Kids and Credibility: An Empirical Examination of Youth, Digital Media Use, and Information Credibility

Kids and Credibility
This report was made possible by a grant from the John D. and Catherine
T. MacArthur Foundation in connection with its grant making initiative
on Digital Media and Learning. For more information on the initiative
visit www.macfound.org.
The John D. and Catherine T. MacArthur Foundation Reports on
Digital Media and Learning
Peer Participation and Software: What Mozilla Has to Teach Government by

David R. Booth
The Future of Learning Institutions in a Digital Age by Cathy N. Davidson

and David Theo Goldberg with the assistance of Zoë Marie Jones
The Future of Thinking: Learning Institutions in a Digital Age by Cathy N.

Davidson and David Theo Goldberg with the assistance of Zoë Marie
Jones
Kids and Credibility: An Empirical Examination of Youth, Digital Media Use,

and Information Credibility by Andrew J. Flanagin and Miriam Metzger
with Ethan Hartsell, Alex Markov, Ryan Medders, Rebekah Pure, and
Elisia Choi
New Digital Media and Learning as an Emerging Area and “Worked Examples”
as One Way Forward by James Paul Gee
Living and Learning with New Media: Summary of Findings from the Digital
Youth Project by Mizuko Ito, Heather Horst, Matteo Bittanti, danah
boyd, Becky Herr-Stephenson, Patricia G. Lange, C. J. Pascoe, and Laura
Robinson with Sonja Baumer, Rachel Cody, Dilan Mahendran, Katynka
Z. Martínez, Dan Perkel, Christo Sims, and Lisa Tripp
Young People, Ethics, and the New Digital Media: A Synthesis from the
GoodPlay Project by Carrie James with Katie Davis, Andrea Flores, John
M. Francis, Lindsay Pettingill, Margaret Rundle, and Howard Gardner
Confronting the Challenges of Participatory Culture: Media Education for the

21st Century by Henry Jenkins (P.I.) with Ravi Purushotma, Margaret
Weigel, Katie Clinton, and Alice J. Robison
The Civic Potential of Video Games by Joseph Kahne, Ellen Middaugh, and
Chris Evans
Kids and Credibility
An Empirical Examination of Youth, Digital Media Use, and

Information Credibility
Andrew J. Flanagin and Miriam Metzger
with Ethan Hartsell, Alex Markov, Ryan Medders, Rebekah Pure, and Elisia Choi
The MIT Press

Cambridge, Massachusetts
London, England
© 2010 Massachusetts Institute of Technology
All rights reserved. No part of this book may be reproduced in any
form by any electronic or mechanical means (including photocopying,
recording, or information storage and retrieval) without permission in
writing from the publisher.
For information about special quantity discounts, please email special_

[email protected].
This book was set in Stone Sans and Stone Serif by the MIT Press.
Printed and bound in the United States of America.
Library of Congress Cataloging-in-Publication Data
Flanagin, Andrew J.
Kids and credibility : an empirical examination of youth, digital media
use, and information credibility / Andrew J. Flanagin and Miriam
Metzger; with Ethan Hartsell ... [et al.].
p. cm.—(The John D. and Catherine T. MacArthur Foundation
reports on digital media and learning)
Includes bibliographical references.
ISBN 978-0-262-51475-0 (pbk. : alk. paper)
1. Mass media and youth—United States. 2. Digital media—United States
—Social aspects. 3. Electronic information resources—United States.
4. Information behavior—United States. 5. Truthfulness and falsehood
—United States. I. Metzger, Miriam J. II. Hartsell, Ethan. III. Title.
HQ799.2.M35F53 2010 302.23’10835—dc22 2009054316
10 9 8 7 6 5 4 3 2 1
Contents
Series Foreword vii

Executive Summary ix
Acknowledgments xvii
Rationale and Overview 1
Research Approach 6
Overview 6
Survey Methodology 9
Sample Characteristics 12
Research Findings 15
Internet Usage among Youth 15
Perceived Trust and Credibility of Web-Based Information 31
Factors Affecting Children’s Credibility Evaluations 57
Child/Parent Dyads and Credibility Assessments 73
Web Site Exposure and Evaluation 82
Conclusions and Implications 105

Summary 105
Implications and Future Directions 108
Conclusion 110
vi Contents
Appendix A: List of Tables and Figures 115

Appendix B: Knowledge Networks Methodology and Panel
Recruitment 119
Notes 127
References 131
Series Foreword
The John D. and Catherine T. MacArthur Foundation Reports

on Digital Media and Learning, published by the MIT Press in
collaboration with the Monterey Institute for Technology and
Education (MITE), present findings from current research on
how young people learn, play, socialize and participate in civic
life. The Reports result from research projects funded by the
MacArthur Foundation as part of its $50 million initiative in
digital media and learning. They are published openly online
(as well as in print) in order to support broad dissemination and
to stimulate further research in the field.
Executive Summary
The enormous amount and variety of information currently

available to people online present both tremendous opportuni-
ties and serious challenges. Readily available Web-based
resources provide extraordinary promise for learning, social
connection, and individual enrichment in a wide variety of
forms. Yet, the availability of vast information resources also
makes the origin of information, its quality, and its veracity less
clear than ever before, resulting in an unparalleled burden on
individuals to accurately assess information credibility.
Contemporary youth are a particularly intriguing and impor-
tant group to consider with regard to credibility because of the
tension between their technical and social immersion with digi-
tal media and their relatively limited development and life
experience compared to adults. Although those who have
grown up in an environment saturated with networked digital
media technologies may be highly skilled in their use of media,
they are also inhibited by their cognitive and emotional devel-
opment, personal experiences, and familiarity with the media
apparatus.
x Executive Summary
Despite these complex realities, examinations of youth and

digital media to date have typically been somewhat simplistic.
To provide a comprehensive look at children and online infor-
mation credibility, this project employed a large-scale, Web-
based survey of a representative sample of 2,747 children with
Internet access in the United States, ages 11 to 18. In addition,
one parent of each child was surveyed to obtain household indi-
cators of digital media use, parental involvement, and various
demographic factors.
Findings from this project constitute the first systematic
survey of youth designed to assess their information-seeking
strategies and beliefs across a wide variety of media and infor-
mation types. As such, findings can be used to inform parents,
educators, and policy makers interested in digital literacy and to
understand the realities of children’s relationship to digital
media and the information they glean from such media.
Key findings of this project can be organized in terms of chil-
dren’s Internet usage, their perceptions of information credibil-
ity and factors affecting these perceptions, child/parent dyads
and credibility assessments, and Web site exposure and
evaluation.
Regarding children’s Internet usage:
The vast majority of children began using the Internet
between second and sixth grades, with a majority of kids online
by third grade. Nearly all kids (97 percent) are online by the
eighth grade. Children use the Internet (not including email)
for an average of almost 14 hours per week, and usage generally
increases with age, from an average of 8 hours weekly among
11-year-olds to 16 hours per week for 18-year-olds.
Executive Summary xi
Overall, children rely fairly heavily on the Internet. The most

important general uses include social networking, virtual usage
(i.e., gaming and the like), information contribution in various
forms (e.g., sharing files with others or creating personal Web
sites, blogs, or journals), and commercial use (which is not very
common among children). Although children generally
acknowledge that information overabundance might pose a
problem for them, nearly two-thirds of children report that
their life would be either a little or much worse overall if they
could not go online again, which is more pronounced with age.
Children believe that they are highly skilled Internet users.
Even 11-year-olds believe that their technical skill, search skill,
and knowledge about Internet trends and features are higher
than other Internet users.
Seventy-five percent of parents control their child’s access and
use of the Internet by placing the computer in a certain loca-
tion in the home, limiting the sites their child can visit, limit-
ing the amount of time their child can go online, or controlling
their children’s Internet access in other ways. Parental oversight
of children’s online activities decreases as kids get older, with
each method of control reported about half as frequently by
parents of older children compared to parents of younger
children.
Regarding children’s perceptions of information credibility:

Young people are concerned about credibility on the Internet,
yet they find online information to be reasonably credible, with
89 percent reporting that “some” to “a lot” of information
online is believable. While the amount of information they find
xii Executive Summary
credible increases with age somewhat, their concern about cred-

ibility does not.
Their concern about credibility could stem from the fact that
73 percent of children have received some form of information
literacy training, and the majority of parents report that they
talk to their kids about whether to trust Internet information.
A third of children reported that they, or someone they know,
had a bad experience due to false information found on the
Internet or through email. In addition, nearly two-thirds said
that they had heard a news report about someone who had a
bad experience because of false information online. These expe-
riences affect how skeptical kids are of Internet information.
Among several options, the Internet was rated as the most
believable source of information for schoolwork, entertain-
ment, and commercial information, as well as second most
believable source for health information and third most believ-
able for news information. Notably, children report that the
Internet is a more credible source of information for school
papers or projects than books.
Kids are not very trusting of blogs, but they do find Wikipedia
to be somewhat believable. Many children report believing
information on Wikipedia substantially more than they think
other people should believe it.
Young people are appropriately skeptical of trusting strangers
or people they meet online and are decidedly more trusting of
people they meet in person.
Children differentiate in reasonable ways among entertain-
ment, health, news, commercial, and school-related informa-
Executive Summary xiii
tion online when deciding which credibility assessment tools to

use and with how much effort to employ them. Although this is
generally encouraging, children also report finding entertain-
ment and health information to be equally believable online,
suggesting a suboptimal degree of skepticism between these
diverse information types that have potentially quite different
consequences.
Older kids also show greater diversity and rigor in assessing
the credibility of online information. Moreover, young people
who are less analytic in their processing of information report
trusting strangers online more and are more likely to be fooled
by false information online.
Children’s concerns about credibility appear to be driven
largely by analytic credibility evaluation processes, which
involve the effortful and deliberate consideration of informa-
tion. By contrast, actual beliefs about the credibility of informa-
tion they find are dictated by more heuristic processes, by
which decisions are made with less cognitive effort and scru-
tiny. This suggests that while most kids take the idea that they
should be concerned about credibility seriously (by invoking a
systematic and analytical approach), many also exhibit a less
rigorous approach to actually evaluating the information they
find online.
There was no clear evidence of a “digital divide” in terms of
the credibility beliefs and evaluations of kids from different
demographic backgrounds. Instead, the rigor with which kids
evaluate information they find online drives much of their
credibility beliefs and concerns.
xiv Executive Summary
Regarding child/parent dyads and credibility assessments:
Parents believe they are more adept at assessing credibility

online than their children, and children almost universally
share this assessment. This is particularly pronounced for
younger children. However, the gap between parents and their
children in this regard narrows with age.
Children and adults both demonstrate an optimistic bias in
their ability to identify credible information when compared to
“typical” Internet users, indicating that they believe they are
better equipped to discern information credibility than the
average user. This is true even among children as young as 11
years old.
Regarding children’s Web site exposure and evaluation:

A majority of children displayed an appropriate level of skep-
ticism when presented with hoax Web sites, a trend that con-
tradicts prior research about this type of site. Nonetheless,
approximately 10 percent of children still believed hoax sites
either “a lot” or “a whole lot,” indicating some lingering and
important concerns.
Children found encyclopedia entries that they believed origi-
nated from Encyclopedia Britannica to be significantly more
believable than those they believed originated from either Wiki-
pedia or Citizendium.
The actual source of an online encyclopedia entry (i.e., taken
from Wikipedia, Citizendium, or Encyclopaedia Britannica) was
irrelevant to how credible the entry was found to be by chil-
dren. However, encyclopedia entries were assessed as less believ-
able when placed on Wikipedia’s site than when they were
Executive Summary xv
placed on the other sites. In addition, entries actually originat-

ing from Wikipedia were perceived as more believable when
they appeared on Citizendium’s web page than if they appeared
on Wikipedia’s page, and even more believable if they appeared
to have originated from Encyclopaedia Britannica. Thus, ironi-
cally, while children find the content of Wikipedia to be most
credible, they find the context of Wikipedia as an information
resource to be relatively low in credibility.
Children largely found product ratings to be credible and
important to their assessments of commercial information.
Average product ratings were significantly more influential
than the number of ratings the product received, and there was
some evidence that older children in particular were influenced
slightly by the combination of average ratings and the number
of ratings considered together.
Overall, this project provides a comprehensive investigation

into youth’s Internet use and their assessment of the credibility
of online information. The findings—which are generalizable to
households in the United States with Internet access—represent
the current state of knowledge on this topic and serve as an
important springboard for future research.
Acknowledgments
We are indebted to a great many people who contributed in

various ways to this project.
First and foremost, we deeply appreciate the support of the
John D. and Catherine T. MacArthur Foundation and in particu-
lar the vision, guidance, and intellect of Connie Yowell and
Craig Wacker. The MacArthur Foundation’s initiative on Digital
Media and Learning (DML) has served as a remarkable resource
for all those it has supported, and we have benefited immensely
from the conversations and help of other DML participants.
We also want to thank Sandra Calvert and Paul Klaczynski,
who served as consultants on this project.
A great deal of support has, of course, come from those at the
University of California, Santa Barbara, as well. We are fortunate
to have as our colleagues the doctoral students also listed on
this report and, in addition, we have benefited enormously
from the research assistance of a number of talented undergrad-
uate students, including Jennifer Bryan, Jennifer Dossett, Westin
Jacobsen, Kaitie Larsen, Cori Ochoa, Kamyab Sadaghiani, Caitie
Ulle, and Arrington Walcott. Finally, Katie Bamburg, Jana
xviii Acknowledgments
Bentley, and Monica Koegler-Blaha have provided invaluable

support for this project through the Institute for Social, Behav-
ioral, and Economic Research at UCSB.
Rationale and Overview
With the sudden explosion of digital media content and infor-

mation access devices in the last generation, there is now more
information available to more people from more sources than
at any other time in human history. Pockets of limited access
by geography or status notwithstanding, people now have
ready access to almost inconceivably vast information reposito-
ries that are increasingly portable, accessible, and interactive in
both delivery and formation. One result of this contemporary
media landscape is that there exist incredible opportunities for
learning, social connection, and individual enhancement in a
wide variety of forms.
At the same time, however, the origin of information, its
quality, and its veracity are in many cases less clear than ever
before, resulting in an unparalleled burden on individuals to
find appropriate information and assess its meaning and rele-
vance. Moreover, wide-scale access to information and the mul-
tiplicity of available sources also make it extremely complex to
assess the credibility of information accurately. And yet, it is
also highly consequential, since not having the skills to
2 Rationale and Overview
accurately assess the credibility of information can have serious

social, personal, educational, relational, health, and financial
consequences in today’s networked world.
While this is true for all users of digital media, youth are a
particularly intriguing group to consider with regard to credibil-
ity because of the tension between their technical and social
immersion with digital media and their relatively limited devel-
opment and lived experience compared to adults. On the one
hand, those who have literally grown up in an environment
saturated with networked digital media technologies may be
highly skilled in their use of media to access, consume, and gen-
erate information. This suggests that in light of their special
relationship to digital tools, youth are especially well positioned
to navigate the complex media environment successfully.
Indeed, forms of credibility assessment that rely on information
to be spread efficiently through social networks highlight some
intriguing advantages for youth populations, who are often
extremely interconnected compared to adults. In such instances,
younger users may actually be better equipped than adults to
transmit information pertaining to an entity’s credibility
quickly and efficiently via their social networks.
On the other hand, youth can be viewed as inhibited in
terms of their cognitive and emotional development, life experi-
ences, and familiarity with the media apparatus. This perspec-
tive suggests that although youth may be talented and
comfortable users of technology, they may lack critical tools
and abilities that enable them to seek and evaluate information
effectively. Children’s relative lack of life experience, for
instance, may put them at greater risk than adults for falsely
accepting a source’s self-asserted credibility, since such assess-

ments are based on accumulated personal experience, knowl-
edge, reputation, and examination of competing resources. As a
group, youth have fewer life experiences to which they might
compare information than do most adults. In addition, youth
may not have the same level of experience with or knowledge
about media institutions, which might make it difficult for
them to understand differences in editorial standards across
various media channels and outlets compared to adults who
grew up in a world with fewer channels and less media conver-
gence. As a consequence, some youth may not have the same
level of skepticism toward digital media or particular sources as
adults do, because these media are not seen as “new” to younger
users who cannot remember a time without them.
Although a good deal of scientific knowledge is accruing
with regard to how people determine the credibility of informa-
tion they get via digital media, extremely little of this work has
focused on children. This is surprising, given the unique rela-
tionship of contemporary youth to media technology. We
know, for example, that youth are more likely than adults to
turn to digital media first when researching a topic for school or
personal use; they are more likely to read news on the Internet
than in a printed newspaper; and they are more likely to use
online social networking tools to meet friends and to find infor-
mation. In other words, the primary sources of information in
their world are often digital, which is quite different from any
prior generation.
Indeed, many have noted that their special relationship to
digital media impacts the way youth approach learning and
4 Rationale and Overview
research. As the first generation to grow up with the Internet,

young people are comfortable collaborating and sharing infor-
mation via digital networks, and do so “in ways that allow them
to act quickly and without top-down direction” (Rainie 2006,
7). Moreover, the interactivity afforded by networked digital
media allows young people to play the roles of both informa-
tion source and receiver simultaneously as they critique, alter,
remix, and share content in an almost conversational manner
using digital tools. These realities, we believe, have profound
implications for how young people both construct and assess
credibility online.
Despite these complex realities, examinations of youth and
digital media have often been somewhat simplistic, focusing for
example on the popular generation gap caricature, where youth
are portrayed as either technologically adept compared to adults
or as utterly vulnerable and defenseless. Such considerations fail
to focus on the most important and enduring by-products of
heavy reliance on digital media: the impact of “growing up digi-
tal” (Tapscott 1997) is that more and more of the information
that drives our daily lives is provided, assembled, filtered, and
presented by sources that are largely unknown to us, or known
to us in nontraditional ways. Yet, we have only begun to explore
what this means for younger users who are not only immersed
in digital media now but will be for the entirety of their lives.
To address these issues, this project provides a comprehen-
sive look at kids and online information credibility, using a
large-scale survey of children in the United States, ages 11 to 18.
The research reported here fills the current void in knowledge
about how youth seek information and assess credibility using
many types of digital media. In the face of increasing disinter-

mediation and media complexity, the practical application of
such knowledge could be employed to empower users to reap
the benefits of the vast digital information environment while
minimizing the risks of relying on information that may be mis-
leading, incomplete, or wholly inaccurate. Overall, data from
this survey constitute the first systematic study of youth
designed to assess their information-seeking strategies and
beliefs across a wide variety of media and information types. As
such, our findings offer unprecedented insight into how young
people think about credibility today.
Findings from this study can be used to inform parents, edu-
cators, and policy makers interested in digital literacy, and to
understand the complex realities of children’s relationship to
digital media and the information they glean from them.
Research Approach
Overview
Although there is a burgeoning literature and empirical work

on adults and credibility (see Flanagin and Metzger 2007),
including informative work on college-age adults (Metzger, Fla-
nagin, and Zwarun 2003; Rieh and Hilligoss 2007), extremely
little research has been conducted on pre-college age youth.
What empirical research does exist is almost exclusively based
on interviews of very small samples of children and adolescents,
which cannot be generalized with any accuracy to the overall
youth population. To redress this shortcoming, this project gen-
erated survey data from a representative sample of young
people in the United States.
The survey instrument used in this study was generated
through a multi-step, multi-method process. The initial survey
topics were based on an extensive review of past literature and
existing surveys on information trust, credibility, and quality.
To better understand cognitive and developmental issues
relevant specifically to youth information assessment and
Research Approach 7
processing, research experts currently working in the fields of

developmental psychology and cognitive psychology were
recruited as project consultants. A draft version of the question-
naire was critiqued and modified through working sessions with
these consultants over the course of multiple days. The outcome
of these sessions was a comprehensive questionnaire pertaining
to digital media use and assessment, informed by contemporary
perspectives from cognitive and developmental psychology.
To gauge the clarity, comprehensiveness, and relevance of
the questionnaire for youth audiences, a small-scale focus group
was next conducted among children ages 9 to 18. Questionnaire
modifications were again made based on this session, and the
consultants’ recommendation that participants should be no
younger than 11 years old was confirmed. Next, to further vali-
date the questionnaire for youth audiences, as well as for the
portion of the survey evaluating parents’ assessments of their
child’s online information behaviors, 40 parent-child pairs were
recruited. These pairs represented a broad range of races, eth-
nicities, and household incomes, as well as roughly equal num-
bers of children in each age cohort and sex. Members of each
child/parent dyad underwent a separate hour-long face-to-face
interview with researchers, in which they provided feedback on
questionnaire content, question wording, and general survey
administration. Once again, this feedback was used to modify
the questionnaire.
This version of the survey was then pilot-tested among 183
undergraduate college students, in order to gauge the reliabili-
ties of the attitudinal and usage scales included in the question-
naire and to uncover any other outstanding issues. Minor
adjustments were made to the questionnaire, which was then
8 Research Approach
forwarded to the research firm that administered the survey to

the target population, as noted in more detail below.
Given the near-ubiquitous use of the Internet among con-
temporary youth, and the fact that this constituted our target
audience, Web-based survey techniques were used to assess
youths’ Web usage behaviors and attitudes about online credi-
bility. The questionnaire was administered to a sample of youth
with Internet access in the United States, ranging in age from 11
to 18 years old. In addition, one parent of each child was sur-
veyed to obtain household indicators of digital media use,
parental involvement in their child’s digital media use, and
demographic factors.
The 2,747 valid responses obtained were a roughly equal rep-
resentation across youth age cohorts (i.e., approximately 340
respondents for each age within the range). Surveying a range of
ages accomplished a number of things: it represented children
at critical junctures in social and cognitive development; it con-
sidered youth at times in their academic development and in
their development as citizens that are key to their future deci-
sions and choices; and it enabled comparisons between children
of various ages, providing relatively precise comparisons across
age cohorts (e.g., junior high versus high school), to pinpoint
the key junctures at which children attend to, and act on, dis-
tinctions in information credibility. Moreover, this sample size
yielded sufficient representation across sex and other demo-
graphic differences to facilitate comparison across these factors.
Finally, because the current project is an extension of ongoing
research on adults, it will also provide direct comparison
between youth and adult populations in future studies, which
will suggest lifespan differences in the key variables of interest.
Research Approach 9
Survey Methodology
Survey Administration
The survey was conducted online by the research firm Knowl-
edge Networks and was fielded between June 17 and July 26,
2009. Knowledge Networks maintains a probability-based panel
of participants and is thus the only online survey source that
meets the standard of federal and peer review, setting the gold
standard in the industry. As mentioned earlier, 2,747 children
in the United States between the ages of 11 and 18 who use the
Internet and who live at home, as well as one parent for each
child participant, completed the survey. Statistical results were
weighted to correct known demographic discrepancies between
the U.S. population and Knowledge Networks’ online panel.
Details on the design, execution, and weighting procedures of
the survey are discussed below. Additional information about
the survey methodology and subject panel used by Knowledge
Networks can be found in Appendix B.
Sample Design
Knowledge Networks has recruited the first online research
panel that is representative of the entire U.S. population. Panel
members are randomly recruited by probability-based sampling
(telephone, mail-, and Web-based surveys), and households are
provided with access to the Internet and hardware if needed
(although this did not apply to the current survey, since our
target sample included only current Internet users). After ini-
tially accepting the invitation by Knowledge Networks to join
the panel, respondents are then profiled online by answering
demographic questions, and maintained on the panel using the
10 Research Approach
same procedures established for research subjects recruited by

random digit dialing. The sample for this study was drawn from
a combination of random digit dialing and address-based sam-
pling methods (taken from the U.S. Postal Service’s Delivery
Sequence File). The combination of these two frames allows
Knowledge Networks to reach homes without a landline tele-
phone, homes with numbers on the do-not-call list, and homes
that use call-screening that normally would be missed by
random digit dialing methods alone.
The typical survey commitment for Knowledge Networks
panel members is one survey per week or four per month, with
a duration of 10 to 15 minutes per survey. Knowledge Networks’
general sampling rule is to assign no more than one survey per
week to members. Knowledge Networks operates an ongoing,
modest incentive program to encourage participation and create
member loyalty. Members can enter special raffles or can be
entered into special sweepstakes to win both cash and other
prizes.
For this study, households with children living at home
between 11 and 18 years of age were identified by Knowledge
Networks within their online panel (18-year-olds not living at
home were excluded from this sample). A sample was drawn at
random from among active panel members. For this survey,
5,936 U.S. adult parents with at least one child age 11 to 18
were selected for the main and pretest surveys.
Contact Procedures
Potential participants received a notification email letting them
know there is a new survey available for them to take. This
Research Approach 11
email notification contained a link that sent them to the survey

questionnaire. No login name or password was required.
Parents were first asked to complete a short screening ques-
tionnaire to confirm that they had a child age 11 to 18 and to
gain consent for the child to participate. Upon completion of
their portion of the survey, parents were asked to have one
selected 11- to 18-year-old complete a longer series of questions
designed to assess the child’s use of the Internet. To accommo-
date participants’ schedules and increase the chances of having
a child complete the survey, parents were told that they could
have their child complete the survey at a later time if that was
more convenient.
A first email reminder was sent to all non-responding panel
members in the sample on July 2, 2009. Second and third email
reminders were sent 7 and 12 days later, respectively. Finally,
calls were made to all remaining non-responding panel mem-
bers starting July 16, 2009 and throughout that weekend.
Incident and Completion Rates

For this survey, 3,136 adult parents with at least one child aged
11 to 18 responded to the invitations, representing a 52.8 per-
cent completion rate. 2,747 parent-child (aged 11 to 18) pairs
completed the survey and qualified for analysis, representing a
91.7 percent qualified rate or 46.3 percent response rate.
Sample Weighting
The survey responses were weighted to provide results that are
generalizable to the U.S. population of Internet households.
Two weighting strategies were employed to compensate for
non-response and other sources of survey error that might bias

the results.
First, a post-stratification adjustment using demographic dis-
tributions from the most recent U.S. Census Bureau’s Current
Population Survey data was used to balance errors due to panel
recruitment methods and panel attrition. Demographic vari-
ables used for this weighting included gender, age, race, educa-
tion, and Internet access.1 This weighting was applied before the
selection of the sample was made for this study.
In addition, a study-specific post-stratification weight was
applied after data collection to adjust for the study’s sample
design and survey non-response. A weight was calculated for all
qualified children to make them comparable to 13- to 18-year-
olds who have Internet access at home.2 Household income was
also included as a weighting variable since education could not
be included (i.e., most of the children in this age range have less
than a high school education). The sample design effect for this
weight is 1.58.
Sample Characteristics
This section provides a detailed profile of the demographic

characteristics for both the parent and child samples.
Parent and Household Demographics

Parents in the sample were 45 years old on average (standard
deviation = 7.25). Most had attended college, with 53 percent
having had at least some college, and 47 percent earning a
bachelor’s degree or higher. In terms of race, 75 percent of the
Research Approach 13
parents were white; 9 percent were black, non-Hispanic; 9 per-

cent were Hispanic; 4 percent were other, non-Hispanic; and 4
percent reported their race as Mixed, non-Hispanic. Thirty-one
percent of the participating parents were male, and 69 percent
were female. Eighty percent of parents were married or living
with a partner, 20 percent were divorced, separated, widowed,
or never married. Seventy-six percent of parents were working
at the time that the data were collected for this study.
Household annual income ranged from less than $5,000 to
more than $175,000, with an average income ranging from
$60,000 to $85,000. Most families (88 percent) had between 3
and 5 members living in the household, and the average
number of children living at home was 2.25 (standard deviation
= 1.39). Participants came from all parts of the United States,
with slightly more coming from the Midwest (31 percent) com-
pared to the Northeast (19 percent), South (28 percent), and
West (23 percent). Table 1 shows the more specific breakdown
of the sample’s geographic distribution:
Child Demographics
The child respondents consisted of 53 percent males and 47
percent females ranging in age from 11 to 18 years, with an
average age of 14.33 (standard deviation = 2.28). Table 2 shows
the percentage of children in the sample within each age group
surveyed. Seventy-five percent of the child respondents
reported that they were white; 9 percent were black, non-His-
panic; 12 percent were Hispanic; 0.4 percent were other, non-
Hispanic; and 4 percent reported being Mixed race,
non-Hispanic.
Table 1
Percent of participants from various U.S. geographic areas
Region Percent Residing
New England 4.9

East–North Central 20.6
East–South Central 4.7
Mid-Atlantic 13.6
South Atlantic 15.1
Mountain 7.2
Pacific 15.5
West–North Central 10.7
West–South Central 7.7
Table 2
Number and percent of participants within each age category sampled
Age Number Percent
11 378 13.8
12 371 13.5
13 385 14
14 323 11.8
15 327 11.9
16 316 11.5
17 368 13.4
18 279 10.2
Research Findings
Internet Usage among Youth
To better understand how young people may be affected by the

Internet, we first assessed the prevalence and nature of their
Internet usage. We examined children’s general use of the Inter-
net by measuring when they first started using the Internet and
how often they go online. We also investigated parental control
of their child’s access to and use of the Internet, since children’s
use behaviors are in some cases not entirely under their own
control. We next asked a number of questions about children’s
online activities, in order to fully understand their usage behav-
iors and activities. Finally, we evaluated young people’s percep-
tions of their Internet skill level and their overall impressions of
the Web environment.
Children’s Use of the Internet

The vast majority of children began using the Internet at some
point between second and sixth grade, with a majority of kids
(52 percent) online by third grade. Nearly all kids (97 percent)
reported being online by the time they were in eighth grade.
16 Research Findings
Children in the study reported that they use the Internet

(not including email), for an average of 13.53 hours per week,
although there was quite a lot of variation in the amount of
time they spent online (standard deviation = 12.44). This means
that the majority of kids (nearly 70 percent) spent anywhere
from 1 to 26 hours a week online, with few spending less than
one hour or more than 26 hours online weekly.
Internet usage did, however, generally increase with age,
with 11-year-olds reporting an average of 8.21 hours (standard
deviation = 6.06) per week online, which doubled by age 18 to
16.38 hours per week on average (standard deviation = 11.63).
Figure 1 shows the average amount of time per week spent
online by children of different ages.
Parental Control of Children’s Access to and Use of the Internet

A large majority of parents (75 percent) control their child’s
access and use of the Internet in some manner. In our total
sample, most parents (53 percent) place the computer in a cer-
tain location in the home in order to keep an eye on what their
child is doing online. Forty-three percent limit the sites their
child can visit, 42 percent limit the amount of time their child
can go online, and 19 percent control their children’s Internet
access in other ways. Among only those parents who control
their child’s access to the Internet in some fashion (the 75 per-
cent of the sample mentioned earlier), 71 percent place the
computer in a certain location in the home, 57 percent limit
the sites their child can visit, 55 percent limit the amount of
time their child can go online, and 25 percent control their
children’s Internet access in other ways.
20
Average hours per week online
15
10
0
11 12 13 14 15 16 17 18
Age of child
Figure 1
Weekly usage of the Internet by age
However, parental oversight of children’s online activities

decreases as kids get older. For example, 94 percent of parents of
11-year-olds report that they control their child’s Internet access
and use, whereas only 45 percent of parents of 18-year-olds
report controlling their child’s online activities. In fact, nearly
each type of control (e.g., placing the computer in a certain
location in the home, limiting the sites their child can visit, and
limiting the amount of time their child may go online) is used
about half as frequently by parents of older children compared
to parents of younger children, as seen in figure 2.
In addition, about two-thirds of parents reported that they sit
with their children while they go online, with only 38 percent
reporting doing this “never” or “rarely.” Nearly half (47 percent)
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
11 12 13 14 15 16 17 18
Age of child
Any type of restriction

Dictate location
Limit sites
Limit time
Other
Figure 2
Percentage of parents restricting their child’s Internet access by age of
child
reported sitting with their child “sometimes,” whereas 16 per-

cent reported sitting with their child “often” or “very often.” As
with the other forms of restrictions we examined, the percent-
age of parents who sit with their children while they are online
decreases with age, with a majority of parents of 11-year-olds
sitting with their children “often” or “sometimes,” while a
majority of parents of 17- to 18-year-olds report doing this
“rarely” or “never.”
Children’s Online Activities

To assess what young people are doing online, we asked a series
of questions about how often they use the Web for a variety of
purposes. Overall, children reported using the Web most often
for watching videos and interacting with others through social
networks, noting that on average they do each of these activi-
ties between “sometimes” and “often” (3 and 4 on a 5-point
scale, respectively). They also reported that they look up infor-
mation on Wikipedia, play games and use avatars, and buy
things online between “rarely” and “sometimes.” Other activi-
ties, including posting information to various groups, sharing
video or music files, and bidding in online auctions, were also
reported, but were reported only “rarely” or less often on aver-
age. Figure 3 shows children’s usage of Web-based information
resources, arranged in order of highest average reported uses to
the least commonly reported uses.
Although the frequency of most uses of the Web was rela-
tively consistent across ages, some notable trends were observed
in children’s Web usage by age. Many uses of the Web increased
with age, or increased up to a certain age, including sharing
videos and music, posting original artwork, photos, stories, or
videos online, and using social networking sites, as represented
in figures 4, 5, and 6. In each of these cases, there appears to be
a notable plateau in the frequency of these uses that occurs at or
about the age of 15. Interestingly, there was actually a decrease
in the extent to which children reported visiting virtual worlds
like Second Life or World of Warcraft as they grew older, as shown
in figure 7.
Some of these usage trends can be better understood in terms
of more general usage behaviors that emerge from the specific
Very often
Frequency of use
Often
Sometimes
Rarely
Never
e on kin g o
ur n p te
es ay bl ip e
Cr o t a F od e
te ine irt s st
W B og/ ass l w re
l e ld
s
it S ph on au nal
rm om c eb n
io h o te
o e
ik ine
Bu t am /p ia
st W m wit nl e
ar rite eth h a ine
so k/s blo on r
e ri fo e
st g vi m
ia
to om os
te or n g ata
ak p nl jo d
pr Pl al/ ik ofil
o l
m to g/ lin
p n
on g in
fo s ne w io
lin ati or kin de
ea nl v ile ca
en g og ed
y sel es rofi
ed
jo o g si
or id on ifi or
at et all si
Li thin es/ ru
bl cl ua ha
M on n o ine d a
a eo
en /s de
a rs e ur
n in nl
W nl
in ell o al ct
on rm etw or vi
Ra tw o in v
ip
n W r
an info l n etw h a
in
ad d cia al n atc
so f
Re Fin so oci W
e
se si
k i
U Vi
e
at e s
Re
U pd s
Po
e
Ed
U
Figure 3
Children’s use of Web-based information resources
Web uses reported by respondents. To assess this, we con-

structed several scales that describe children’s types of Web
usage.3 We derived measures describing (a) the frequency of
children’s social network use, (b) the extent to which they con-
tribute information to others, (c) the degree to which their use
of the Web is virtual (for activities like playing games), and (d)
the degree to which they use the Web for commercial purposes.
Social network use was based on how often children use social
networking sites, read others’ blogs, profiles, or journals online,
and update their own social networking profiles (Cronbach’s
alpha = 0.87). Subsequent analyses showed that older children
were more likely to be heavier users of social networks, as were
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 4
Frequency of file sharing by age
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 5
Frequency of posting original content by age
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 6
Frequency of using social networking sites by age
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 7
Frequency of visiting virtual worlds by age
more skilled users, girls, and those who spent more time online
per week. Children who earned lower grades in school and had
less parental control of their Internet access and use also showed
higher use of social networking sites. In addition, those relying
on group processes to discern credibility and those with “heuris-
tic” decision making styles (as opposed to more analytical styles,
covered in a later section of this report, “Factors Affecting Chil-
dren’s Credibility Evaluations”) were also heavier social network
users.
Information contribution scores were derived from such things
as the extent to which children reported creating personal Web
sites, blogs, or journals, posting information to groups or shar-
ing files with others, or rating people or things online, through
sites like Amazon, eBay, IMDb, or YouTube (Cronbach’s alpha =
0.77). Greater Internet skill, reliance on group processes for
making credibility decisions, time online per week, and age
resulted in greater information contribution. In addition,
among other factors, grades in school were negatively related to
information contribution, girls were more likely to contribute,
and high Internet social trust led to greater contribution,
although higher general social trust led to lower information
contribution.
Virtual uses consisted of representing oneself with an avatar,
visiting virtual worlds, or playing games with others who are
also online (Cronbach’s alpha = 0.71). The strongest predictors
of heavy virtual uses were hours online per week, sex (with girls
more likely to engage in virtual uses), Internet skill, and age. In
addition, virtual users tended to rely on group processes more in
making credibility assessments, had been taught at some point
about credibility issues online, tended to get good grades, and

had been online a long time.
Finally, commercial use was composed of bidding in online
auctions, buying or selling merchandise online, and frequent-
ing sites like Craigslist.org to look at classified ads (Cronbach’s
alpha = 0.69). Those with higher Internet skill and who were
older use the Web for commercial purposes more than do kids
with lower skill and who are younger, and commercial use of
the Internet decreased as parents controlled their children’s
Internet access and usage more. Boys tended to use the Web for
commercial purposes more than girls, and commercial users
tended to rely on others and also invoke analytic methods when
determining credibility online, though they also had high faith
in their own intuition in determining what, and who, is
trustworthy.
The frequency of social networking behaviors increases
rather dramatically between the ages of 11 and 15, as shown in
figure 8, after which it remains relatively constant. Whereas
11-year-olds report using social networking sites only between
“never” and “rarely” on average, older teenagers use them on
average between “sometimes” and “often.” This confirms find-
ings from past studies that have shown increases in social net-
work site usage between younger (12- to 14-year-olds) and older
(15- to 17-year-olds) kids (Lenhart and Madden 2007), and indi-
cates that as children grow older they may be using more digital
forms of communication to connect and interact within their
social networks.
Largely consistent with our earlier observation about visiting
virtual worlds, figure 9 shows that the frequency with which
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 8
Frequency of social networking use by age
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 9
Frequency of “virtual” Web use by age
children’s use of the Web for “virtual” purposes varies little by

age. Contrary to popular conceptions of older teens as more fre-
quent participants in virtual worlds, our data suggest that chil-
dren’s use of the virtual features of the Web is not only fairly
low (indicating that they “rarely” use the Web in this way), but
that this low usage is consistent across age. Not surprisingly,
commercial use of the Web increases with age, as shown in
figure 10, likely due to the availability of greater discretionary
income and increased independence from parents.
Figure 11 shows differences in how often children use the
Web for contributing information in various forms to others.
Once again, these results suggest that as children get older they
contribute more information, although it should be noted that
information contribution remains rare overall, with average
contributions of even older children not quite reaching even a
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 10
Frequency of commercial Web use by age
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Figure 11
Frequency of online information contribution by age
level of doing so “rarely.” Nonetheless, sharing information in

various ways (via personal Web sites, blogs, or journals; posting
information to groups or other people, or rating people or
things online) is indicative of one of the most notable features
of the Internet—its ability to enable information consumers to
simultaneously be information providers, a behavior that
appears to increase slightly during childhood.
Together, these findings show that children are going online
for a variety of purposes, although to varying degrees according
to their age. To evaluate the extent to which they have become
reliant on the Internet, we asked them to assess how their lives
would be different if they could never use the Internet again.
Only 1 percent reported that their lives would be better overall
if they could not go online again. By contrast, 64 percent
reported that their life would be either much or a little worse
overall, and 23 percent said that it would not make any differ-
ence to their life overall if they could not go online again. In
general, as their age increases children report that their lives
would be progressively worse without Internet access, indicat-
ing that older children experience increased reliance on the
Internet.
Children’s Web and Internet Skill Levels

To evaluate children’s self-perceptions of their skill levels with
Internet technologies, we asked them three questions, designed
to assess their technical skills (e.g., fixing connection problems
or changing computer settings), search skills (i.e., ability to find
what they are looking for online), and knowledge about Inter-
net trends and features. All questions were rated on a 0 to 10
scale, where 0 indicated that they were much less skilled/knowl-
edgeable than other Internet users and 10 indicated that they
were much more skilled/knowledgeable than other Internet
users. The scale midpoint of 5 indicated that respondents
thought they were “about as knowledgeable/skilled as other
Internet users.”
Results show that, overall, average perceived skill levels were
high. For technical skill, search skill, and knowledge about
trends and features, even 11-year-olds perceived that they were
on average more skillful/knowledgeable than other Internet
users (average scores = 5.32, 5.36, and 6.85, respectively). More-
over, across all three measures of Internet skill, children of all
ages perceived themselves to be more skillful than other Inter-
net users (average scores were all above the scale midpoint), and
saw their search skills as significantly better (average = 7.47)
than their technical skills (average = 6.49) or knowledge of

Internet trends and features (average = 6.48).
In addition, a trend emerged for all skill measures where
between ages 11 and 15 children felt they were generally more
skillful on average with each passing year. However, for all three
skill measures, self-perceived average skill and knowledge typi-
cally decreased slightly after the peak at age 15, although gener-
ally not to levels that achieved statistical significance. Overall,
this shows that children’s average self-perceived skills appear to
rise until the age of 15 or so, after which they generally either
level off or decrease modestly.
There were also intriguing sex differences with regard to skill
self-assessments. Setting aside children’s ages for the moment,
boys rated themselves as significantly more skillful on average
than did girls, for both search skills and technical skills (but not
for knowledge of Internet trends and features). Considering the
age of respondents, however, paints a slightly more nuanced
picture of children’s skill assessments: 11-year-old boys rated
themselves on average as more technically skillful than 11-year-
old girls did; 12- and 14-year-old boys and girls did not differ on
any self-assessed skill measure; and boys aged 15 through 18
tended to rate themselves as more skillful than did 15- to18-
year-old girls, on average, on at least one measure of skill, and
sometimes on all three measures. Thirteen-year-old girls, how-
ever, are distinct in that they rated themselves as significantly
more skillful than 13-year-old boys rated themselves, on both
search skills and on knowledge of Internet trends and features.
Overall, meaningful sex differences on self-perceived skill show
up as children reach their mid-teens, when boys believe
themselves to be more highly skilled than girls do. Prior to that

time, differences in skill are negligible other than a surprising
peak among 13-year-old girls in terms of their perceived Inter-
net and Web skills.
Children’s Impressions of the Web Environment

When asked about the amount of information available online,
children generally acknowledged that information overabun-
dance might pose a problem for them. Specifically, children
were asked, on a scale of 0 (not enough information online) to
10 (too much information online), what they thought about
the amount of information available online overall. The aver-
age response was 7.57, indicating that they thought there was
generally too much information available online. Moreover, 61
percent of children indicated responses ranging from 6 to 10
(too much information), whereas only 13 percent gave
responses ranging from 0 to 4 (not enough information). 26
percent of children, however, did indicate that the amount of
information online was “just right.” There were no statistically
significant differences in opinions about the amount of infor-
mation available online by age or sex, indicating that this find-
ing applied equally to children of all ages and both sexes.
Summary
Overall, findings indicate that children are actively using the
Internet for a variety of reasons. Other than the reported fre-
quency with which children’s use of the Web is virtual, it seems
that use of the Internet for myriad purposes increases with age
and, therefore, with the total amount of time spent online.
This increase in time spent online with age may be due to the
corresponding reduction of parental control. Interestingly,
although many children said there is too much information
online, the majority of children also indicated their Internet
skill was fairly high, suggesting that while children may be
overwhelmed or bombarded with information, they are confi-
dent in their ability to decipher and filter through it. This sug-
gests that children believe they are relatively capable of
successfully navigating the digital world, a premise that we
explore in much greater depth in the remainder of this mono-
graph, particularly as it relates to information credibility.
Perceived Trust and Credibility of Web-Based Information
This section and the next discuss survey results intended to

assess children’s perceptions of the credibility of information
on the Web. We begin by considering general issues regarding
the credibility of Web-based information, including children’s
trust of people they meet online and their experiences with
false information on the Internet. We then look at differences
in children’s perceptions of online credibility across various
types of information and information sources, as well as across
several media (e.g., Internet, television, newspapers), and we
investigate the behaviors children engage in when deciding
which information and what sources to trust.
General Issues Regarding the Credibility of Web-Based Information

In general, children found information on the Web to be
relatively believable, with 59 percent reporting that “some”
information was believable, and 30 percent reporting that “a

lot” of the information found online was believable (see figure
12). There was also a small but significant tendency for per-
ceived information believability to increase with age: 18-year-
olds found more of the information online to be credible than
11- through 14-year-olds.
When asked how often they think about credibility, as well
as how concerned others should be about the credibility of
online information, children showed a healthy level of concern
about these issues. Seventy-nine percent of children in the
sample said they think about whether they should believe infor-
mation they find online “sometimes” or more often, and 71
percent said that people should be “somewhat” to “very” con-
cerned about the believability of online information. Figures 13
and 14 summarize these results. Age did not matter much in
these findings, although 18-year-olds felt people should be more
70%
60%
Percent believing
50%
40%
30%
20%
10%
0%
None of it A little of it Some of it A lot of it All of it
Figure 12
Amount of online information that children find credible
60%
50%
40%
30%
20%
10%
0%
Never Rarely Sometimes Often Very often
Figure 13
Frequency of thinking about credibility when online
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Not at all A little Somewhat Concerned Very
concerned concerned concerned concerned
Figure 14
How concerned others should be about credibility
concerned about how believable information online is than

both 12- and 14-year-olds.
Trust and the Negative Consequences of False Information Online

When asked whether they felt people could be trusted, children
reported that they trusted people they knew or met in person
more than they trusted people online. Specifically, on a 4-point
scale (where higher scores indicate higher trust), children
reported an average trust score of 2.65 for people in person,
whereas online this score decreased to 1.17. Figures 15 and 16
show the specific differences in scores, as well as the categories
of children’s responses to the trust questions.
Similar to this, children tended not to trust strangers they
meet online (e.g., in social networking sites, forums, online
70%
60%
50%
40%
30%
20%
10%
0%
Most people can Most people Most people Most people can
almost always can usually usually cannot almost never
be trusted be trusted be trusted be trusted
Figure 15
General social trust
70%
60%
50%
40%
30%
20%
10%
0%
Most people on Most people Most people Most people on
the Internet can on the Internet on the Internet the Internet can
almost always can usually usually cannot almost never
be trusted be trusted be trusted be trusted
Figure 16
General Internet social trust
communities, etc.), with nearly half indicating they do not trust

people under these circumstances at all, and only about 2 per-
cent saying they trust strangers online “a lot” or “a whole lot.”
About a third of children did indicate they trust strangers online
“a little bit” and 15 percent said they trust them “some” (see
figure 17). It should be noted, however, that nearly 42 percent
of the children in the sample indicated that they have “never”
met a stranger online. There was little variance in children’s
level of stranger trust by age, although older children (16–18
years old) indicated they were slightly more trusting of strangers
online than younger children (11–14 years old). Figure 18 illus-
trates these findings.
60%
50%
40%
30%
20%
10%
0%
Not at all A little bit Some A lot A whole lot
Figure 17
Degree that children trust strangers they meet online
Not at all
A little bit
Some
A lot
A whole lot
11 12 13 14 15 16 17 18
Age of child
Figure 18
Degree that children trust strangers they meet online by age
One possible reason for these diminished levels of trust

online is negative experiences in children’s pasts. To assess this,
we explored the extent to which children had negative experi-
ences online, either firsthand or through others’ experiences.
Thirty-two percent of children reported that they, or someone
they know, had a bad experience due to false information found
on the Internet or through email. This did not vary by age.
Nearly twice as many children (62 percent) reported that they
had heard a news report about someone who had a bad experi-
ence because of false information online. In this case, differ-
ences were only found between 11- and 18-year-olds (with
18-year-olds reporting higher scores).
To mitigate or avoid negative experiences, children can of
course be instructed in recognizing bad information or evaluat-
ing information in general. To assess the extent to which chil-
dren have been taught about various issues regarding the
credibility of online information, we asked two questions. First,
we asked children if they had ever had someone (like a teacher,
parent, librarian, or friend) teach them ways to decide what
information from the Internet they should believe. Results show
that 73 percent have indeed been instructed by someone on
how to assess the credibility of information online.
Second, we asked parents how often they talk with their
child about whether information on the Internet is trustworthy.
Most parents (84 percent) reported that they talk with their
children about whether the information on the Internet is
trustworthy at least occasionally, with only 16 percent reporting
doing this “never” or “rarely.” Forty-five percent reported talk-
ing with their child “sometimes,” and 31 percent reported
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Never Rarely Sometimes Often Very often
Figure 19
Frequency that parents talk with their children about online credibility
talking with their child “often.” However, only 8 percent talk

with their child “very often” (see figure 19). Interestingly, this
does not change much with the age of the child, as a majority of
parents within each age category say they talk with their chil-
dren “sometimes” to “often” about whether information on the
Internet is trustworthy.
Credibility Differences by Information Type and Source of

Information
Past research on credibility has found that the degree to which
adults believe information they find online varies by the type
or topic of information they might search for. For example,
people tend to be less likely to find commercial information or
information coming from special interest groups to be credible,
presumably because they recognize the strong potential for bias
(Flanagin and Metzger 2000, 2007). We wanted to see if similar

patterns were found for younger Internet users. In addition,
because most past work has focused on the credibility of Web
sites, we wanted to explore young people’s perceptions of newer
information sources that they likely encounter online, includ-
ing blogs and Wikipedia.
Information Type Despite indications that children tend not to

be terribly trusting of others online (as noted earlier), they indi-
cate a great deal of faith in the Internet as a source of conse-
quential information, compared to more trivial information
pursuits. We asked children how likely they are to believe infor-
mation on the Internet about a number of topics or types of
information, including health or medical issues, news, some-
thing they may want to buy, entertainment information (e.g.,
about movies, musicians, celebrities, etc.), other people they
meet online, and information they find for school papers or
projects. Results showed that children varied in their likelihood
of believing information across these topics.
Specifically, children were on average most likely to believe
information on the Internet about schoolwork, followed by
news, then entertainment and health information (which chil-
dren were on average equally likely to believe), commercial
information, and information about people they met online.
Figure 20 illustrates these results. Although there were minor
age differences with these findings (e.g., some of the older chil-
dren did not distinguish between the believability of health and
commercial information, and children of some ages did not dis-
tinguish between commercial and entertainment information),
the general pattern of findings endured regardless of age.
Very likely
Likelihood of believing
information online
Likely
Somewhat
likely
A little bit
likely
Not at all
likely
Health News Something Entertain- People Information
to buy ment you meet for
online schoolwork
Information type
Figure 20
Credibility of online information by information type
Although it makes sense that news information, for example,

is highly regarded in terms of its credibility, it is potentially
problematic that children believe health and entertainment
information equally. Indeed, these findings suggest that kids
may not be processing the credibility of health information
with any more rigor than they are assessing the trustworthiness
of entertainment information, in spite of the fact that the pos-
sible negative consequences of believing false health informa-
tion are far greater in scope and scale than are the negative
consequences of false entertainment information. Nonetheless,
there is also encouraging news, inasmuch as children tend to
rate as least believable information that might be presented
with persuasive, or even potentially nefarious, intent (i.e., com-
mercial information and information about people they meet
online).
Yet, it is unclear from these data whether children view infor-

mation found for school as highly credible because of a high
level of diligence in searching for or verifying it, or if they base
such assessments on the convenience of finding information on
the Internet.
Past research on college-age students (Metzger, Flanagin, and
Zwarun 2003), however, shows that students not only rely on
Web-based information for their schoolwork quite heavily, but
verify its veracity less than adults, suggesting that convenience
may be a more critical factor here.
Blogs Unlike their parents, children today are growing up in a

news media environment full of sources that do not have the
journalistic checks present in traditional media like newspa-
pers, magazines, or radio and television broadcasts. Our survey
assessed children’s level of trust in both old and new informa-
tion sources, with some interesting results.
Overall, kids do not find (news) blogs to be very credible. Sev-
enty-nine percent say they are either “much less” or “somewhat
less” believable than newspaper and television news. This does
not vary much by age. It should be noted, however, that many
kids were unsure about the comparative credibility of blogs and
mainstream news, with 37 percent of all kids answering “I don’t
know” about their relative credibility and 8 percent of the total
sample indicating that they did not know what a blog is. Figure
21 illustrates responses for those with an opinion on the topic.
Wikipedia Another recent information source phenomenon is

Wikipedia, the online encyclopedia to which anyone can con-
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Blogs are Blogs are Blogs are Blogs are Blogs are
much less somewhat equally somewhat much more
believable less believable believable more believable believable
Figure 21
Credibility of blogs compared to newspapers and television
tribute information anonymously. Wikipedia currently boasts

more than 3 million entries and is among the top 10 most popu-
lar Web sites in the United States (Alexa 2009; Quantcast 2009).
Nearly all kids (99 percent) who completed our survey had
heard of Wikipedia, and the vast majority of them (84 percent)
have used it to look up information, with most reporting they
“sometimes” look up information on Wikipedia. However, few
(12 percent) have ever written or changed some information in
Wikipedia and those who have done so report doing it only
“rarely.” This does vary by age, to some degree: older kids are
about 10 percent more likely to have done both activities (look
up information and write or change information in Wikipedia)
than the youngest kids in our sample.
However, when asked to identify what Wikipedia is from a

list of seven possibilities (e.g., whether it is an online encyclope-
dia where anyone can contribute information, a social network-
ing site, a Web site where you can play games, an online
encyclopedia where only experts may contribute information, a
company that sells books online), 9 percent admitted that they
do not know what it is, and only 78 percent made the correct
identification. Moreover, there was a small tendency for older
kids (ages 16+) to more accurately understand what Wikipedia is.
Because this distinction is important, in subsequent analyses we
assessed only those children who correctly identified what Wiki-
pedia is and how it operates.
Overall, children find Wikipedia to be fairly believable. Most
children believe information from Wikipedia at least “some” (43
percent) or “a lot” (28 percent). However, children were slightly
more skeptical about how much people should believe Wikipe-
dia, with 23 percent saying it should be believed “a little bit,” 49
percent saying it should be believed “some,” and 20 percent
saying it should be believed “a lot.” Indeed, the extent to which
children say people should believe information in Wikipedia is
significantly lower than they report believing it themselves.
These results are illustrated in figure 22. There were no differ-
ences between younger and older children either in how much
they themselves believed or in how much they thought that
people should believe the information found on Wikipedia.
Differences in Credibility across Media

To determine which channel of information delivery children
think provides the most believable information for a variety of
60%
50%
40%
30%
20%
10%
0%
How much information the How much information the

child believes on Wikipedia child thinks others should
believe on Wikipedia
Figure 22
Credibility of Wikipedia
purposes, we asked them to indicate which among several alter-

natives (including the Internet, television, books, magazines,
newspapers, radio, and someone they talk to in person) pro-
vides the most believable information.
Consistent with past research (Flanagin and Metzger 2000),
differences emerged across technologies depending on the type
of information sought. When looking for health or medical
information, 39 percent of children indicated that they would
most believe someone they talk to in person, followed by the
Internet (21 percent) and books (20 percent), which were
roughly equivalent. Children indicated that the most believable
news information originated from television (54 percent),
followed by newspapers (24 percent), and then the Internet (11

percent). Commercial information was best retrieved from the
Internet (41 percent) or in person (33 percent), followed by tele-
vision and magazines (10 percent each). The most believable
entertainment information, according to children, can be found
on the Internet (40 percent), then television (28 percent), then
in magazines (11 percent). Lastly, 53 percent of children noted
that the most believable information for school papers or proj-
ects can be found on the Internet, followed by books (34 per-
cent), and then people they talk to in person (7 percent). These
results are summarized in figure 23.
Overall, children appear to rely fairly heavily on the Internet
to access different types of information. It was rated as the most
believable source of information for schoolwork, entertainment,
and commercial information, as well as second most believable
for health information and third most believable for news
information.
Some age differences emerged in children’s indication of
which channels they believe most for specific types of informa-
tion. For instance, older kids tended to believe entertainment
information from the Internet and newspapers more than
younger kids did, and entertainment information from books
and the radio less than older kids did. Additionally, older chil-
dren believed health information from the Internet, books, and
magazines more than younger children did and health informa-
tion from the radio less than younger kids did.
With news information, older children believed the Inter-
net, books, and magazines more than younger kids, and in-
person and radio sources less than them. For school-related
60%
50%
40%
30%
20%
10%
0%
Health News Commercial Entertainment School
Which source would you believe most for this type of information?
Internet
Television
Books
Magazines
Newspapers
Radio
In Person
Figure 23
Most credible source by information type
information, older children believed books and magazines more

than did younger kids and in-person sources less than them.
Finally, the Internet and newspapers are seen as more credible
channels for commercial information for older kids than for
younger ones, while television was seen as a less credible source
of commercial information by older versus younger children.
We next asked children how much people should believe the
information they find via particular media channels, including
newspapers, television, and the Internet. Children indicated sig-
nificantly different assessments of which medium should be
believed, noting that newspapers should be believed the most,
followed by television, and finally the Internet, as shown in
figure 24. These assessments did not vary with the age of chil-
dren. Once again, when considered together with the findings
above about credible sources by information type, it appears
that in some ways children’s own use of the Internet may
exceed the extent to which they think others should rely on it
for credible information.
Methods of Determining Information and Source Credibility

Past research on credibility and on decision making more gen-
erally has suggested that there are several ways that people may
approach information processing when evaluating information
(Metzger 2007; Scott and Bruce 1995). One method is to care-
fully analyze the information and its features, another is to use
a more holistic and intuitive approach based on feelings, and a
third method is to draw upon others in one’s social circle for
advice and guidelines. Indeed, our own research on adults has
found evidence for these three strategies in people’s credibility
A whole lot
A lot
Some
A little
Not at all
Newspapers Television Internet
Figure 24
Degree to which others should find information credible by medium
determinations, which we call “analytic,” “heuristic,” and

“group-based” (Metzger, Flanagin, and Medders, forthcoming).
(See also the next section, “Factors Affecting Children’s Credi-
bility Evaluations,” for a more detailed description of these
methods).
We asked children the extent to which they based their cred-
ibility assessments on heuristic (e.g., by relying on their gut feel-
ings, making decisions based on feelings, making quick
decisions), analytic (by carefully considering the information,
double checking facts, gathering a lot of information, and con-
sidering all views), or social (by getting advice from others or
asking for others’ help) criteria when evaluating whether to
believe information online. Kids reported that they used ana-
lytic techniques to carefully evaluate the credibility of informa-
tion online “sometimes” to “often” (3 and 4 on the scale,
respectively) whereas they used social and heuristic methods
less often overall.
Although this pattern of using analytic methods most often,

followed by heuristic and then social methods, was similar
across all age groups, the frequency with which kids used each
of these strategies increased with age. In other words, there was
a general trend in that older children reported applying all three
methods of credibility evaluation more often than younger kids
(see figure 25).
Interestingly, these results do not comport with research on
adults, who indicate that they often use heuristic methods of
credibility evaluation. Without further study, however, it is
impossible to say whether this difference is due to true differ-
ences between kids and adults in their strategies for evaluating
Frequency of using evaluation method
Very often
Often
Sometimes
Rarely
Never
11 12 13 14 15 16 17 18
Age of child
Social Heuristic Analytic
Figure 25
Use of credibility evaluation methods by age
credibility, or to the specific question wording or research

method used in this study (i.e., survey methods versus focus
groups). Indeed, the question itself may have prompted kids to
think about situations in which knowing the credibility of the
information they sought was important, rather than consider-
ing how they evaluate credibility across the full range of infor-
mation-seeking situations (e.g., the question asked how often
they used analytic, heuristic, or group-based strategies when
deciding what to believe, rather than simply asking how often
each strategy is used while looking at information online).
Moreover, survey responses on items like these are suscepti-
ble to social desirability response biases, where study partici-
pants want to sound like they are more diligent and informed
than they really are. So, while these results are intriguing and
even encouraging in that they suggest children are carefully
considering the credibility of information they find on the
Internet, they should be interpreted with caution until further
research can be conducted.
In order to further understand the ways in which children
make situation-specific (as opposed to general) judgments about
the credibility of information they find online, we asked them
how important a number of cues/elements were when they
were determining whether to believe information they found
online. To do this, each child was asked to imagine a situation
in which he or she was seeking information about a certain
topic or type of information (i.e., health information, news
information, entertainment information, information about
something they wanted to buy online, or information for a
school paper or project).
Across all five types of information, the most important

cues/elements involved the currency of the information, the
security of the Web site, information completeness, and the
authority of the information source (for example, if the infor-
mation originated from experts). Next most important were a
number of items that dealt with social endorsement and reputa-
tion; those cues reported as least important dealt with Web site
design and general feelings about the Web site (see table 3 for
details).
It is interesting to note that across all of the credibility cues
and elements presented, responses ranged from 3 (“somewhat
important”) to 4 (“important”), and that no cue/element was
considered by children on average to be either “very important”
or “not at all important.” Also, while age differences were not
dramatic, there was a slight trend toward many of these cues/
elements to increase in importance for older children. Advances
in cognitive development that come with age, accumulated
(positive and negative) experiences with online information, or
having had information literacy training could explain these
results.
Regardless of their explanation, however, the results overall
suggest that kids do pay more attention to the “right” cues
when determining credibility, at least according to digital liter-
acy advocates and educators who stress the importance of
source authority or expertise, security, and information cur-
rency and scope as credibility markers (e.g., Alexander and Tate
1999), and that learning to evaluate credibility by examining
many facets of Web sites and information provided on them is
a process that develops over time.
Table 3
Importance of various credibility cues
Credibility cue Mean SD
The information on the Web site is up-to-date 3.85 1.05

The Web site seems safe and secure* 3.80 1.07
The information is very complete 3.71 1.04
Experts believe the information (like your doctor, 3.68 1.09
teacher, etc.)
The information is from an expert on the topic 3.66 1.07
The information seems reasonable to you* 3.54 0.99
You ask an expert (like your doctor, teacher, etc.) who 3.49 1.16
you know in person
It does not try to convince you to do something or buy 3.49 1.24
something
The Web site is easy to use* 3.45 1.16
People you know, such as friends and family, believe the 3.43 1.08
Web site or information source
There are high ratings, positive comments, or good 3.41 1.11
reviews
Others recommend the Web site or information source 3.39 1.05
You get more than just one person’s opinion* 3.38 1.09
You have heard good things about the information 3.37 1.11
source or Web site creator*
The information is well written, and you see no typing 3.30 1.23
mistakes
The information on the Web site is similar to informa- 3.28 1.11
tion on other Web sites
There is information about the source’s or author’s edu- 3.20 1.19
cation or training
You have heard of the source or information creator 3.13 1.14
before*
A lot of other people use the Web site* 3.07 1.17
The information you find is similar to what you already 3.06 1.10
think
The Web site address has a certain ending (like “.gov” or 2.96 1.34
“.edu” or “.com”)
Note: 1 = not at all important, 2 = a little important, 3 = somewhat im-

portant, 4 = important, 5 = very important; *this item did not vary by
information type
Methods of Determining Information and Source Credibility by

Information Type We also compared the cues and elements
that kids use to evaluate the credibility of various types of infor-
mation individually (i.e., health, news, entertainment, com-
mercial, and school-related information). Regardless of the type
of information sought, children reported that the following
things were equally important in determining credibility:
Reputation having heard good things about the information

source or Web site creator and having heard of the source or
information creator before
Endorsement getting more than one person’s opinion and the
fact that a lot of other people use the Web site
Security the Web site seems safe and secure
Site design the Web site is easy to use, the Web site looks good,
and just liking the Web site
Information plausibility the information on the site seems
reasonable
Differences in the importance of some credibility cues/ele-

ments did emerge for different types of information on several
items. When it comes to more consequential information, spe-
cifically, health information and information for their schoolwork,
kids felt the following cues/elements were particularly impor-
tant for determining credibility:
Expertise experts believe the information, the information is

from an expert, there is information about the source’s educa-
tion/training (expertise), and they ask an expert whom they
know in person
Endorsement others recommend the Web site

Professionalism the information is well written and there are
not typographical errors
Information currency and comprehensiveness the information is
current and complete
Information consistency the information is similar to informa-
tion on other Web sites
Lack of bias the Web site does not try to sell you something, or
the Web site address ends in “.gov” or “.edu”
It is interesting to note that the various expertise items were

rated as most important for health information, whereas informa-
tion currency and its completeness, professionalism, consis-
tency with other Web sites, endorsement by others, and lack of
commercial motive/bias were rated as most important for school-
related information. A different picture emerged for commercial
information, where the most important cues/elements for deter-
mining credibility were endorsement (i.e., people you know
believe the site or there are high ratings and positive reviews)
and similarity of the information to what children already
believed.
Nearly all credibility cues and elements were rated as only
moderately important for news information, except that kids said
it was less important for news information to be similar to what
they already think or to have high ratings, as compared to most
other types of information where they felt these were important
factors in credibility assessment. More interestingly, kids felt it
was less important that news information be endorsed by others
(for example, when people you know believe the Web site or
source or when others recommend it), as compared to the other

information types. Most surprisingly, children did not rate the
importance of news information’s completeness and currency
as being any higher than for entertainment and e-commerce
information, although older kids saw currency of news informa-
tion as a more important credibility cue than did younger kids.
In almost every case, all credibility cues/elements were rated
as least important for deciding whether entertainment informa-
tion was believable. It appears that children recognize that
entertainment information is less consequential to evaluate for
its credibility compared to the other information types.
In terms of age differences, there was a slight tendency for
younger children (11- to 12-year-olds) to feel that most of the
cues/elements are less important for evaluating the credibility of
each type of information than older children. This may reflect
the fact that older kids have had more online experience that
would lead them to consider more facets of online information
in their credibility assessments than younger kids, or it may
reflect simple developmental or experiential differences.
Summary
Overall, kids view information on the Web as relatively credi-
ble, and they distinguish across information type and informa-
tion source when determining the credibility of online
information. They show an awareness of the possibility of nega-
tive consequences stemming from false information online,
and admit that they should probably believe some information,
such as that on Wikipedia, less than they actually do. Moreover,
children trust information more or less depending on its type,
with some information, such as that used for school projects,

seen as more believable than information from strangers they
meet online, for example.
Kids also appear to employ different credibility-assessment
tools depending on the type of information they are seeking.
Our data demonstrate that kids differentiate among news, enter-
tainment, health, commercial, and school-related information
when choosing which tools to use. For instance, kids tend to
look at fewer credibility cues, and apply these cues with less
rigor, when looking for entertainment information than they
do when looking for news, health, commercial, or school-
related information.
Our study also reveals some slight differences across age
groups in the ways kids view the credibility of online informa-
tion as well as the tools they use to assess information credibil-
ity. For instance, older kids are slightly more likely to correctly
identify what Wikipedia is, and are more likely to use heuristic,
social, and analytic methods of assessing credibility simultane-
ously than are younger kids. Moreover, older kids appear to
trust the Internet more as an information source than younger
children do, with credibility ratings for entertainment, news,
commercial, and health information increasing with age. Older
kids (in this case 18-year-olds) also are more likely to report that
people should be more concerned about online information
credibility than younger kids (12- to14-year-olds).
Interestingly, age differences did not surface in several other
areas. For instance, differences do not exist in how much older
and younger children believe Wikipedia or in how much they
thought they should believe it. Children’s rating of information
credibility across information types also is steady across age

groups.
Factors Affecting Children’s Credibility Evaluations
In addition to understanding the general parameters of trust

among youth, we also analyzed the data to assess what factors
play a role in kids’ credibility judgments. Surprisingly little
research has focused on questions about what types of people
are more or less likely to believe information they find online,
the extent to which young people exhibit particular Internet
usage patterns or online information evaluation practices, or
how prior experiences online drive young people’s credibility
beliefs.
To address these issues, we examined how various factors
impact both kids’ credibility concern as well as their beliefs about
the credibility of Internet information of various sorts. These factors
include:
demographic or background characteristics
patterns of Internet usage, access, and experiences
personality traits
specific strategies or methods for evaluating credibility
The demographic or background characteristics we examined

included young people’s sex, age, household income, race, and
grades in school. Previous research on credibility evaluation has
paid scant attention to these sorts of factors, although there is
reason to believe that children’s information evaluation strate-
gies and opportunities may vary developmentally across age
(see Eastin 2008), income (see van Dijk 2006), or by other types
of demographic groupings.
We also suspected that certain patterns of Internet usage, access,
and experiences could impact young people’s perceptions of
credibility online. In terms of usage, we investigated how much
time young people spend with the Internet (per week as well as
the number of years they have been using the Internet), their
level of skill (technical and search skill, as well as knowledge of
the latest online trends and features, as detailed earlier under
“Children’s Web and Internet Skill Levels” in the section on
“Internet Usage among Youth”), and their use of the Internet for
specific activities (social networking, contributing information
online, visiting virtual worlds, and using the Web for commercial
purposes—see the earlier discussion of “Children’s Online Activi-
ties” in the section on “Internet Usage among Youth”).
We also looked at the extent to which children’s parents con-
trolled or restricted their access and use of the Internet. To do
this, we created a measure that reflected how many restrictions
the parent of each child reported imposing in the home, rang-
ing from 0 (parent sets no restrictions on child’s use of the Inter-
net) to 4 (parent implements all of the control mechanisms
detailed previously under “Parental Control of Children’s Access
to and Use of the Internet” in the section on “Internet Usage
among Youth”).
Young people’s prior experiences with online information
and its evaluation were also included here in order to under-
stand their effect on credibility perceptions. Specifically, we
asked children whether they had ever had a bad experience
using some information they found online that turned out not
to be credible, or whether they had ever heard of this happen-

ing to others. We also asked them about whether they have had
instruction in evaluating the credibility of Internet information
and parents’ reports of how often they talk to their children
about the trustworthiness of information online (see “Trust and
the Negative Consequences of False Information Online” in the
previous section, “Perceived Trust and Credibility of Web-based
Information”).
Several personality traits were also explored for their contribu-
tion to young people’s credibility beliefs and practices, includ-
ing cognitive dispositions or “thinking styles” that have been
shown to influence how people approach information. “Need
for cognition,” for example, reflects the degree to which people
engage in and enjoy thinking deeply about problems or infor-
mation and, thus, are willing to exert effort to understand and
scrutinize information. Another personality trait that we exam-
ined, “flexible thinking,” measures how willing people are to
consider opinions that differ from their own, which we felt
might impact the way they process contradictory or contrasting
information when judging credibility online. “Faith in intu-
ition” reflects people’s tendency to trust based on their first
impressions, instincts, and feelings. Survey items were adapted
from standard measures of these factors (e.g., Epstein et al.
1996; Kokis et al. 2002) and were pilot tested to ensure that
children ages 11 to 18 could comprehend them easily, as noted
in the overview section of the earlier chapter, “Research
Approach.”
A final personality trait examined young people’s propensity
to trust others, which was measured by questions asking kids
how much they feel other people can be trusted, either online
or in person (see “Trust and the Negative Consequences of False
Information Online” in the previous section, “Perceived Trust
and Credibility of Web-based Information”).
The last factor we considered was strategies or methods for
evaluating credibility, or how the process of evaluating informa-
tion online influences the assessments that young information
consumers make. We analyzed two sets of strategies for infor-
mation evaluation. First, we posed the question, “When you
decide what information to believe on the Internet, do you . . .
[give careful thought to the information, rely on your gut feel-
ings, ask for help from other people, etc.].” Research in cogni-
tive psychology, information processing, and, especially,
adolescent decision making (Jacobs and Klaczynski 2005), indi-
cates that adolescents primarily approach information analyti-
cally or heuristically when making decisions (Klaczynski 2001).
Analytic processing involves effortful and deliberate consider-
ation of information; heuristic decisions are made more quickly,
with less cognitive effort and scrutiny. Additional research sug-
gests another decision-making strategy that may be relevant to
evaluating credibility online, namely, relying on others to help
make decisions (Scott and Bruce 1995). Given the Internet’s vast
and increasing capacity for social interaction, and the impor-
tance of relying on social means of credibility assessment found
in previous work on adults (Metzger, Flanagin, and Medders,
forthcoming), we included heuristic, analytic, and social or
group approaches or strategies of evaluating credibility in our
analysis (see also “Methods of Determining Information and
Source Credibility,” in the previous section on “Perceived Trust
and Credibility of Web-based Information”).
In addition to these approaches, there exist more specific

strategies that young people may use in the context of evaluat-
ing the credibility of information online, for example, by focus-
ing more or less on certain credibility cues or elements (see
“Methods of Determining Information and Source Credibility”).
Further analyses on these various credibility cues showed that
they boil down to three strategies: evaluating credibility via
social confirmation (e.g., consulting others and looking to see if
information is from expert sources), evaluating credibility via
information quality (e.g., looking at the currency and complete-
ness of the information), and evaluating credibility via Web site
design (e.g., considering the site’s appearance and navigability).
These constituted the second set of credibility evaluation strate-
gies that we examined to see whether they play a role in young
people’s credibility perceptions and beliefs.4
Concern about Credibility

The degree to which young people are concerned about the
credibility of information online is of key interest to this proj-
ect. Beyond examining levels of young people’s concern about
whether they can trust the information they find online (which
was done in the previous section, “Perceived Trust and Credibil-
ity of Web-based Information”), we wanted to understand who
is more and less likely to be concerned about credibility, and
the degree to which credibility concern is attributable to spe-
cific Internet usage and information evaluation patterns.
Toward this end, we looked at how young people’s demo-
graphic or background characteristics, Internet usage and expe-
rience, personality traits, and strategies for evaluating credibility
influence their level of concern about credibility online. Analy-

ses showed that the type of strategies that young people use to
evaluate credibility affect their concern about credibility. Spe-
cifically, those children who are more concerned about the
credibility of Internet information tend to use a more analytic
than heuristic approach to evaluating information, and rely less
on evaluating credibility by means of social confirmation and
Web site design. Kids’ online experiences and education matter
also: having had a bad experience or even hearing about others
who have trusted bad information online, having parents talk
to them about the trustworthiness of information found online,
and having had formal instruction in credibility evaluation all
contribute to greater concern about the credibility of informa-
tion on the Internet.
These results make good sense, since personal or even vicari-
ous negative experiences with online information and formal or
informal information literacy training are both likely to sensi-
tize kids to the dangers of using information that is not credible.
Also, being more critically minded and thorough in evaluating
information may come from a sense of concern about credibil-
ity or may contribute to that concern.
The ways in which kids engage with the Internet and partici-
pate in content creation also mattered in their concerns about
credibility, although less so. More specifically, those who use
the Internet to immerse themselves in virtual worlds more often
(including playing games such as World of Warcraft) and those
who contribute information online less, show higher levels of
concern about credibility. Also, kids who were more highly
skilled and who had been online for a greater number of years
were more concerned about credibility. These results indicate

that as kids engage more, and more deeply, with various aspects
of the Internet, they may develop a healthy sense of skepticism
and concern about the believability of information available
online. This finding refutes some adults’ fears that kids will
become more accepting and less critical of Internet information
as they deepen their experience and participation in online
activities.
Only two traits, flexible thinking style and Internet social
trust, emerged as being related to kids’ level of credibility con-
cern. As kids are more flexible in considering information that
runs counter to their own beliefs and are less trusting of others
online, they express greater concern about credibility. Again,
this makes sense because attending to contradictory informa-
tion would naturally raise concern about whose view to trust, as
would having little confidence in the trustworthiness of others
online.
Interestingly, young people’s demographic characteristics did
not seem to matter much, with one exception: race made a very
small contribution to users’ concern about credibility. Kids who
reported themselves to be minorities expressed slightly greater
concern about credibility than did white children, which may
reflect subcultural differences found in many surveys for trust of
all sorts among minority populations (Alesina and La Ferrara
2002). It is noteworthy that overall age did not impact concern
about credibility (i.e., other factors accounted for differences in
credibility concern when considered collectively), despite the
fact that older kids have more online experience and more life
experience.
Beliefs about the Credibility of Online Information

While young people’s concern about the credibility of informa-
tion online seems to be driven to some extent by analytic pro-
cesses of evaluating information, this is not the case for their
actual trust of online information, both in terms of the amount
of information on the Internet they feel is credible and their
likelihood of trusting information they personally find online.
Indeed, young people’s beliefs about credibility appear to be
more a function of heuristic processes, as evidenced by the fact
that young people who rated online information as more credi-
ble tended to use a more heuristic than analytic approach to
evaluating information online. Consistent contributors to
young people’s actual credibility beliefs were evaluating infor-
mation based on the Web site’s design and using heuristic cred-
ibility evaluation strategies, such as relying on gut feelings and
making quick credibility judgments. Personality traits related to
these heuristic strategies also contributed significantly to beliefs
about credibility, whereby youth possessing lower need for cog-
nition and higher faith in intuition thinking styles rated infor-
mation on the Internet as more credible.
These results are not surprising in light of what is known
from past research on adults (see Metzger 2007), which finds a
good deal of people’s credibility evaluations are based on cur-
sory rather than thorough examination of online information.
However, the fact that heuristic processes figure so prominently
in how much online information kids find credible and how
likely they are to believe the information they find online is a
little disconcerting—particularly for digital literacy advocates
who stress the need for kids to apply critical thinking skills to
Internet-based information, due to its unique characteristics

that make discerning credible from non-credible information
particularly complex and difficult (see Metzger, Flanagin, et al.
2003). Another personality trait that influenced young people’s
views of the credibility of online information was their trusting
nature. Questions that tapped into the degree to which kids felt
others could be trusted both generally and online were signifi-
cant and positive predictors of how much of the information
online they felt was believable.
Kids’ demographic characteristics mattered more for their
actual beliefs about the credibility of online information than
they did for their concern about credibility. Specifically, young
people who were from families of higher income said they
believed more information on the Internet, and both younger
kids and girls were more likely to believe the information they
find online compared to older kids and boys, respectively.5
This could be due to differences in girls’ and boys’ Internet
usage or experiences interacting with others online, and to the
fact that older children are more likely to have had greater over-
all exposure to online information generally, and thus perhaps
more experiences with bad information, as well as being more
likely to have had formal information literacy training than
have younger children.
Indeed, the data show that Internet usage and experiences
also factor into kids’ credibility beliefs. In particular, young
people who rated themselves as more technically skilled on-
line felt Internet information was more credible, as did those
who use the Internet to visit virtual worlds more often. Past
negative experiences with false or non-credible information also
mattered in that having such experiences led kids to say that

less Internet information is believable and that they were less
likely to believe the information they found online, as one
would expect.
Beliefs about the Credibility of Other People Online

An area of particular concern among both parents and educa-
tors has focused on children’s trust of strangers they meet
online through online chat groups or forums, social networking
sites, virtual communities, and other Internet venues. As dis-
cussed earlier (see “Trust and the Negative Consequences of
False Information Online” in the previous section), we probed
kids about the extent to which they trust people they meet
online. Although children did not express a great deal of trust
of strangers on the Internet overall, kids’ specific uses of the
Internet seemed to increase their trust of strangers online, par-
ticularly uses such as spending time using the Internet to visit
virtual worlds and contributing information (to blogs, personal
Web sites, online groups, etc.). While at first glance this may
seem troubling, we also found that kids whose parents con-
trolled their Internet access and use to a greater extent showed
greater trust of strangers. These findings, coupled with the fact
that kids did not express high levels of stranger trust online
overall, may indicate that kids who immerse themselves in
virtual worlds and contribute online content interact with
strangers in spaces that are reasonably safe (or at least parent-
approved) and thus they feel they can trust the strangers they
meet in those online environments.
It also appears that the strategies kids employ to evaluate

credibility contribute to how trusting they are of strangers
online. Kids who use more group-based and heuristic credibility
evaluation methods, and those who use less analytic methods,
trust strangers more. In other words, using more methods to
evaluate credibility and being more meticulous in evaluating
information leads kids to be more cautious about trusting
strangers. Supporting this, the trait of need for cognition also
emerged as a factor in online stranger trust, such that kids who
were higher in need for cognition expressed that they are less
trusting of people they meet online than those with lower need
for cognition.
The only other trait that was found to impact stranger trust
was kids’ general propensity toward trusting others, which is
not surprising. Two demographic characteristics, race and age,
played a significant but minimal role in the degree to which
kids trust people they meet on the Internet. Older kids expressed
more trust of strangers, while white kids said they trusted
strangers less than did nonwhites.
Of course it is difficult to say exactly why these patterns
emerged without knowing more about where kids meet strang-
ers online. Overall, though, our results suggest that there may
be relatively little reason for adults to fear kids trusting the
strangers they meet online because young people seem to be
fairly aware that there are risks to being too trusting of strang-
ers, and because kids can be taught to use more rigorous credi-
bility evaluation strategies that may help increase their acumen
for deciding not only what, but whom, to trust online.
Beliefs about the Credibility of Blogs and Wikipedia as Information

Sources
As relatively new sources of online information that rely heav-
ily on the contributed knowledge of largely unknown others,
we were interested in understanding what drives young peo-
ple’s credibility perceptions of blogs and Wikipedia.
Our analyses showed that, despite kids feeling that the infor-
mation in blogs is not as credible as news and political informa-
tion in newspapers and television news overall, kids who used
fewer analytic and more group credibility evaluation strategies
(and who relied on Web site design more in their evaluation
practices), who felt more strongly that people could be trusted
online generally, whose parents talk with them more often
about the credibility of online information, and who were from
households with lower incomes were more likely to find blogs
credible than kids who did not possess these characteristics.
Among kids who have some familiarity with Wikipedia, those
who express greater trust of others online, employ more heuris-
tic credibility evaluation strategies and rely more on site design
to discern credibility, and who use the Internet more frequently
for visiting virtual worlds, believe Wikipedia to be more credible
than kids who do not. Grades were the sole demographic factor
at play here, where kids with higher grades rated Wikipedia as
more believable than did kids earning lower grades in school.
Finally, kids who have had a bad experience themselves (or who
personally know others who have had bad experiences) by
trusting information that later turned out to be false find Wiki-
pedia to be less believable compared to young people who have
not had these negative online experiences.
While these results show that kids’ perceptions of the credi-

bility of blogs and Wikipedia are each affected by some unique
factors, both show that heuristic processes or strategies of deter-
mining credibility, as well as individuals’ propensity to trust
others online, are prominent drivers of children’s credibility
judgments of these two relatively new Web-based information
resources.
Factors Contributing to Children’s Credibility Evaluation Methods

Because the methods by which kids evaluated credibility online
emerged as a significant factor in young people’s credibility
attitudes and judgments across several of the aforementioned
analyses, we wanted to understand better what leads to particu-
lar styles or methods of credibility assessment. To do this, we
explored what predicted whether children used a more ana-
lytic, heuristic, or group-based (i.e., social) strategy when evalu-
ating online information credibility.
As discussed in the introduction to this section, using an
analytic strategy to evaluate credibility is somewhat effortful,
since it involves careful inspection of the information and its
author or source (e.g., gathering a lot of information and double
checking it). Young people who tend to use this strategy exhibit
a number of related behaviors and traits; for example, they
enjoy thinking hard about things, are more flexible thinkers,
get higher grades in school, and have more technical Internet
skill. They are also older and look more to the quality of the
information and less to site design when evaluating the credi-
bility of information online. In other words, a pattern of ana-
lytic evaluation appears to be related to individuals’ intellectual
prowess and experience, which comes as no surprise. Other fac-

tors that predict whether someone uses analytic evaluation
strategies include skepticism about whether others online can
be trusted, using the Internet for commercial purposes (which
may breed skepticism), and being from lower, rather than
higher, income households.
Heuristic credibility evaluation strategies tend to be based on
emotion and are made quickly, without much consideration of
evidence, sources, or information. Major predictors of this style
of evaluation tended to fall into the category of personality
traits. For example, the trait of trusting one’s instincts or
hunches and going by one’s gut feelings to evaluate informa-
tion were the most significant factors predicting heuristic evalu-
ation strategies. Related to this, using Web site design cues to
guide credibility decisions, and not relying on information pro-
vided by experts were also important. Another trait, need for
cognition, negatively predicted using a heuristic strategy, such
that those who avoid thinking hard about problems and do not
enjoy activities that are cognitively demanding used this strat-
egy more. Finally, individuals’ trust of others online increased
their tendency to evaluate credibility heuristically.
The last strategy we examined was using group-based meth-
ods of evaluating credibility (i.e., seeking the advice of others to
help discern whether some information or person online is
credible). Here, various patterns of Internet usage and experi-
ence were the best predictors. Contributing information online,
engaging in social exchange, and using the Internet for “vir-
tual” purposes all resulted in relying to a greater extent on
others to help discern credible information online. This makes
sense, of course, as kids who participate in these sorts of online

activities would be likely to have a larger and more (inter)active
social circle that they can turn to for advice and guidance while
online.
Group-based credibility assessments were also predicted by
greater technical skill and trust of others online, as well as by
more life experience (age) and years of Internet experience.
Moreover, kids who indicated they used group-based strategies
were more likely to look at Web site design and to rely on social
confirmation, and they were less likely to consider the quality
of the information in determining credibility. Children using
social evaluation strategies were also less disposed to relying on
gut feelings for trust in their general lives.
Summary
As we suspected, data show that young people’s uses of the
Internet, demographic and personality traits, and specific strat-
egies for evaluating credibility all played a role in the judg-
ments that children make about whether to trust information
and people they encounter online.
Among the more consistent predictors of both credibility
concern and beliefs were the personality traits of need for cog-
nition and kids’ propensity to trust others, either offline or
online. Online experiences also mattered, as credibility beliefs
seem to be shaped by past negative experiences with finding
and using bad information. The old adage of “once bitten, twice
shy” definitely seemed to operate in this arena.
Usage was important too, and in particular using the Inter-
net to immerse oneself in virtual environments (i.e., playing
online games, participating in online communities) was associ-

ated not only with greater concern about credibility, but also
with stronger beliefs that online information is credible. This
suggests that, contrary to some views, online gamers and par-
ticipants in virtual worlds have an optimistic yet appropriately
skeptical sense of the credibility of information and others
online.
The most consistent influence on young people’s credibility
perceptions was the strategies kids used to evaluate information
(which were themselves influenced by a number of the other
credibility predictors, such as traits like need for cognition).
Whether kids used analytic, heuristic or, in a few cases, group-
based credibility assessments impacted their credibility concern,
as well as their beliefs about credibility. Interestingly, although
concern about credibility appears to be driven by analytic cred-
ibility evaluation processes, actual beliefs about the credibility
of information found is dictated by more heuristic processes.
This suggests that while most kids take the idea that they should
be concerned about credibility seriously (by invoking a system-
atic and analytical approach), those who feel the Web is more
credible arrive at that judgment by taking a more lax approach
in actually evaluating the information they find online.
Finally, it is interesting that demographic characteristics did
not emerge as particularly important or consistent predictors.
This suggests that, at least in the area of credibility perceptions,
there is little evidence of a digital divide. Similarly, Internet skill
and frequency/duration of Internet usage were minimal factors
in kids’ credibility beliefs, as was instruction in information lit-
eracy practices.
Child/Parent Dyads and Credibility Assessments
Past research has demonstrated that an individual tends to feel

that he or she is less susceptible to negative influence than are
other people. This phenomenon has its roots in a cognitive
process known as the “optimistic bias” (Weinstein 1980), which
is the tendency to see oneself as less likely than others to expe-
rience negative life events (or, conversely, more likely to experi-
ence positive life events).
Research on the notion of optimistic bias has thoroughly
examined its impact on the beliefs and behaviors of individuals
in both health (Clarke et al. 2000; Weinstein 1982) and non-
health (Weinstein 1980) contexts, and has demonstrated the
stability of this phenomenon across a wide range of demo-
graphic variables, including age, sex, and education (Weinstein
1987). However, little research has focused on the occurrence
of the optimistic bias in a digital media environment (for
an exception, see Campbell et al. 2007) and research to date
has not examined child-parent dyads with regard to this
phenomenon.
To fill this void, we were interested in seeing if the same psy-
chological processes underpinning the optimistic bias phenom-
enon operate in the context of judging the credibility of
information online. We were particularly interested in chil-
dren’s perceptions of their own ability to evaluate the credibil-
ity of information online, and to differentiate between good
and bad information compared to (a) their parents and to (b) a
typical Internet user. For children, survey questions designed to
assess this were as follows:
Who is more likely to believe false information on the Inter-

net, you or your parent?
Who is more likely to question the information they find on
the Internet, you or your parent?
Who is better at figuring out which information is good or
bad on the Internet, you or your parent?
Who is more likely to believe false information on the Inter-
net, you or a typical Internet user?
Who is more likely to question the information they find on
the Internet, you or a typical Internet user?
Who is better at figuring out which information is good or
bad on the Internet, you or a typical Internet user?
Moreover, we took advantage of our study design to query

parents and their children to look at whether parents also
exhibited an optimistic bias. To do so we posed a parallel set of
six questions that compared parents’ perceptions of their own
ability to discern credible information online to that of (a) their
child, and (b) a typical Internet user.
Our data on the optimistic bias are unique in that they are
the first to examine this phenomenon in the context of credibil-
ity judgments in digital media environments, as well as to pro-
vide a detailed look at the phenomenon in the context of
parents and children, and not just between individuals and
“typical others.”
Children’s and Parents’ Comparisons to Each Other

Figure 26 shows that across all ages children consistently feel
that their parents are better than they are at figuring out which
information is good or bad online. Similarly, parents seem to

agree, as they also consistently reported that they were better
than their children at discerning good from bad information on
the Internet. So, while the parents’ overall responses display
evidence of an optimistic bias, the children’s responses do not.
There is an interesting trend with age, however. As children
get older, the gap between their perceptions of their own and
their parent’s ability to differentiate between good and bad
information online narrows. So, while younger children feel
their parents are significantly better at discerning credible
information, by about age 16 they feel they are about equally
good as their parents at doing so. Interestingly, while parents do
My child/parent
Who is better at figuring out good
is much better
than me
or bad information?
We are equally
good
I am much better
11 12 13 14 15 16 17 18
Age of child
Child's response Parent's response
Figure 26
Child/parent differences in discerning good versus bad information on-
line
perceive their children’s ability to tell good from bad informa-

tion online as improving with age, they never report feeling
their children are their equal in this regard.
When it comes to who is more likely to believe false infor
mation online, we see a similar pattern of responses, such that
parents always feel their children are more likely to believe
false information than they are. Younger children feel that they
are slightly more likely than their parents to believe false infor-
mation, but again by age 16 kids feel they are about as likely
as their parents to be tricked by false information on the Inter-
net (see figure 27). Notably, at age 16 children actually cross
over the midpoint of the scale, indicating that on average
they feel they are less likely than their parents to believe false
My child/parent
is much more
Who is more likely to believe
likely than me
false information online?
We are equally
likely
I am much
more likely
11 12 13 14 15 16 17 18
Age of child
Figure 27
Child/parent differences in discerning false information online
information online. On average, 17- and 18-year-olds also ap-

proach this threshold quite closely.
When asked about who is more likely to question informa-
tion they find online, a slightly different pattern emerges. Figure
28 shows that while parents still consistently believe they are
more likely than their children to question information found
online—and children seem to agree (although not as strongly)—
the lines do not change as dramatically across the age groups as
they did for the other comparison questions.
This indicates that both parents’ and kids’ perceptions of how
likely they are to question information online stay fairly consis-
tent across the age groups, with only slight upward and down-
ward trends for the parents and their children, respectively.
My child/parent
Who is more likely to question the
is much more
information they find online?
likely than me
We are equally
likely
I am much
more likely
11 12 13 14 15 16 17 18
Age of child
Figure 28
Child/parent differences in the likelihood of questioning information
online
This could suggest a generational difference in the level of

skepticism by which parents (so-called digital immigrants) and
children (so-called digital natives) approach online information,
or at least a difference in the amount of effort that each group is
willing to expend in thinking about the credibility of the infor-
mation they find. Effort, of course, may be tied to the type of
information sought online, with the idea that people will
expend more energy questioning information that is more per-
sonally and objectively consequential (e.g., health, financial
information). Moreover, it is likely that adults spend more time
overall while searching online for these types of information
than do children.
Children’s and Parents’ Comparisons to Typical Internet Users

In comparison to a typical Internet user, even the youngest
children saw themselves as slightly better on average in their
ability to figure out which information is good and bad online,
thus showing an optimistic bias. Older children viewed them-
selves as even more capable than a typical Internet user in this
regard (see figure 29).
Parents of children across all age groups also felt they were
better than the typical Internet user in discerning credible infor-
mation online. That said, by age 17 kids’ and their parents’ rat-
ings of their own ability are very similar to each other, and they
both see themselves as more skilled than typical Internet users.
Parents and children across all ages felt that they were less
likely than a typical Internet user to believe false information
online, again showing clear evidence of an optimistic bias for
both groups. In addition, parents always view themselves as
A typical user
Who is better at figuring out good
is much better
you or a typical Internet user?
than me
or bad information online,
We are equally
good
I am much
better
11 12 13 14 15 16 17 18
Age of child
Figure 29
Child/parent differences in discerning good versus bad information on-
line, compared to a typical Internet user
more capable in this regard than their children. Interestingly,

older kids (ages 15 and up) and their parents were both slightly
more likely than younger kids and their parents to report a
favorable comparison to typical Internet users when it came to
believing false information online, as can be seen in figure 30.
An optimistic bias was also seen for both groups, although
more strongly for parents, when they were asked whether they
or a typical Internet user were more likely to question informa-
tion found on the Internet. As seen in figure 31, parents consis-
tently rated themselves as more likely to question information
than a typical Internet user, as do their children, and there is an
increasing tendency to do so as kids get older. Again, parents
always view themselves as more capable than their children.
information online, you or a typical
A typical user
Who is more likely to believe false
is much more
likely than me
Internet user?
We are equally
likely
I am much
more likely
11 12 13 14 15 16 17 18
Age of child
Figure 30
Child/parent differences in the likelihood of believing false informa-
tion online, compared to a typical Internet user
Summary
Overall, compelling evidence exists for an optimistic bias in
individuals’ perceptions of their ability to evaluate the credibil-
ity of information online, especially on the part of parents.
Clear evidence for an optimistic bias exists across several mea-
sures, for parents compared to their children and for both par-
ents and children compared to a typical Internet user.
Although compared to their children parents might be realis-
tic in their estimations, it is unlikely that, as a group, the par-
ents and kids surveyed are uniformly more accurate in their
judgments of credible information online than are typical Inter-
net users. Our data are consistent with some of the preliminary
trends that have emerged from research into the optimistic bias
A typical user
is much more
Who is more likely to question
you or a typical Internet user?
likely than me
information online,
We are equally
likely
I am much
more likely
11 12 13 14 15 16 17 18
Age of child
Figure 31
Child/parent differences in the likelihood of questioning information
online, compared to a typical Internet user
in a digital media context, which has shown that people believe

themselves to be less likely to have their credit card or identity
stolen online, and more likely to avoid being misled by infor-
mation online, as compared to their peers (Campbell et al.
2007).
That said, it is clear that more research into this phenome-
non needs to be conducted, in order to solidify the trends that
have emerged in this study and to better understand how the
optimistic bias operates in an environment in which much of
our information about others is produced and consumed elec-
tronically. However, our results serve as an important and com-
prehensive first step in this endeavor, offering a valuable look at
the nuances of the optimistic bias both within and across

generations.
Web Site Exposure and Evaluation
In order to simulate children’s Web-browsing experiences, and

to evaluate their reactions to specific Web content, we included
a quasi-experimental component in our survey. We presented
each child with two stimuli, one at a time: first, they viewed a
screenshot of a Web page, which was presented as a “picture of
a Web page from the Internet.” Depending on the experimental
condition to which they were randomly assigned, the child saw
either an image from an online encyclopedia or from Amazon
.com, followed by questions about the site they had seen (as
described in greater detail later). Second, children were pre-
sented with another Web page screenshot, this time of a “hoax”
site that currently exists on the Web. Children were again asked
some questions about the site they viewed. To avoid any influ-
ence on our respondents from other questions on the survey
dealing with credibility that might cue them to this concern,
this portion of the survey preceded all other sections.
Online Encyclopedia Exposure and Reactions

A critical feature of the contemporary Internet environment is
the ability of users to be both information consumers and infor-
mation providers. Indeed, the Internet’s very design facilitates
wide-scale collaboration among individual users (Flanagin, Flan-
agin, and Flanagin 2010), which can take a number of forms,
ranging from the provision of valuable consumer information
to the organization of political protests. Such collaborative

efforts often rely on individuals pooling their efforts to create
collectively held resources that none could produce without the
aid of others (see, for example, Benkler 2006 and Jenkins 2006).
One venue in which collectively produced information has
burgeoned is in online encyclopedias such as Wikipedia, where
anyone can anonymously contribute encyclopedia entries or
edit those provided by others. As mentioned in an earlier sec-
tion (“Perceived Trust and Credibility of Web-based Informa-
tion”), in its short history Wikipedia has risen to among the top
10 most popular Web sites in the United States today, with
more than 3 million user-generated entries (Alexa 2009; Quant-
cast 2009).
Given that all of the content in Wikipedia is provided by
anonymous individuals, there has been a great deal of contro-
versy and concern regarding the credibility of this information,
particularly as compared to more established encyclopedias
such as Encyclopaedia Britannica, which also has an online pres-
ence. In addition, exclusively online encyclopedias like Citizen-
dium have also emerged, relying on user-contributed content
that is not anonymous and that is provided or vetted by experts
prior to its acceptance.
In spite of concerns that user-generated content may be less
credible than its expert-produced counterparts, studies suggest
that the differences may not be particularly great. For example,
research has shown that user-created entries in Wikipedia have
been judged to be nearly as accurate as well-regarded print ency-
clopedias like Encyclopaedia Britannica (Giles 2005), and entries
from Wikipedia have been evaluated as relatively credible, and
even more so by area experts than by non-experts (Chesney

2006).
In light of the relatively heavy use of and reliance on Wikipe-
dia by children (e.g., 84 percent have used it to look up informa-
tion, as detailed in the previous section), and the fact that no
research to date has systematically considered children’s use of
online encyclopedias, we endeavored to assess children’s percep-
tions of the credibility of information in online encyclopedias.
To do so, we performed several different quasi-experiments.
In the first, we had a subset of children who took the survey
view a screenshot of an encyclopedia entry, which was pre-
sented as coming from one of three different online encyclope-
dias. In reality, encyclopedia entries were actually identical in
all cases within each experimental condition, and were derived
from information gleaned from all three of the sources.
The notable difference among the encyclopedias was the pur-
ported source of the information, which was reflected in the
description of the encyclopedia that children were given: chil-
dren were instructed that they would see a picture of a Web
page from (a) “the online encyclopedia Wikipedia, where anyone
can add or change information at any time without giving their
real names,” or from (b) “the online encyclopedia Citizendium,
where anyone can contribute entries, as long as they are identi-
fied by their real names. All contributions, however, are
reviewed by experts before being accepted,” or from (c) the
online version of “Encyclopaedia Britannica, whose entries have
been contributed by respected experts worldwide since 1768.”
To ensure that children in this study understood these differ-
ences, they were asked to later identify which method of
selection for entries was used by the encyclopedia they viewed.

Those children who did not correctly identify the method of
selecting entries for the encyclopedia (41 percent of respon-
dents) were excluded from all further analyses.
To assess differences across encyclopedia entry topic, three
different types of entries were provided and were assigned ran-
domly (an entry on an entertainment topic, an entry on a news
topic, and an entry on a health topic). Tests showed no differ-
ences across encyclopedia entry topic, so these entries were col-
lapsed for subsequent analyses.
Encyclopedia entries were also constructed to be either one-
sided in their presentation of information or balanced in their
presentation. Pretests with a different sample of young adults
confirmed that stories were perceived as appropriately one-sided
or balanced. Data from the children in the present study con-
firmed this as well. Figure 32 shows an example encyclopedia
entry, presented as a balanced entry on the topic of autism
originating from Encyclopaedia Britannica.
Results from this quasi-experiment yielded two interesting
discoveries. First, children found the balanced encyclopedia sto-
ries to be significantly more believable than the one-sided ver-
sion of the same stories, as illustrated in figure 33. Second,
children found the entries that they believed had originated
from Encyclopaedia Britannica to be significantly more believable
than those they believed originated either from Wikipedia or
Citizendium. However, children did not distinguish between
encyclopedia entries they believed originated from Wikipedia or
Citizendium, in terms of how believable they thought they were.
Figure 34 shows this relationship. The relative balance of stories
Figure 32
Example encyclopedia entry
and their source did not demonstrate any statistically signifi-

cant interaction effects.
These findings indicate that children readily attended both
to the information content in the entries and to the source of
the information when asked to evaluate its credibility. Content
was important inasmuch as children rated the balanced stories
as more credible than one-sided presentations, consistent with
past studies that demonstrate higher perceptions of credibility
for more balanced presentations, at least within some contexts
A whole lot
How much do you believe
this information?
A lot
Some
A little
Not at all
One-sided entry Balanced entry
Figure 33
Credibility of one-sided versus balanced encyclopedia entries
A whole lot
A lot
this entry?
Some
A little
Not at all
Wikipedia Citizendium Encyclopaedia
Britannica
Figure 34
Credibility of encyclopedia entry by perceived source
(Kamins et al. 1989). Information source was also important,

confirming past research that has demonstrated differences in
credibility based on the perceived information source (Flanagin
and Metzger 2007). However, it is unclear whether any precon-
ceived notions they had about the three sources in this study
attenuated or exacerbated the effect of the source above and
beyond that which our descriptions were designed to provoke.
We next endeavored to assess whether encyclopedia entries
that actually originated from these various online sources (as
opposed to those we created) were viewed differently among
children with regard to their credibility. We also evaluated
whether it made a difference from which among the three online
encyclopedias children believed the entry to have originated.
To do this we selected actual entries on two different topics
(global warming and homeopathy) from each of the three
online encyclopedia Web sites and edited them very slightly to
be of roughly the same length (content was not changed). Tests
once again showed no differences in believability across ency-
clopedia entry topic, so data from these entries were collapsed
for subsequent analyses.
We showed children a screenshot of one encyclopedia entry,
presented as if it originated from one of the three encyclope-
dias. However, the encyclopedia entry may have actually origi-
nated from any of the three encyclopedias. In this manner, we
created 18 different page images, representing each possible
combination of encyclopedia entry topic (global warming or
homeopathy), original encyclopedia entry source (Wikipedia,
Citizendium, or Encyclopaedia Britannica), and the placement of
the encyclopedia entry (Wikipedia, Citizendium, or Encyclopaedia
Britannica). Figure 35 shows an example of an entry from Wiki-

pedia on the topic of global warming, although the text for this
entry actually came from Citizendium. Again, only those chil-
dren who correctly identified the encyclopedia’s actual method
of selecting entries for the encyclopedia were included in subse-
quent analyses.
Results showed that, by itself, where the entry actually origi-
nated (i.e., the original and actual source of the entry) was irrel-
evant to how believable the entry was found to be by children.
Thus, the source of the encyclopedia entry was not important
Figure 35
Example encyclopedia entry
with regard to its perceived believability. The placement of the

entry, however, was critical in children’s credibility evaluations:
encyclopedia entries were assessed as significantly less believ-
able when placed on Wikipedia’s site than when they were
placed on either Citizendium’s or Encyclopaedia Britannica’s sites
(and children did not report statistically significant differences
between these two sites), as illustrated in figure 36.
Moreover, the entry placement also interacted in meaningful
ways with the entry source, such that, for example, entries actu-
ally originating from Wikipedia were perceived as significantly
more believable when they appeared on Citizendium’s Web page
than if they appeared on Wikipedia’s page, and even more believ-
able if they appeared to have originated from Encyclopaedia Bri-
tannica. Put another way, the encyclopedia entries from
Wikipedia were seen as significantly more believable than those
A whole lot
A lot
this entry?
Some
A little
Not at all
Britannica
Placement of encyclopedia entry
Figure 36
Credibility of encyclopedia entry by placement of entry
from both Citizendium and Encyclopaedia Britannica, but only

when children thought they were actually from Citizendium or
Encyclopaedia Britannica. Figure 37 illustrates these results.
Similar to the previous encyclopedia quasi-experiment, chil-
dren in this case show strong evidence of attending carefully to
the entry (in this case, its placement and apparent source).
Interestingly, the fact that Wikipedia content was deemed more
credible if children thought it originated from Citizendium, and
most credible under the banner of Encyclopaedia Britannica,
could be taken as signaling the high quality of Wikipedia infor-
mation, despite popular cries that it cannot be of high credibil-
ity since it is provided by anyone who cares to contribute it.
A whole lot
A lot
this entry?
Some
A little
Not at all
Britannica
Placement of encyclopedia entry
Actual source of encyclopedia entry:

Wikipedia Citizendium Encyclopaedia Britannica
Figure 37
Credibility of encyclopedia entry by actual encyclopedia source
Commercial Web Site Exposure and Reactions

People are increasingly relying on the Internet for commercial
information and e-commerce transactions that range from
small personal items to home purchases. Although due to
financial constraints and other factors children constitute only
a small proportion of consumers online, the knowledge and
habits learned as children are likely to influence their use of the
Internet in this capacity well into adulthood. It is therefore
important to understand their perceptions of the credibility of
commercial Web site information and the factors they find
important in their evaluative processes.
Overall, children’s consumption of commercial information
online is low, and they do not find it to be very credible as com-
pared to other types of information. Yet, kids believe that com-
mercial information is better retrieved online than via other
sources. Moreover, both their use of commercial information
and their faith in its credibility increase with age. Finally, they
tend to rely on product endorsements by other Internet users
when making purchasing decisions, suggesting the need to
assess their facility in evaluating such cues in relatively natural-
istic commercial environments (see the previous sections of this
report for more complete discussion of these findings).
To gauge the degree to which children make credibility
assessments of commercial information online, and the factors
that influence their evaluations, we presented a subset of the
children surveyed with one screenshot from a set of product
pages from Amazon.com, which were slightly modified in order
to highlight particular features available on these pages. Three
different products were shown (a digital camera, an electric
toothbrush, and rolling luggage), in order to determine if the

type of product in question influenced young people’s assess-
ments of information credibility and product quality.
Given recent attention in studies of Web-based information
credibility to the influence of others’ opinions in credibility
assessments (Flanagin and Metzger 2008; Metzger, Flanagin,
and Medders, forthcoming), we focused on the prevalence and
nature of user-generated feedback in forming children’s assess-
ments of commercial products. Specifically, we varied (a) the
number of ratings and (b) the average rating provided about
products by other users, by altering this information on the
Web page screenshots used in the study. The pages thus showed
the number of user ratings as 4, 16, 102, or 1,002 and average
“star” ratings (on a 1–5 scale, where 5 is the best rating) of 1.6,
2.23, 3.0, 3.68, 4.4, 4.84, or 5.0. In this manner, we created 84
different page images, representing each possible combination
of number of ratings, average ratings, and product. Factors other
than these were held constant across all pages. Because we
found that children’s interest in each product varied, we statisti-
cally controlled for interest in the product in all analyses. An
example page viewed by children in the study is shown in figure
38, demonstrating a digital camera presented as receiving an
average user rating of 3.68, across 102 individual user ratings.
Overall, children found user-provided commercial informa-
tion to be credible and important, demonstrated by the facts
that (a) their assessments of product quality and (b) their likeli-
hood of buying the product in question depended on the prod-
uct ratings they viewed. There were, however, distinct effects for
the number and nature of product ratings: specifically, there
Figure 38
Example product Web page
were strong differences in children’s assessment of product

quality and likelihood of purchasing the product based on the
average ratings of the product, but only extremely minor differ-
ences based on the number of ratings the product received. Thus,
although the average ratings positively influenced children’s
product quality assessments and their likelihood of purchasing
the product, the number of ratings for the most part did not.
This difference is illustrated in figure 39, which shows the aver-
age perceived product quality by the number of ratings, and
figure 40, which demonstrates the average perceived product
quality by the average product rating. Age was not a major
factor in these findings, although there was some evidence that

older children were more influenced by the combination of aver-
age ratings and the number of ratings together.
Although ratings are clearly credibility cues affecting chil-
dren’s product quality assessments and intent to purchase,
other factors are reported to be even more important as children
look for things to buy online. As illustrated in figure 41, ratings
are seen by children as less important than a product’s price and
product details, but more important than who makes or sells
the product.
The quality is
much better
than average
How would you rate the
quality of this product?
The quality is
about average
The quality is
much worse
than average 4 ratings 16 ratings 102 ratings 1,002 ratings
received received received received
Number of ratings for the product
Avg. rating of 1.6 Avg. rating of 4.4

Avg. rating of 3.68
Figure 39
Product quality by number of ratings, across average product ratings
The quality is
much better
than average
How would you rate the
quality of this product?
The quality is
about average
The quality is
much worse
than average 1.6 2.32 3.0 3.68 4.4 4.84 5.0
Average product rating
4 ratings received 102 ratings received

16 ratings received 1,002 ratings received
Figure 40
Product quality by average product ratings, across number of ratings
As with the encyclopedia quasi-experiments, children dem-

onstrated an ability to attend to specific aspects of the informa-
tion available to them that then played a major role in
determining credibility and had an influence on their attitudes.
In this case, the average rating for a product seemed to override
concern for the number of people who rated it (though there
was limited evidence that older children were slightly less prone
to this), a potentially detrimental oversight given the question-
able accuracy of such ratings under circumstances when a small
number of people provide feedback. For example, one disgrun-
tled consumer has a very large impact on the overall rating of a
product when there are only 4 ratings and a very small impact
Very
important
Importance of credibility cue
Important
Somewhat
important
A little
important
Not at all
important
Who sells Who makes Customer Product details Price
the product the product reviews (size, color, etc.)
or ratings
Figure 41
Importance of credibility cues for purchasing decisions
when there are over 1,000, so decoupling the number of ratings

from the average rating, as the children in our survey did,
reflects a critical deficiency in young people’s ability to correctly
interpret the meaning of online ratings.
Hoax Web Site Exposure and Reactions

Research (see Krane 2006; Leu et al. 2008) has shown that even
among seventh-graders (typically 12 years of age or so) identi-
fied as their schools’ most proficient online readers, many fail
to distinguish fake from legitimate information online. In a
study of 25 such students, for example, all believed the infor-
mation on a hoax Web site advocating the protection of the
Pacific Northwest tree octopus, with 96 percent of the students
rating the site as “very credible” and recommending the site to
others. Moreover, even after learning that the site was a fake,
these children had difficulty indicating why this was the case,
in spite of clear cues present on the site.
To further assess the degree to which children believe ficti-
tious information online, we presented children with one of
two “hoax” Web sites6 currently present on the Internet. The
first site detailed “the first male pregnancy,” and included infor-
mation about the pregnant individual and testimonials and
links to media coverage of the pregnancy (see figure 42). The
Figure 42
Male pregnancy hoax Web site
Figure 43
Rennets hoax Web site
second hoax site was devoted to a cause to “save the rennets,”

described as “small hamster like rodents” who are used in the
production of cheese (see figure 43). Tests showed that these
two hoax sites did not differ in terms of their believability,
although there were some other minor differences between
them (as detailed later.)
Six percent of children reported having seen the male preg-

nancy site before the study and 5 percent reported they had
seen the rennet site previously. After removing these children
from subsequent analyses, an additional 2 percent for the male
pregnancy site and 7 percent for the rennet site indicated, when
asked, that they had taken time out from responding to the
survey to search the Web for information on the sites. Although
these children’s responses were also removed from further anal-
yses, this is an intrinsically interesting result in itself, since it
demonstrates children’s active use of the Internet to verify
information whose credibility may on its surface be suspect.
Unclear, however, is whether or not this bit of additional infor-
mation seeking produced lower or higher levels of trust in the
Web site. Neither age nor sex of the child was indicative of the
likelihood to seek out additional information on these hoax
sites.
Overall, and somewhat unlike past studies, children were rel-
atively unlikely to believe the information on these hoax sites.
For the male pregnancy site, the average believability score was
1.97 (on a 5-point scale), corresponding to an indication that
children believed the information on the site “a little bit” (see
figure 44). Moreover, 48 percent found the information to be
“not at all believable” and an additional 41 percent reported
that they believed the information either “a little bit” or “some.”
Eleven percent of children, however, reported believing the
information either “a lot” or “a whole lot.” When asked about
particular features of this hoax site that might inform their
assessments, children on average tended to mildly disagree with
statements that the information on the site was reasonable,
authoritative, well written, and similar to their own beliefs.

Also, they tended on average to disagree that the site looked
good and that there was evidence on the site supporting the
claims made.
Results were similar for the rennet site, where the average
believability score was 1.93, indicating that children believed
the information on the site “a little bit.” Forty-nine percent
found the information to be “not at all believable” and an addi-
tional 39 percent reported that they believed the information “a
little bit” or “some.” Eleven percent of children did report
believing the information either “a lot” or “a whole lot,” how-
ever. By and large, children on average tended to mildly disagree
with statements that the information on the site was reason-
able, authoritative, and based on evidence, even though they
tended to find the information to be somewhat well written and
60%
50%
40%
30%
20%
10%
0%
How much do you believe the information on this Web site?
Figure 44
Credibility of male pregnancy hoax Web site
the site to look good. On average, they found that the informa-
tion on the site was not similar to their own beliefs. Figure 45
shows the extent to which children found the site to be
believable.
Finally, we were interested to know whether there were cer-
tain background and demographic characteristics, Internet
usage and experience patterns, personality traits, or methods of
evaluating credibility that were related to believing the hoax
sites.
We found no differences in income, race, or school grades
among children on how believable they found the information
on either of the hoax Web sites to be. However, females and
younger children were more likely to believe the hoax sites.
More specifically, 11-year-olds were significantly more likely to
60%
50%
40%
30%
20%
10%
0%
How much do you believe the information on this Web site?
Figure 45
Credibility of rennets hoax Web site
believe information on the rennet hoax site than were 15- to

18-year-olds.
Credibility-evaluation strategies were also related, such that
those who used more heuristic methods of credibility evalua-
tion (i.e., rely on site design) found the hoax cites to be more
credible than those who used analytic methods of evaluating
online information.
Two surprising results were: (a) greater self-reported skill in
using the Internet and (b) having had information literacy
training by a teacher or parent resulted in higher believability of
the hoax sites. In some ways, this is consistent with the work of
Leu et al. (2008) who found that even high-performing online
readers who had formal information literacy training had trou-
ble discerning whether a hoax site was credible. This suggests
that a new strategy for digital literacy training may be
necessary.
Results from this quasi-experiment are in many ways quite
heartening, although there remain indications for some con-
cern. A majority of children displayed an appropriate level of
skepticism when presented with either of the hoax sites, a trend
that contradicts prior research on this type of site. Additionally,
they seemed able to identify important credibility cues on these
Web sites, such as whether information was reasonable, author-
itative, and well-evidenced. Perhaps most important, and con-
trary to past experiments looking at information verification
behaviors (Flanagin and Metzger 2007), a number of respon-
dents even reported searching the Internet for more informa-
tion on these sites after initial exposure to the site. This
demonstrates a familiarity and comfort with information-search
strategies and credibility-assessment behaviors that is hoped for

from digital natives. Nonetheless, approximately 10 percent of
children still believed these hoax sites either “a lot” or “a whole
lot,” indicating some lingering and important concerns, espe-
cially for younger Internet users.
Summary
This series of quasi-experiments was designed to place children
to some degree in the kind of information environment they
might reasonably be expected to occupy during their time on
the Internet. Children were tested on their ability to detect both
good and bad information, and for the most part seemed able
to do so, across both informational and commercial contexts.
Their tendency to overlook some of the nuances of information
presented on the Amazon.com pages might be explained as the
byproduct of a general unfamiliarity with that context, as chil-
dren do not typically have the resources necessary to engage in
a large number or variety of e-commerce transactions. And,
they seemed able (at least implicitly) to pick up on the nuances
of balanced versus one-sided information in the encyclopedia
entries, as well as important information-sourcing cues. Finally,
children were (mostly) successful in seeing through the hoax
sites they encountered. Although such sites make up a minority
of the Internet, the skills children reported drawing upon to
handle them are universally applicable.
Conclusions and Implications
This report describes in detail how the activities that young

people between the ages of 11 and 18 engage in online, as well
as a number of their traits and attitudes, affect their assessments
of the credibility of information, and how they go about form-
ing those assessments. Results described herein are generalizable
to households in the United States with Internet access.
Summary
The research outlined in this study describes youth who have

been using the Internet for much of their lives and who use it
for a wide variety of purposes. In many ways, the results are
encouraging. For example, the young people in our survey dem-
onstrated an understanding of the potential negative conse-
quences of believing false information online, a tendency to
question information that comes from deceptive sources like
hoax Web sites, the ability to differentiate between one-sided
and two-sided information presentations, general feelings of
distrust toward strangers on the Internet, and the inclination to
106 Conclusions and Implications
put more effort into assessing the credibility of highly con

sequential information (e.g., health information) than less
consequential information (e.g., entertainment information).
Moreover, parents, teachers, and others provide children with
some level of guidance and training with regard to issues of
credibility. Consequently, worries that all adolescents are help-
less and at the mercy of unscrupulous others on the Internet
appear to be generally overstated.
On the other hand, whereas children’s concern about credi-
bility appears to be driven largely by analytic credibility evalua-
tion processes (which involve the effortful and deliberate
consideration of information), those who find Internet informa-
tion most credible use more heuristic (hasty and feeling-based)
processes to evaluate it. This finding, coupled with the fact that
most kids said that people should be concerned about the cred-
ibility of information online, suggests that while kids take the
issue of credibility seriously, actual decisions about credibility
are not always based on a stringent approach to evaluating the
information they find online.
In addition, children report being equally likely to believe
entertainment and health information online, which implies
potentially problematic outcomes since these types of informa-
tion should typically warrant different levels of skepticism. Also,
children consistently overestimate their own skill levels and
capacity to discern good from bad information as compared to
others. Such overconfidence is troubling, inasmuch as it implies
a correspondingly reduced level of vigilance or attention. And,
although most children displayed a healthy level of skep-
ticism toward the hoax sites presented to them in this study,
approximately 10 percent of the children still believed the

information on these hoax sites “a lot” or “a whole lot.” Find-
ings such as these illustrate that although youth exhibit encour-
aging signs of achieving appropriate skills and attitudes about
online information credibility, there remain important gaps in
their knowledge and abilities.
A number of factors appear to partially explain the coexis-
tence of these encouraging and discouraging findings. For
example, as kids get older, their Internet use increases both in
scope and in time spent online. This increase may be due in part
to decreased regulation by parents, and is accompanied by an
increase in the variety of tools used to assess the credibility of
information online. Older teens also trust the Internet more as
an information source than do younger kids but think that
people should be more concerned about the quality of informa-
tion online than do younger children. This might indicate that
as kids become more experienced with the Internet they have a
greater appreciation for the potential of deceptive information
online as well as greater confidence in their ability to find cred-
ible information sources.
Indeed, various forms of experience play a critical role in
youth’s credibility perceptions and information evaluation
behaviors. Kids who have been using the Internet for a longer
period of time, who spend a lot of time in virtual worlds, or who
have contributed information to an online source (e.g., a blog,
Wikipedia, etc.) think about credibility more and find more of
the information and people they meet online to be credible.
Also, older kids and kids who report having had or heard about
bad experiences online report lower levels of belief in online
information. In addition, our research indicates a positive rela-

tion between experience using the Internet and the use of ana-
lytic strategies for assessing the credibility of online information:
as kids become more experienced using the Internet, they show
more concern for the believability of information online, use
more cognitively demanding tools to assess its credibility, and
show a higher level of trust toward people and information
online. When it comes to actively processing credibility cues to
assess information credibility, practice appears to reap real
rewards.
Implications and Future Directions
Findings from this study reveal a relationship between youth,

the Internet, and credibility that is far more nuanced than pre-
vious research has suggested. Our study indicates that a combi-
nation of experience using the Internet over time and vital
cohort-related changes in youth’s cognitive development inter-
act to promote better awareness of general credibility concerns
and the ability to evaluate information found online. This has
implications for several domains, including education and the
creation of media literacy curricula, children’s use of the Inter-
net, policy formulations, and future research endeavors.
For example, based on our findings, online media literacy
programs should emphasize a structured but graduated
approach to guiding children’s use of the Internet, which
stresses the accumulation of personal experience online, early
parental involvement, and the sharing of positive and neg-
ative online experiences at an early age. Curricula should be
developed with these factors in mind, and should be assessed in

terms of developmental and experiential differences among
children.
This study also indicates that although overall experience
may be a good predictor of credibility concern, it may also lull
youth and even parents into believing they are better at discern-
ing the credibility of information online than they actually are.
Therefore, educational efforts regarding credibility evaluation
should be ongoing, and should be targeted at youth with vary-
ing experience and skill levels in order to remain relevant.
Indeed, quite different approaches appear warranted for younger
versus older children and for those with lower versus higher
online experience and skill.
It is also important to note that a number of limitations
inherent in the survey methodology color our findings. For
example, as use accumulates over time, children appear to
appraise their ability to discern good versus bad information
inaccurately. However, since survey data cannot accurately
assess people’s actual ability to find credible information suc-
cessfully, techniques other than surveys should be used to vali-
date and reveal any biases that result from this overconfidence.
A possible direction for future research into this area is to inves-
tigate youth’s evaluation of consequential information, such as
information sought for schoolwork, during an actual informa-
tion-seeking task. This could be done experimentally, via obser-
vation, or by other means.
As another example of the limitation of the survey method,
we relied on screenshots of Web pages in our quasi-experiments
to represent actual Web pages. Although this method has the
advantage of experimental control, it suffers from its non-natu-

ralistic nature. The screenshots simply cannot fully represent
children’s actual information seeking or browsing experiences,
which would require methods that retain the context of such
experiences. Considering this context could, of course, affect the
results presented here, in ways that are not entirely predictable.
Additionally, while it appears most kids were appropriately
concerned about the believability of the hoax Web sites repre-
sented in this study, there is a need to determine what charac-
teristics and contextual factors led the minority to believe this
information, above and beyond simple ignorance that may
remain irrespective of the presentation of information via the
Internet. To better understand this outcome, future research
may investigate the effect of developmental states from age 11
forward, for example, on evaluating Web site credibility.
Overall, the findings presented here not only represent the
current state of knowledge on this topic, but also serve as an
important springboard for future research. Based on our find-
ings, research should consider the development of children’s
information evaluation styles and strategies over time, differ-
ences in and the effects of parental involvement, the role of
negative experiences online, the evolution and influence of
false confidence in information evaluation abilities, and the
most appropriate educational efforts to enhance and assess
online information literacy.
Conclusion
One goal of this study was to move away from the simplistic
treatments children often receive in examinations of youth and
digital media, which cast children as either substantially more

tech-savvy than adults, and therefore as superior in their use of
digital media, or as universally vulnerable, and therefore in
need of constant protection. Such accounts are prone to unnec-
essarily provoke either alienation or outrage, depending on the
perspective taken. Our data suggest that neither view is particu-
larly warranted, and that children’s relation to digital media
with regard to credibility is significantly more nuanced than
either of these positions suggests.
In the end, and in spite of some evidence to the contrary, the
reality seems to be largely what we would hope for as citizens,
fellow Internet users, and parents: children are for the most part
aware of the issues surrounding information verity on the Inter-
net and appear generally capable of making informed and
appropriate decisions in this regard. Thus, the best strategy to
help children become more skillful Internet information con-
sumers would appear to be from a perspective that empowers
them and capitalizes on their unique upbringing in an all-digital
world. Indeed, in a future in which the information that drives
their lives is assembled, transmitted, shared, and processed digi-
tally, children need to develop the skills necessary to navigate
that information environment effectively. Perhaps the most
encouraging conclusion from our data so far is that, for the
most part, children seem to be making inroads toward that goal.
Appendixes
Appendix A: List of Tables and Figures
Tables
Table 1: Percent of participants from various U.S. geo- 14

graphic areas
Table 2: Number and percent of participants within each 14
age category sampled
Table 3: Importance of various credibility cues 52
Figures
Figure 1: Weekly usage of the Internet by age 17

Figure 2: Percentage of parents restricting their child’s 18
Internet access by age of child
Figure 3: Children’s use of Web-based information 20
resources
Figure 4: Frequency of file sharing by age 21
Figure 5: Frequency of posting original content by age 21
Figure 6: Frequency of using social networking sites by 22
age
116 Appendix A
Figure 7: Frequency of visiting virtual worlds by age 22

Figure 8: Frequency of social networking use by age 25
Figure 9: Frequency of “virtual” Web use by age 25
Figure 10: Frequency of commercial Web use by age 26
Figure 11: Frequency of online information contribution 27
by age
Figure 12: Amount of online information that children 32
find credible
Figure 13: Frequency of thinking about credibility when 33
online
Figure 14: How concerned others should be about 33
credibility
Figure 15: General social trust 34
Figure 16: General Internet social trust 35
Figure 17: Degree that children trust strangers they meet 36
online
Figure 18: Degree that children trust strangers they meet 36
online by age
Figure 19: Frequency that parents talk with their children 38
about online credibility
Figure 20: Credibility of online information by informa- 40
tion type
Figure 21: Credibility of blogs compared to newspapers 42
and television
Figure 22: Credibility of Wikipedia 44
Appendix A 117
Figure 23: Most credible source by information type 46

Figure 24: Degree to which others should find information 48
credible by medium
Figure 25: Use of credibility evaluation methods by age 49
Figure 26: Child/parent differences in discerning good 75
versus bad information online
Figure 27: Child/parent differences in discerning false 76
information online
Figure 28: Child/parent differences in the likelihood of 77
questioning information online
Figure 29: Child/parent differences in discerning good 79
versus bad information online, compared to a typical
Internet user
believing false information online, compared to a typical
Internet user
questioning information online, compared to a typical
Internet user
Figure 32: Example encyclopedia entry 86
Figure 33: Credibility of one-sided versus balanced ency- 87
clopedia entries
Figure 34: Credibility of encyclopedia entry by perceived 87
source
Figure 35: Example encyclopedia entry 89
118 Appendix A
Figure 36: Credibility of encyclopedia entry by placement 90

of entry
Figure 37: Credibility of encyclopedia entry by actual 91
encyclopedia source
Figure 38: Example product Web page 94
Figure 39: Product quality by number of ratings, across 95
average product ratings
Figure 40: Product quality by average product ratings, 96
across number of ratings
Figure 41: Importance of credibility cues for purchasing 97
decisions
Figure 42: Male pregnancy hoax Web site 98
Figure 43: Rennets hoax Web site 99
Figure 44: Credibility of male pregnancy hoax Web site 101
Figure 45: Credibility of rennets hoax Web site 102
Appendix B: Knowledge Networks Methodology and
Panel Recruitment
Knowledge Networks has recruited the first online research

panel that is representative of the entire U.S. population. Panel
members are randomly recruited by probability-based sampling,
and households are provided with access to the Internet and
hardware if needed.
Knowledge Networks selects households using random digit
dialing (RDD) and address-based sampling methods. Once a
person is recruited to the panel, they can be contacted by email
(instead of by phone or mail). This permits surveys to be fielded
very quickly and economically. In addition, this approach
reduces the burden placed on respondents, since email notifica-
tion is less obtrusive than telephone calls, and most respondents
find answering Web questionnaires to be more interesting and
engaging than being questioned by a telephone interviewer.
Beginning recruitment in 1999, Knowledge Networks estab-
lished the first online research panel (now called Knowledge-
Panel®) based on probability sampling that covers both the
online and offline populations in the United States. The panel
members are randomly recruited by telephone and by
120 Appendix B
self-administered mail and Web surveys. Households are pro-

vided with access to the Internet and hardware if needed. Unlike
other Internet research that covers only individuals with Inter-
net access who volunteer for research, Knowledge Networks
surveys are based on a dual sampling frame that includes both
listed and unlisted phone numbers, telephone and non-tele-
phone households, and cell-phone-only households. The panel
is not limited to current Web users or computer owners. All
potential panelists are randomly selected to join the Knowl-
edgePanel; unselected volunteers are not able to join.
Random-Digit-Dialing Sample Frame
Knowledge Networks initially selects households using random-

digit-dialing (RDD) sampling and address-based sampling (ABS)
methodology. In this section, we will describe the RDD-based
methodology, while the ABS methodology is described in a sep-
arate section below.
KnowledgePanel recruitment methodology uses the quality
standards established by selected RDD surveys conducted for
the federal government (such as the Centers for Disease Con-
trol-sponsored National Immunization Survey).
Knowledge Networks utilizes list-assisted RDD sampling tech-
niques based on a sample frame of the U. S. residential landline
telephone universe. For efficiency purposes, Knowledge Net-
works excludes only those banks of telephone numbers (a bank
consists of 100 numbers) that have fewer than two directory-
listings. Additionally, an oversample is conducted among a stra-
tum of telephone exchanges that have high concentrations of
Appendix B 121
African American and Hispanic households based on census

data. Note that recruitment sampling is done without replace-
ment, thus numbers already fielded do not get fielded again.
A telephone number for which a valid postal address can be
matched occurs in about 70 percent of the sample. These
address-matched cases are all mailed an advance letter inform-
ing them that they have been selected to participate in Knowl-
edgePanel. For efficiency purposes, the unmatched numbers are
under-sampled at a current rate of 0.75 relative to the matched
numbers. Both the over-sampling mentioned above and this
under-sampling of non-address households are adjusted appro-
priately in the panel’s weighting procedures.
Following the mailings, the telephone recruitment begins for
all sampled phone numbers using trained interviewer/recruiters.
Cases sent to telephone interviewers are dialed for up to 90
days, with at least 14 dial attempts on cases where no one
answers the phone, and on numbers known to be associated
with households. Extensive refusal conversion is also performed.
The recruitment interview, about 10 minutes long, begins with
informing the household member that they have been selected
to join KnowledgePanel. If the household does not have a com-
puter and access to the Internet, they are told that in return for
completing a short survey weekly, they will be provided with a
laptop computer (previously a WebTV device was provided) and
free monthly Internet access. All members in a household are
then enumerated, and some initial demographic and back-
ground information on prior computer and Internet use are
collected.
122 Appendix B
Households that inform interviewers that they have a home

computer and Internet access are asked to take their surveys
using their own equipment and Internet connection. Per survey
incentive points, redeemable for cash, are given to these “PC”
respondents for completing their surveys. Panel members who
were provided with either a WebTV or a laptop computer (both
with free Internet access) do not participate in this per survey
points incentive program. However all panel members do
receive special incentive points for select surveys to improve
response rates and for all longer surveys as a modest compensa-
tion of burden.
For those panel members receiving a laptop computer (as
with the former WebTV), prior to shipment, each unit is custom
configured with individual email accounts, so that it is ready for
immediate use by the household. Most households are able to
install the hardware without additional assistance, though
Knowledge Networks maintains a telephone technical support
line. The Knowledge Networks Call Center also contacts house-
hold members who do not respond to email and attempts to
restore contact and cooperation. PC panel members provide
their own email addresses, and weekly surveys are sent to that
email account.
All new panel members receive an initial survey to both wel-
come them as new panel members and familiarize them with
how online survey questionnaires work. They also complete a
separate profile survey that collects essential demographic infor-
mation such as gender, age, race, income, and education to
create a personal member profile. This information can be used
to determine eligibility for specific studies, is used for weighting
Appendix B 123
purposes, and operationally need not be gathered with each and

every survey. (This information is updated annually with each
panel member.) Once new members are “profiled,” they are des-
ignated as “active” and ready to be sampled for client studies.
(Note: Parental or legal guardian consent is also collected for
conducting surveys with teenage panel members, age 13–17.)
Once a household is contacted by phone—and additional
household members recruited via their email address—panel
members are sent surveys linked through a personalized email
invitation (instead of by phone or mail). This permits surveys to
be fielded quickly and economically, and also facilitates longitu-
dinal research. In addition, this approach reduces the burden
placed on respondents, since email notification is less obtrusive
than telephone calls, and allows research subjects to participate
in research when it is convenient for them.
Address-Based Sampling (ABS) Methodology
When Knowledge Networks started KnowledgePanel® panel

recruitment in 1999, the state of the art in the industry was that
probability-based sampling could be cost effectively carried out
using a national random-digit-dial (RDD) sample frame. RDD at
the time allowed access to 96 percent of the U.S. population.
This is no longer the case. They introduced the ABS sample
frame to rise to the well-chronicled changes in society and tele-
phony in recent years that have reduced the long-term scien-
tific viability of the RDD sampling methodology: declining
respondent cooperation to telephone surveys; do not call lists;
call screening, caller-ID devices, and answering machines,
124 Appendix B
dilution of the RDD sample frames as measured by the working

telephone number rate; and finally, the emergence of house-
holds that no longer can be sampled by RDD—the cell-phone-
only households (CPOHH).
According to the Centers for Disease Control, approximately
21 percent of U.S. households cannot be contacted through
RDD sampling: 18 percent as a result of CPOHH status and 3
percent because they have no phone service whatsoever. Among
some segments of society, the sample non-coverage is substan-
tial: almost one-third of young adults age 18–24 reside in
CPOHHs. After conducting an extensive pilot project in 2008,
Knowledge Networks made the decision to add an address-based
sample (ABS) frame in response to the growing number of cell-
phone-only households that are outside of the RDD frame.
Before conducting the ABS pilot, they also experimented with
supplementing their RDD samples with cell-phone samples.
However, this approach was not cost effective and raised a
number of other operational, data quality, and liability issues
(e.g., calling people’s cell phones while they were driving, for
example).
The key advantage of the ABS sample frame is that it allows
sampling of almost all U.S. households—an estimated 99 per-
cent of U.S. households are “covered” in sampling nomencla-
ture. Regardless of households’ telephone status, they can be
reached and contacted. Second, the ABS pilot project revealed
some other advantages beyond the expected improvement in
recruiting adults from CPOHHs as well:
Improved sample representativeness for minority racial and
ethnic groups.
Appendix B 125
Improved inclusion of lower educated and low income

households.
Exclusive inclusion of CPOHHs that have neither a landline
telephone nor Internet access (approximately 4 percent to 6
percent of U.S. households).
ABS involves probability-based sampling of addresses from

the U.S. Postal Service’s Delivery Sequence File. Randomly sam-
pled addresses are invited to join KnowledgePanel through a
series of mailings and in some cases telephone follow-up calls to
non-responders when a telephone number can be matched to
the sampled address. Invited households can join the panel by
one of several means:
Completing and mailing back a paper form in a postage-paid
envelope.
Calling a toll-free hotline maintained by Knowledge Net-
works.
Going to a designated Knowledge Networks Web site and
completing the recruitment form.
As mentioned earlier, after initially accepting the invitation

to join the panel, respondents are then profiled online by
answering demographic questions and maintained on the panel
using the same procedures established for the RDD-recruited
research subjects. Respondents not having an Internet connec-
tion are provided a laptop computer and free Internet service.
Respondents sampled from the RDD and ABS frames are pro-
vided the same privacy terms and confidentiality protections
that Knowledge Networks has developed over the years and
have been reviewed by dozens of institutional review boards.
126 Appendix B
Because Knowledge Networks has recruited panelists from

two different sample frames—RDD and ABS—they take several
technical steps to merge samples sourced from these frames.
This approach preserves the representative structure of the over-
all panel for the selection of individual client study samples. An
advantage of mixing ABS frame panel members in any Knowl-
edgePanel sample is a reduction in the variance of the weights.
An ABS-sourced sample tends to align more true to the overall
population demographic distributions, and thus the associated
adjustment weights are somewhat more uniform and less varied.
This variance reduction efficaciously attenuates the sample’s
design effect and confirms a real advantage for study samples
drawn from KnowledgePanel with its dual frame construction.

Notes
Andrew Flanagin is a Professor in the Department of Communication at

the University of California, Santa Barbara, where he also serves as the
Director of the Center for Information Technology and Society. Dr. Fla-
nagin’s research focuses on the ways in which communication and
information technologies structure and extend human interaction,
with particular emphases on credibility, collective organizing, social
media, and collaborative groups.
Miriam Metzger is an Associate Professor in the Department of Com-
munication at the University of California, Santa Barbara. Dr. Metzger’s
research focuses on how information consumers assess the credibility of
information in the new media environment and on issues of online
information privacy.
Drs. Flanagin and Metzger are recognized as leading experts on cred-
ibility and digital media. They were among the first to conduct and
publish empirical research on this topic, and have written several arti-
cles and reviews on Web credibility over the last decade, including the
most comprehensive treatment of credibility in the online context to
date. They have served as expert advisors on credibility for several orga-
nizations, including the MacArthur Foundation, the American Library
Association, the Center for Media Literacy, and the National Library of
Medicine. In addition, Drs. Metzger and Flanagin recently received a
grant from the General Program of the MacArthur Foundation for their
128 Notes
study “Credibility and Digital Media: Helping People Navigate Informa-

tion in the Digital World” and coedited the volume Digital Media,
Youth, and Credibility (MIT Press, 2008) as part of the MacArthur Foun-
dation Series on Digital Media and Learning.
Ethan Hartsell, Alex Markov, Ryan Medders, Rebekah Pure, and
Elisia Choi are graduate students in the Department of Communica-
tion, at the University of California, Santa Barbara.
1. Benchmark distributions for Internet access among the U.S. popula-
tion of adults are obtained from KnowledgePanel recruitment data since
this measurement is not collected as part of the Current Population
Survey.
2. Since Knowledge Networks does not collect profile data for 11- and
12-year-olds, to set up the benchmarks of those with Internet access,
they first weighted all 13- to 18-year-olds to look like the 11- to 18-year-
old general population using Current Population Survey benchmarks.
Thirteen-year-olds were treated as if they were 11 and 12 years old; thus
13-year-olds were weighted to be 36.17 percent of this population
instead of 15.53 percent within all profiled members ages 13 to 18.
Then, based on the weights for all 13- to 18-year-old KnowledgePanel
members, Knowledge Networks derived the benchmarks based on those
who have Internet access from home and weighted the child respon-
dents to these Internet benchmarks.
3. The scales were constructed by relying on the results of principal

components factor analyses, and were informed by factor loadings and
the face validity of the questions we asked on the survey.
4. Multiple regression analysis was used to produce all results presented
in this section. Detailed statistical information is available from the
authors.
5. It is interesting that younger children said they were more likely to
believe information they find online than did older children in light of
our earlier finding that older children said they believe more of the
information on the Internet than do younger children (see the section,
Notes 129
“Perceived Trust and Credibility of Web-based Information”). This

could be due to younger children’s realization that they are particularly
susceptible to believing misinformation online, and older children’s
greater accumulation of positive experiences online, in terms of finding
information that is useful and credible—a few of the factors that played
into kids’ tendency to trust (or not trust) people they encountered
online.
6. The Web sites were modified slightly from their original online ver-
sions for size.

References
Alesina, A., and E. La Ferrara. 2002. “Who Trusts Others?” Journal of

Public Economics 85: 207–234.
Alexa. 2009. “Top Sites: The Top 500 Sites on the Web.” http://www
.alexa.com/topsites (accessed on August 25, 2009).
Alexander, J. E., and M. A. Tate. 1999. Web Wisdom: How to Evaluate and
Create Information Quality on the Web. Hillsdale, NJ: Erlbaum.
Benkler, Y. 2006. The Wealth of Networks: How Social Production Trans-

forms Markets and Freedom. New Haven: Yale University Press.
Campbell, J., N. Greenauer, K. Macaluso, and C. End. 2007. “Unrealistic

Optimism in Internet Events.” Computers in Human Behavior 23:
1273–1284.
Chesney, T. 2006. “An Empirical Examination of Wikipedia’s Credibil-

ity.” First Monday 11, no. 11. http://outreach.lib.uic.edu/www/issues/
issue11_11/chesney/index.html (accessed on October 30, 2009).
Clarke, V. A., H. Lovegrove, A. Williams, and M. Macpherson. 2000.

“Unrealistic Optimism and the Health Belief Model.” Journal of Behav-
ioral Medicine 23, no. 4: 367–376.
Eastin, M. 2008. “Toward a Cognitive Developmental Approach to

Youth Perceptions of Credibility.” In Digital Media, Youth, and Credibil-
132 References
ity, ed. M. J. Metzger and A. J. Flanagin, 29–47. Cambridge, MA: MIT

Press.
Epstein, S., R. Pacini, V. Denes-Raj, and H. Heier. 1996. “Individual Dif-

ferences in Intuitive-Experiential and Analytical-Rational Thinking
Styles.” Journal of Personality and Social Psychology 71, no. 3: 390–405.
Flanagin, A. J., C. Flanagin, and J. Flanagin. 2010. “Technical Code and

the Social Construction of the Internet.” New Media & Society 12, no. 2:
179–196.
Flanagin, A. J., and M. J. Metzger. 2000. “Perceptions of Internet Infor-

mation Credibility.” Journalism & Mass Communication Quarterly 77:
515–540.
Flanagin, A. J., and M. J. Metzger. 2007. “The Role of Site Features, User
Attributes, and Information Verification Behaviors on the Perceived
Credibility of Web-based Information.” New Media & Society 9, no. 2:
319–342.
Flanagin, A. J., and M. J. Metzger. 2008. “Digital Media and Youth:

Unparalleled Opportunity and Unprecedented Responsibility.” In Digi-
tal Media, Youth, and Credibility, ed. M. J. Metzger and A. J. Flanagin, 5–
27. Cambridge, MA: MIT Press.
Giles, J. 2005. “Internet Encyclopaedias Go Head to Head.” Nature 428

(December 15): 900–901.
Jacobs, J. E., and P. A. Klaczynski, eds. 2005. The Development of Judg-

ment and Decision Making in Children and Adolescents. Mahwah, NJ:
Erlbaum.
Jenkins, H. 2006. Convergence Culture: Where Old and New Media Collide.
New York: New York University Press.
Kamins, M. A., M. J. Brand, S. A. Hoeke, and J. C. Moe. 1989. “Two-

Sided versus One-Sided Celebrity Endorsements: The Impact on Adver-
tising Effectiveness and Credibility.” Journal of Advertising 18, no. 2:
4–10.
References 133
Klaczynski, P. A. 2001. “The Influence of Analytic and Heuristic Process-

ing on Adolescent Reasoning and Decision Making.” Child Development
72: 844–861.
Kokis, J., R. Macpherson, M. Toplak, R. F. West, and K. E. Stanovich.

2002. “Heuristic and Analytic Processing: Age Trends and Associations
with Cognitive Ability and Cognitive Styles.” Journal of Experimental
Child Psychology 83: 26–52.
Krane, B. 2006 (November 13). “Researchers Find Kids Need Better

Online Academic Skills.” The Advance. http://advance.uconn.edu/
2006/061113/06111308.htm (accessed on August 6, 2009).
Lenhart, A., and M. Madden. 2007 (April). Teens, Privacy, and Online
Social Networks. Pew Internet & American Life report. http://www
.pewinternet.org/Reports/2007/Teens-Privacy-and-Online-Social-Net-
works. aspx?r=1 (accessed on August 25, 2009).
Lenhart, A., Madden, M., & Hitlin, P. 2005 (July). Teens and Technology:
Youth Are Leading the Transition to a Fully Wired and Mobile Nation. Pew
Internet & American Life report. http://www.pewinternet.org/report
_display.asp?r=162 (accessed on April 17, 2006).
Leu, D. J., J. Coiro, J. Castek, D. Hartman, L. A. Henry, and D. Reinking.

2008. “Research on Instruction and Assessment in the New Literacies of
Online Reading Comprehension.” In Comprehension Instruction:
Research-Based Best Practices, ed. C. Collins Block, S. Parris, and P.
Afflerbach, 321–346. New York: Guilford.
Metzger, M. J. 2007. “Making Sense of Credibility on the Web: Models

for Evaluating Online Information and Recommendations for Future
Research.” Journal of the American Society for Information Science and Tech-
nology 58, no. 13: 2078–2091.
Metzger, M. J., and A. J. Flanagin, eds. 2008. Digital Media, Youth, and
Credibility. Cambridge, MA: MIT Press.
Metzger, M. J., A. J. Flanagin, K. Eyal, D. R. Lemus, and R. McCann.

2003. “Credibility in the 21st Century: Integrating Perspectives on
134 References
Source, Message, and Media Credibility in the Contemporary Media

Environment.” In Communication Yearbook 27, ed. P. Kalbfleisch, 293–
335. Mahwah, NJ: Erlbaum.
Metzger, M. J., A. J. Flanagin, and R. B. Medders. Forthcoming. “Social

and Heuristic Approaches to Credibility Evaluation Online.” Journal of
Communication.
Metzger, M. J., A. J. Flanagin, and L. Zwarun. 2003. “College Student

Web Use, Perceptions of Information Credibility, and Verification
Behavior.” Computers & Education 41, no. 3: 271–290.
Quantcast. 2009. “Quantcast US Site Rankings.” http://www.quantcast

.com/top-sites-1 (accessed on August 26, 2009).
Rainie, L. 2006. “Life Online: Teens and Technology and the World to
Come.” Keynote address, annual conference of the Public Library Asso-
ciation, Boston, MA, March 23. http://www.pewinternet.org/ppt/
Teens%20and%20technology.pdf (accessed on November 7, 2006).
Rieh, S. Y., and B. Hilligoss. 2007. “College Students’ Credibility Judg-

ments in the Information Seeking Process.” In Digital Media, Youth, and
Credibility, ed. M. J. Metzger and A. J. Flanagin, 49–71. Cambridge, MA:
MIT Press.
Scott, S. G., and R. A. Bruce. 1995. “Decision-Making Style: The Devel-

opment and Assessment of a New Measure.” Educational and Psychologi-
cal Measurement 55, no. 5: 818–831.
Slater, M. D., and D. Rouner. 1997. “How Message Evaluation and

Source Attributes May Influence Credibility Assessment and Belief
Change.” Journalism & Mass Communication Quarterly 73: 974–991.
Tapscott, D. 1997. Growing Up Digital: The Rise of the Net Generation. New
York: McGraw-Hill.
van Dijk, J. 2006. “Digital Divide Research, Achievements and Short-

comings.” Poetics 34: 221–235.
References 135
Weinstein, N. D. 1980. “Unrealistic Optimism about Future Life

Events.” Journal of Personality and Social Psychology 39, no. 5: 806–820.
Weinstein, N. D. 1982. “Unrealistic Optimism about Susceptibility to

Health Problems.” Journal of Behavioral Medicine 5, no. 4: 441–460.
Weinstein, N. D. 1987. “Unrealistic Optimism about Susceptibility to

Health Problems: Conclusions from a Community-Wide Sample.” Jour-
nal of Behavioral Medicine 10, no. 5: 481–500.

Kids and Credibility: An Empirical Examination of Youth, Digital Media Use, and Information Credibility

Uploaded by

Copyright:

Available Formats

Kids and Credibility: An Empirical Examination of Youth, Digital Media Use, and Information Credibility

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Kids and Credibility: An Empirical Examination of Youth, Digital Media Use, and Information Credibility

Uploaded by

Copyright:

Available Formats

Kids and Credibility

Peer Participation and Software: What Mozilla Has to Teach Government by

The Future of Learning Institutions in a Digital Age by Cathy N. Davidson

The Future of Thinking: Learning Institutions in a Digital Age by Cathy N.

Kids and Credibility: An Empirical Examination of Youth, Digital Media Use,

Confronting the Challenges of Participatory Culture: Media Education for the

An Empirical Examination of Youth, Digital Media Use, and

Andrew J. Flanagin and Miriam Metzger

The MIT Press

For information about special quantity discounts, please email special_

Library of Congress Cataloging-in-Publication Data

Series Foreword vii

Rationale and Overview 1

Conclusions and Implications 105

Appendix A: List of Tables and Figures 115

The John D. and Catherine T. MacArthur Foundation Reports

The enormous amount and variety of information currently

Despite these complex realities, examinations of youth and

Overall, children rely fairly heavily on the Internet. The most

Regarding children’s perceptions of information credibility:

credible increases with age somewhat, their concern about cred-

tion online when deciding which credibility assessment tools to

Regarding child/parent dyads and credibility assessments:

Parents believe they are more adept at assessing credibility

Regarding children’s Web site exposure and evaluation:

placed on the other sites. In addition, entries actually originat-

Overall, this project provides a comprehensive investigation

We are indebted to a great many people who contributed in

Bentley, and Monica Koegler-Blaha have provided invaluable

With the sudden explosion of digital media content and infor-

accurately assess the credibility of information can have serious

accepting a source’s self-asserted credibility, since such assess-

research. As the first generation to grow up with the Internet,

many types of digital media. In the face of increasing disinter-

Although there is a burgeoning literature and empirical work

processing, research experts currently working in the fields of

forwarded to the research firm that administered the survey to

same procedures established for research subjects recruited by

email notification contained a link that sent them to the survey

Incident and Completion Rates

non-response and other sources of survey error that might bias

This section provides a detailed profile of the demographic

Parent and Household Demographics

parents were white; 9 percent were black, non-Hispanic; 9 per-

Region Percent Residing

New England 4.9

Age Number Percent

Internet Usage among Youth

To better understand how young people may be affected by the

Children’s Use of the Internet

Children in the study reported that they use the Internet

Parental Control of Children’s Access to and Use of the Internet

However, parental oversight of children’s online activities

Any type of restriction

reported sitting with their child “sometimes,” whereas 16 per-

Children’s Online Activities

Web uses reported by respondents. To assess this, we con-

about credibility issues online, tended to get good grades, and

children’s use of the Web for “virtual” purposes varies little by

level of doing so “rarely.” Nonetheless, sharing information in