely than non-mobile internet users to say that they
feel as though they “would like to do more” to protect the privacy of their personal information
online (67% vs. 52%).

0
ed their own digital footprints, 66% say
Among those who have used a search engine to check up on
iew
that they would like to do more, compared , v just 53% of those who have not searched for
with
results connected to their name online.
659
2-4
1express a greater desire to take additional steps when compared with
Similarly, social mediao.
N users
non-users (66% vs. 55%).
Content creators employ multiple strategies for managing their identity when posting
online
One of the ways that people cope with the challenges of managing their privacy online is to employ
multiple strategies for managing identity and reputation across different networks and
transactions online. As previous findings from the Pew Research Internet Project have suggested,
users bounce back and forth between different levels of disclosure depending on the context. This
survey also finds that when people post comments, questions or other information, they do so
using a range of identifiers—using a screen name, using their actual name, as well as doing so
anonymously.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 42 of 57

41
PEW RESEARCH CENTER

Among all adults:




59% have posted comments, questions or other information online using a user name or
screen name that people associate with them
55% have done so using their real name
42% have done so anonymously

Younger adults are generally more likely to post
content online when compared with older
adults, and these younger users also tend to
post under a wide range of identifiers. This is
especially true when it comes to screen names
or user names—some 82% of 18-29 year olds
have posted content online using some type of
screen name that people associate with them.
Along with younger adults, college graduates
are more likely to post content online using a
dedicated screen name or user name (69% have
done so) than are those who have attended but
not graduated from college (57%) or those who
have not attended college (54%). And those
with at least some college experience, or an
annual household income of $50,000 and up,
are more likely to post anonymously online
than are those with lower levels of income or
education.

.
No

% of adults who post comments, queries or information
on the internet using the following…
18-29

30-49

Screen name or
user name

ed

/
/03
08

50-64

015
2

Anonymously

65+
82
64
47
40
66
54
52
49

Real name

w

e
, vi
9

5

-46
12

Strategies for self-presentation online
vary by age

59
48
29
27

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.
PEW RESEARCH CENTER

In addition, social networking site users are more likely than non-SNS users to post content using
each of these identifiers—using their real names (70% vs. 30%), a screen name that people
associate with them (74% vs. 34%), as well as anonymously (48% vs. 31%).

Few adults think it is “easy” for them to be anonymous online
Just 24% of adults “agree” (20%) or “strongly agree” (3%) with the statement: “It is easy for me to
be anonymous when I am online.” By contrast, 74% “disagree” (52%) or “strongly disagree” (22%)
that it is easy for them to be anonymous.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 43 of 57

42
PEW RESEARCH CENTER

Those who have heard a lot about government
collection of personal information were more
likely than those less aware to “strongly
disagree” that it is easy for them to be
anonymous online (30% vs. 16%). Likewise,
those who have searched for themselves online
are less confident about the ease with which one
can be anonymous. A quarter “strongly disagree”
that it is easy, compared with 16% of those who
have not searched for themselves online.

Nearly nine in ten adults agree that if
inaccurate information was posted about
them online, it would be very difficult to
remove

Those who are aware of their own digital
footprints more pessimistic about ease
of anonymity online
Among those who have searched for themselves online
vs. those who have not, the % who “strongly disagree”
that it is easy to be anonymous online

Have not searched for
yourself online

16

Have searched for yourself
online

26

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.

015
3/2
8/0

PEW RESEARCH CENTER
Fully 88% of adults “agree” (49%) or “strongly
agree” (39%) that it would be very difficult to
remove inaccurate information about them online. Just 9% “disagree” and 1% “strongly disagree”
that it would be very difficult to remove.

d
we

0

e
, vi fear the possibility that they would not be able
Even those that think their privacy is well-protected
659(31%) of those who say they do enough to protect their
to correct inaccurate information.4 third
1it2- A be difficult to remove inaccurate information, and another
privacy “strongly agree” that would
o.
Nunsurprisingly, those who feel as though they could do more to protect their
53% “agree.” Perhaps
privacy are more likely to “strongly agree” that removing inaccurate information would be
difficult. But overall, these two groups show relatively similar levels of concern.

Yet, relatively few say they have had any bad experiences because embarrassing or
inaccurate information was posted about them online
About one in ten (11%) of adults say they have had bad experiences due to embarrassing or
inaccurate information that was posted about them on the internet, while 87% say this has never
happened to them. Adults ages 65 and older are less likely to have had a negative experience like
this; for instance, just 2% report having a bad experience due to embarrassing or inaccurate
information, while 16% of young adults ages 18-29 report this.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 44 of 57

43
PEW RESEARCH CENTER

There are no notable variations between men and women or adults across income groups.18
Variations by education are not consistent enough to suggest a clear pattern.
Mobile internet users (13%) are more likely than non-mobile internet users (7%) to report bad
experiences. Similarly, social media users (14%) have a greater tendency to report bad experiences
than non-users (6%). Those who have searched for results connected to their name online, are
also considerably more likely to say they have had bad experiences (16% vs. 2%).

16% of adults say they have asked someone to remove or correct information about them
that was posted online
One in six adults have asked someone to remove
or correct some kind of information—including
things like photos or videos—that was posted
about them on the internet. While men and
women are equally likely to say they have asked
someone to remove information, experiences
vary considerably by age. Nearly one in three
(32%) of young adults say they have asked
someone to remove or correct information about
them that was posted online, compared with just
17% of those 30-49, 12% of those 50-64 and 2%
of those ages 65 and older.

9, v
65

4
12.

No
Among social media users, 21% have asked for
something to be taken down or corrected,
compared with 9% of adults who do not use
social media. Mobile internet users, who also
tend to be younger, are more likely than nonmobile users to say they have asked for
something to be removed (22% vs. 7%).

Young adults most likely to have asked
someone to correct or remove info
about them online
Total
Age
18-29
30-49

d

e
ew
i

50-64

/
/03
08

015
2

65+

16%
32
17
12
2

Social media
Uses social media
Does not use social media

21
9

Mobile internet use
Uses internet on a mobile device

22

Does not use internet on mobile device

7

Self-searching activity
Used search engine to search for name

23

Has not self-searched

6

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.
PEW RESEARCH CENTER

Those who have searched for information about
themselves online were almost four times as likely to say they have requested a correction or
takedown of information when compared those who have not checked up on their digital
footprints (23% vs. 6%).
18 There were not enough cases in this sample to analyze variations among young women and young men. However, a separate report from
the Pew Research Center’s Internet Project found that young women, ages 18-24, experience certain severe types of online harassment at
disproportionately high levels when compared with men of the same age: http://www.pewinternet.org/2014/10/22/online-harassment/

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 45 of 57

44
PEW RESEARCH CENTER

Photos and videos are the most common types of information that people request to be
removed or corrected
There is a wide range of material that people request to have changed or taken down, but the most
common requests involve images and informal written material, rather than official records or
statements. Among adults who have asked someone to remove or correct material about them
online:




65% have asked that a photo or video be removed or corrected.
39% have asked that written material like a comment or blog posting be removed or
corrected.
13% have asked that something else, such as a court record or financial statement be
removed or corrected.

24% of employed adults say their employer has rules or guidelines about how they present
themselves online

015
One in four (24%) employed Americans say that
Work-related self-presentation online by
3/2
0
their employer has rules or guidelines about
education /
08
how they are allowed to present themselves
d
we
online (such as what they can post or what
e
information they are allowed to share about v
, i
themselves). One in ten employed Americans
659
4
(11%) say that their job requires them to
12.
promote themselves through social media or
No
% of adults in each education group who promote
themselves online or whose employers have rules about
self-presentation online
HS grad or less

other online tools.

Employees with at least some college
experience are more likely than those who have
not attended college to say that they need to
promote themselves professionally online, and
that their employer has policies about how they
are allowed to present themselves digitally.

However, few have set up automatic alerts
to monitor results connected to their name
online

Promote self
professionally
online

Employer has rules
about how you
can present yourself
online

Some college

College grad

4

13
16

15
28
28

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older.
PEW RESEARCH CENTER

Just 6% of adults have set up some sort of automatic alert to notify them when their name is
mentioned in a news story, blog, or elsewhere online. Young adults are somewhat more likely than

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 46 of 57

45
PEW RESEARCH CENTER

older adults to have done this—11% of 18-29 year olds have set up alerts—but just 1% of those 65
and older have done so. Perhaps unsurprisingly, “self-searchers” (people who have used a search
engine to look up what is available about them online) are more likely than non-self-searchers to
have done this. But even among this group the practice of setting up automatic alerts is far from
universal—some 9% of self-searchers have set up automatic alerts, compared with 2% of non-selfsearchers.

Many still do not assume that people they meet will search
for information about them online
Almost half of respondents (47%) say that they generally
assume that people they meet will search for information about
them on the internet, while 50% do not. Younger adults under
age 50 are more likely to say they assume new acquaintances
will search for information about them (53%) than older adults
(40%). College graduates (53%) are also more likely to assume
people will search for information about them online compared
with those who have not attended college (40%). Overall, 54% of
adults who have searched for information about themselves
online expect others will look for information about them,
compared with 36% of those who are not self-searchers.

d
we

e

.
No

, vi
659

4
12-

Do you assume new
acquaintances will search
for you online?
Among all adults, the % who agree
that new people they meet might
search for information about them
on the internet

015
2

Strongly agree

/
/03
08

Agree

37

Disagree

Strongly disagree

40
11

Source: Pew Research Privacy Panel Survey,
January 2014. N=607 adults, ages 18 and
older.
PEW RESEARCH CENTER

www.pewresearch.org

10

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 47 of 57

46
PEW RESEARCH CENTER

Appendix—Detailed Demographic Tables
The tables below include detailed demographic data for the nine most sensitive categories of
information. For each of these kinds of data, at least 40% of adults considered the information to
be “very sensitive.”

Those who view their social security
number as “very sensitive”
Total

90

Gender
Men

90

Women

89

Age
18-29

93

015
3/2
8/0

30-49

85

50-64

92

65+

d
we

Education
HS grad or less

59,
6

Some college

4

College graduate

12.
No

vie

0

92

85
92
94

Household income

Less than $30,000/year

81

$30,000-$49,999

86

$50,000-$74,999

91

At least $75,000

95

Gov’t surveillance awareness
Heard “a lot”

95

Heard “a little” or “nothing at all”

89

Self-searching activity
Used search engine to search for name

94

Has not self-searched

85

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 48 of 57

47
PEW RESEARCH CENTER

Those who view health information as
“very sensitive”
Total

55%

Gender
Men

55

Women

55

Age
18-29

56

30-49

55

50-64

59

65+

47

Education

015
3/2
8/0

No college

49

Some college

57

College graduate

d
we

Household income

e
, vi
659

Less than $30,000/year
$30,000-$49,999

0

61

49
43

4
12-

61

Heard “a lot”

65

Heard “a little” or “nothing at all”

49

$50,000-$74,999

.
No

At least $75,000

60

Gov’t surveillance awareness

Self-searching activity
Used search engine to search for name

61

Has not self-searched

45

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 49 of 57

48
PEW RESEARCH CENTER

Those who view the content of phone
conversations as “very sensitive”
Total

54%

Gender
Men

59

Women

50

Age
18-29

56

30-49

52

50-64

56

65+

51

Education
No college

50

Some college

55

015
3/2
8/0

College graduate

59

Household income
Less than $30,000/year

d
we

$30,000-$49,999

e
, vi
659

$50,000-$74,999
At least $75,000

4

12.
No

0

49

42
56

62

Gov’t surveillance awareness

Heard “a lot”

Heard “a little” or “nothing at all”

66
48

Self-searching activity

Used search engine to search for name

59

Has not self-searched

48

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 50 of 57

49
PEW RESEARCH CENTER

Those who view the content of email as
“very sensitive”
Total

52

Gender
Men

57

Women

47

Age
18-29

59

30-49

54

50-64

51

65+

42

Education
HS grad or less

45

Some college

55

015
3/2
8/0

College graduate

60

Household income
Less than $30,000/year

d
we

$30,000-$49,999

e
, vi
659

$50,000-$74,999
At least $75,000

4

12.
No

0

43

38
56

61

Gov’t surveillance awareness

Heard “a lot”

Heard “a little” or “nothing at all”

62
45

Self-searching activity

Used search engine to search for name

58

Has not self-searched

44

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 51 of 57

50
PEW RESEARCH CENTER

Those who view the details of their
physical location over a period of time
as “very sensitive”
Total

50%

Gender
Men

46

Women

53

Age
18-29

58

30-49

50

50-64

47

65+

43

Education
HS grad or less

44

015
3/2
8/0

Some college

53

College graduate

55

Household income
Less than $30,000/year

d
we

e
, vi
659

$30,000-$49,999
$50,000-$74,999

4
12-

At least $75,000

.
No

0

44

40
53
56

Gov’t surveillance awareness

Heard “a lot”

58

Heard “a little” or “nothing at all”

45

Self-searching activity
Used search engine to search for name

56

Has not self-searched

40

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 52 of 57

51
PEW RESEARCH CENTER

Those who view the numbers called or
texted from their phone as “very
sensitive”
Total

45%

Gender
Men

47

Women

43

Age
18-29

53

30-49

43

50-64

48

65+

36

015
3/2
8/0

Education
HS grad or less

40

Some college
College graduate

e

, vi
659

Household income

Less than $30,000/year

d
we

4
12-

$30,000-$49,999

.
No

0

50
47

36
45

$50,000-$74,999

45

At least $75,000

50

Gov’t surveillance awareness

Heard “a lot”

57

Heard “a little” or “nothing at all”

37

Self-searching activity
Used search engine to search for name

51

Has not self-searched

35

Mobile internet use
Accesses internet on mobile device

50

Does not access internet on mobile device

39

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 53 of 57

52
PEW RESEARCH CENTER

Those who view the content of text
messages as “very sensitive”
Total

49%

Gender
Men

50

Women

47

Age
18-29

55

30-49

48

50-64

50

65+

38

Education

015
3/2
8/0

HS grad or less

41

Some college

52

College graduate

d
we

Household income

e
, vi
659

Less than $30,000/year
$30,000-$49,999

0

57

39
38

4
12-

50

Heard “a lot”

56

Heard “a little” or “nothing at all”

45

$50,000-$74,999

.
No

At least $75,000

58

Gov’t surveillance awareness

Self-searching activity
Used search engine to search for name

54

Has not self-searched

41

Mobile internet use
Accesses internet on mobile device

55

Does not access internet on mobile device

40

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 54 of 57

53
PEW RESEARCH CENTER

Those who view their birth date as “very
sensitive”
Total

41%

Gender
Men

43

Women

39

Age
18-29

40

30-49

40

50-64

44

65+

39

Education
HS grad or less

38

Some college

41

015
3/2
8/0

College graduate

45

Household income
Less than $30,000/year

d
we

$30,000-$49,999

e
, vi
659

$50,000-$74,999
At least $75,000

0

39

30
50

43

HS grad or less

4

38

Some college

41

College graduate

45

12.
No

Education

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 55 of 57

54
PEW RESEARCH CENTER

Those who view relationship history as
“very sensitive”
Total

40%

Gender
Men

44

Women

37

Age
18-29

40

30-49

37

50-64

44

65+

41

Education
HS grad or less

29

Some college

45

015
3/2
8/0

College graduate

52

Household income
Less than $30,000/year

d
we

$30,000-$49,999

e
, vi
659

$50,000-$74,999
At least $75,000

4

12.
No

0

32

35
41

48

Gov’t surveillance awareness

Heard “a lot”

Heard “a little” or “nothing at all”

52
33

Self-searching activity

Used search engine to search for name

47

Has not self-searched

29

Source: Pew Research Privacy Panel Survey, January 2014. N=607
adults, ages 18 and older. For adults living in households earning
$30,000-$49,000 per year, n=97.
PEW RESEARCH CENTER

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 56 of 57

55
PEW RESEARCH CENTER

Methods
About this survey
The analysis in this report is based on a survey conducted January 10-27, 2014 among a sample of
607 adults, 18 years of age or older. The survey was conducted by the GfK Group using
KnowledgePanel, its nationally representative online research panel. GfK selected a representative
sample of 1,537 English-speaking panelists to invite to join the subpanel and take the first survey.
Of the 935 panelists who responded to the invitation (60.8%), 607 agreed to join the subpanel and
subsequently completed the first survey (64.9%). This group has agreed to take four online surveys
about “current issues, some of which relate to technology” over the course of a year and possibly
participate in one or more 45-60-minute online focus group chat sessions. A random subset of the
subpanel receive occasional invitations to participate in these online focus groups. For this report,
a total of 26 panelists participated in one of three online focus groups conducted during August
2013 and March 2014.

015
3/2
8/0

KnowledgePanel members are recruited through probability sampling methods and include both
those with internet access and those without. KnowledgePanel provides internet access for those
who do not have it and, if needed, a device to access the internet when they join the panel. A
combination of random digit dialing (RDD) and address-based sampling (ABS) methodologies
have been used to recruit panel members (in 2009 KnowledgePanel switched its sampling
methodology for recruiting panel members from RDD to ABS). The panel comprises households
with landlines and cellular phones, including those only with cell phones, and those without a
phone. Both the RDD and ABS samples were provided by Marketing Systems Group (MSG).

d
we

e

, vi
659

0

4

12.
No

KnowledgePanel continually recruits new panel members throughout the year to offset panel
attrition as people leave the panel. Respondents were selected randomly from eligible adult
household members of the panel.19 All sampled members received an initial email on January 10,
2014 to notify them of the survey and included a link to the survey questionnaire. One standard
follow-up reminder was sent three days later to those who had not yet responded.
The final sample for this survey was weighted using an iterative technique that matches gender,
age, education, race, Hispanic origin, household income, metropolitan area or not, and region to
parameters from the March 2013 Census Bureau’s Current Population Survey (CPS). In addition,
the sample is weighted to match current patterns of internet access from the October 2012 CPS
survey. This weight is multiplied by an initial base or sampling weight that corrects for differences

19 When a household is selected to be invited to KnowledgePanel, GfK attempts to enroll the entire household (age 14+). Subsequent
samples from the full KnowledgePanel are then selected using a “one member per household” criterion to avoid clustering affects.

www.pewresearch.org

Appeal: 12-4659

Doc: 170-2

Filed: 08/05/2015

Pg: 57 of 57

56
PEW RESEARCH CENTER

in the probability of selection of various segments of the sample and by a panel weight that adjusts
for any biases due to nonresponse and noncoverage at the panel recruitment stage (using all of the
parameters mentioned above as well home ownership status).
Sampling errors and statistical tests of significance take into account the effect of weighting at
each of these stages. Sampling error for the total sample of 607 respondents is plus or minus 4.6
percentage points at the 95% level of confidence.20 Sample sizes and sampling errors for other
subgroups are available upon request. The margins of error reported and statistical tests of
significance are adjusted to account for the survey’s design effect, a measure of how much
efficiency is lost from the weighting procedures.
In addition to sampling error, one should bear in mind that question wording and practical
difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.

d
we

e

, vi
659

015
3/2
8/0

0

4

12.
No

The original margin of error provided by the vendor and published here (3.98) used a less conservative
design effect. However, the analysis in the report was based on the more conservative design effect and the
significance of the findings is not affected by this change.

20

www.pewresearch.org