This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
         If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line, 
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").

(Jan 2006)
Thank you.
Elliott Parker
====================================================================

Internet Use, Digital Literacy and Purposive Web Use: An Emerging Model
internet is used.
Abstract
Internet studies scholars are calling for more focus in internet use 
research on the
nature of use by those who have access. User sophistication and 
digital literacy are two
concepts currently being defined and explored, especially in relation 
to research on new
types of digital inequality, or "digital divides" associated with 
differentials in how the
In order to further explore the concept of "digital literacy," this 
study presents
findings from an international survey of high-access internet users 
consisting of mass
communications educators, researchers, graduate students and 
undergraduate students,
exploring the relationships between internet use, digital literacy, 
and purposive web use.
In this study, purposive web use takes the form of online political 
participation, online
opinion leadership, online alternative view-seeking, and online political news
consumption. Findings confirm past work in general populations 
indicating that even in
this highly media literate population, age is a barrier to internet 
use and digital literacy.
Heirarchical regression analysis suggests that age, frequency of 
internet use, and other
more detailed internet use variables serve as indicators for digital literacy.
Additionally, heirarchical regression analysis shows that when 
controlling for age
and internet use, digital literacy still serves as a predictor for 
online political participation
and other forms of purposive web use, suggesting that digital 
literacy may be a
prerequisite to such activity. And finally, when controlling for age, 
internet use and
digital literacy, online political participation serves as an 
indicator of offline participation
among this population of high-access users.
2
Internet Use, Digital Literacy and Purposive Web Use
Introduction
Digital literacy denotes one's depth and breadth of understanding of digital
technologies, and has roots in the umbrella concept of media 
literacy. Sonia Livingstone
(2004) suggests that media literacy, along with media and 
communications in general,
contribute in our media-saturated knowledge society to three societal 
purposes: a)
democracy, participation and active citizenship; b) knowledge 
economy, competitiveness
and choice; and c) lifelong learning, cultural expression and 
personal fulfillment. Digital
literacy may be relevant to this model as well, given the following 
societal purposes that
may be fulfilled through internet use: a) opportunities online to 
become politically
engaged, b) the upward mobility and competitive edge that technology 
skills can afford
individuals, and c) the learning and cultural gains that can be 
achieved through both
consumption and production of online content. Digital literacy is 
therefore an emerging
concept of importance spanning multiple disciplines.
Digital divide researchers are also interested in the concept of 
digital literacy.
National phone and internet surveys have given us descriptive data on 
how segments of
the population use the internet generally, and how digital 
inequality, or the "digital
divide," is evidenced in frequency and patterns of internet use 
across demographics.
While these findings alone hold major policy implications, Hargittai 
(2004) urges
researchers to continue moving the internet use agenda forward to 
include a greater
understanding of multiple dimensions of digital inequality, 
specifically technological
access, autonomy, social support, skill, and types of uses. She also 
suggests that we must
explicate the meaning of "internet use" given the diversity of 
functions the internet serves
3
Internet Use, Digital Literacy and Purposive Web Use
(2004). Hargittai has thus begun to address the variable of user 
sophistication (2003) and
digital literacy (2005). Similarly, DiMaggio (2002) has pointed out 
that while digital
divide research focuses on differences among the haves and have nots 
across the
population, a deeper understanding is needed on how the "haves" use 
digital technologies
to which they hold varying levels of access, and the interactions 
among uses. And
Livingstone (2005b) too points out that among those who already have 
access, the two
key considerations for audience use may reside within the concepts of
motivation/interest, and technology skills or depth of knowledge (literacy).
This study addresses digital literacy differences in a population of mass
communications educators, graduate students and undergraduate 
students, who are
assumed to all hold at least in-school access to the internet and 
thus fall into the category
of "users with internet access." Additionally, because of their 
position as educators and
students of the media industries, their level of media use will 
likely be more advanced
than the general population. Within this population of 
university-affiliated, media-savvy
scholars, this study will address a number of key digital literacy 
variables in relation to
purposive internet use, an endeavor identified by Livingstone (2005b) 
as an important
area for future research. Examples of purposive internet use in this 
study include online
political participation, online alternative view-seeking, online 
opinion leadership, and
online news consumption. The study addresses the extent to which 
digital literacy may be
a prerequisite for these activities.
Literature review
Media literacy. The term "literacy" is difficult to define, because as Bawden
(2001) suggests there are many meanings for the word, and the term 
has a history of
4
Internet Use, Digital Literacy and Purposive Web Use
changing meanings over time. Friere and Macedo (1987) state, 
"literacy is not the reading
and writing of words in and of themselves, as if the reading and 
writing of words did not
imply another reading, anterior to and simultaneous with the first -- 
the reading of reality
itself." Literacy is not just made up of task-oriented skills, but is 
also tied up with
meaning creation. Media literacy addresses the role of the media in 
reflecting and
constructing meaning, real or otherwise, for the public, and is 
concerned with audiences'
ability to interpret mediated messages.
Information literacy. Another legacy term for digital literacy is that of
"information literacy," which Livingstone (2004) suggests should 
include a number of
skills. The main distinction she points out between media and 
information literacy is that
the latter emphasizes the identification, location, evaluation and 
use of media materials
(2004). According to Isbell and Hammond (1993), "An information 
literate person
recognizes the different levels, types and formats of information and 
their appropriate
uses. The ability to place information in a context and an awareness 
of information access
issues (copyright, privacy, globalization, currency of information, 
etc.) are key to
information literacy." Information literacy has been a concern for 
those in the field of
library science and information studies, and as online database 
technologies have
advanced, the term for some has come to include if not entirely 
denote internet search
skills.
Integrating media and internet-oriented literacies. Livingstone, Van Couvering
and Thumim (2005b) do not adopt the term "digital literacy," instead 
choosing to
integrate digital and information technologies into an overall model 
for media literacy in
general for adults. This model includes multiple media literacy 
categories, which the
5
Internet Use, Digital Literacy and Purposive Web Use
authors describe medium by medium, including an internet/mobile 
technology section in
each media literacy category. For instance, in the first of eight 
categories, "media access
and ownership," the internet component includes the topic of the 
digital divide, for which
Livingstone et al. suggest, citing Murdock (2002), should be 
addressed as a continuum
with degrees of marginality and multiple different types of divide, 
given that the
"goalposts of what constitutes acceptable access are continually 
shifting" as new
technologies emerge and old ones evolve (2005b, p. 14). And the 
second category,
"navigating—basic media competences," refers for the internet to the 
ability to use its
essential features such as a browser, mouse, keyboard, etc.
Within these categories, Livingstone et al.'s (2005b) useful review 
of existing
literature points to an overall set of key barriers to digital and 
media literacy among
adults that have been found in empirical research in the U.S.,U.K, 
and other countries,
including age, socio-economic status, gender, disability, ethnicity, 
and proficiency in
English (2005b, p. 50). The authors propose that in contrast, 
enablers of digital literacy
include the interface design of technologies and content, adult 
education opportunities,
consumer information and awareness, perceived value of media goods 
and services, selfefficacy
(skills and confidence), social networks of support, family 
composition including
having children in the household, work involving computers and new 
technologies, and
institutional stakeholders (2005b, p. 50). The authors point to a 
number of key areas for
future research (pp. 59-62), including the following: addressing 
inequalities and excluded
population segments, tracking of more advanced forms of access such 
as using public
databases and advanced internet searching, a better understanding of 
how people manage
multiple media together, how people understand online news and 
political information,
6
Internet Use, Digital Literacy and Purposive Web Use
critical awareness of promotional practices, the range of experiences 
of content creation
by audiences, the social and other benefits of content creation, 
quality or sophistication of
media use, evaluation/assessments, effectiveness of interventions, 
medium-specificity of
media literacy, accessibility issues built into technology designs. 
This study addresses a
few facets of Livingstone's model: individuals' use and understanding 
of online news and
their engagement in online political activity, as these relate to 
their general level of taskoriented
digital literacy.
Digital literacy. Paul Gilster (1997) may have been the first to coin 
the phrase
digital literacy in publication, which is presented as a concept in a 
guidebook for internet
users and educators. Under most formal definitions, "digital 
literacy" includes the
interpretation of images and other digital forms besides text. Bawden 
(2001) cites
Gilster's work as positing that digital literacy is
... cognition of what you see on the computer screen when you use a networked
medium. It places demands upon you that were always present, though less
visible, in the analog media of newspaper and TV. At the same time, 
it conjures
up a new set of challenges that require you to approach networked computers
without preconceptions. Not only must you acquire the skill of 
finding things, you
must also acquire the ability to use these things in your life.
Gilster's definition emphasizes use of technology, and his work also 
discusses the
role of the network in expanding information in many forms retrieved 
through computer
interfaces. This book made popular the idea that with new media arise 
the need for a new
set of skills for interpreting meaning of digital messages, and for 
navigating networked
technologies.
7
Internet Use, Digital Literacy and Purposive Web Use
Eshet (2004, pp. 94-102) also emphasizes both interpretation and use 
of digital
media in her helpful conceptual model for digital literacy, and she 
and Amichai-
Hamburger (2005) present empirical experimental data to support the 
model. The study
included the age categories, young (11th graders averaging 16.9 years 
of age), middle
(third-year college students averaging 26.4 years of age), and older 
(college graduates
between thirty and forty, averaging 36.5 years of age). The 
experiment presents
participants with a scenario for each dimension that requires them to 
complete a task,
often resulting in an artifact being created by the participant. She 
and a team of evaluators
then assessed their work. In Eshet, et al.'s model, photo-visual 
literacy indicates one's
ability to "read" and interpret visual representations. They find 
that when participants
with little direction used a theatrical stage design software program 
that involved MS
Office-type windows navigation, there were no significant differences 
between young
and middle groups' assessment scores, but both were significantly 
different from the
oldest group's scores which were lower (2005).
Eshet (2004) defines reproduction literacy as the ability to create 
new meanings
by recombining existing information in any form of media (text, 
graphic, or sound). Her
stimulus for this dimension tested text reproduction literacy and 
involved editing a
neutral text story about a child getting ready for school (2005). 
Results indicated that the
oldest group scored significantly highest, followed by the middle and 
the youngest,
contrary to other research suggesting that age is a barrier to 
digital literacy, but likely due
to low ecological validity given that the test is more about creative 
writing than digital
skills (2005).
8
Internet Use, Digital Literacy and Purposive Web Use
The branching literacy of hypermedia and non-linear thinking dimension is
characterized by those who have "a good sense of multidimensional 
spatial orientation"
(Eshet, 2004, p. 97). Her results indicate that there were no 
significant differences
between the two younger groups, who both scored significantly higher 
than the oldest in
planning an online trip to Spain (2005).
Information literacy, or the art of skepticism to identify bias and 
persuasion in
content, was operationalized using seven internet news sources of 
multiple political
persuasions, which participants were asked to analyze, looking for 
bias and contradiction,
and presenting an overall written summary of opinion on quality and 
reliability for each
source (2005). Comparisons by age indicate that the oldest group 
scored highest,
followed by the middle and then the low (2005).
The final dimension in Eshet's model is socio-emotional literacy in online
environments, characterized by the ability to sociologically 
recognize for instance a virus
email, and ability to behave appropriately and communicate 
effectively in online
collaboration (2004). For this dimension, participants entered a chat 
room and engaged
in dialogue on a political topic. Her experimental findings indicate 
that the middle group
scored highest, followed by the youngest, followed by the oldest on a 
combined variable
that included cognitive presence, social presence, and emotional 
presence (2005).
Overall, within her model and stimuli operationalizations, Eshet et 
al.'s findings
indicate that younger scholars are not always better than older 
scholars in executing
digital literacy tasks, as they have defined them. These empirical 
results for the concept
of "digital literacy" are interesting, and warrant further 
exploration in order to continue
9
Internet Use, Digital Literacy and Purposive Web Use
building out and refining the definition of the concept. They will be 
addressed in relation
to this study's findings in the discussion section.
Survey measures of digital literacy. Outside of education, others 
interested in the
concept of digital literacy include scholars in sociology addressing 
digital divide issues.
Hargittai (2002) suggests that it is becoming increasingly important 
to distinguish
varying levels of online skill, i.e., the ability to efficiently and 
effectively find
information on the Web. This definition falls within the "skill 
(effectiveness of use)"
category of DiMaggio and Hargittai's (2001) five possible dimensions 
of digital divide,
which also includes the categories technical means (software, 
hardware, connectivity),
use patterns (types of use), autonomy of use (location of access, 
freedom), social support
networks (others to turn to for help and increased size of networks of online
communication). These scholars see digital literacy as a concept that 
can help researchers
achieve a better understanding of depth of internet use in the 
context of "digital
inequality" (DiMaggio, Hargittai, Celeste & Shafer, 2004).
Hargittai (2005) offers a set of survey items for the concept of 
digital literacy as
she conceives it, that can stand in as a proxy for peoples' actual 
skill. Her model is more
narrow than Eshet's, but she finds a number of useful items that she 
presents as a new
digital literacy index (2005).
This study uses Hargittai's (2005) survey variables to examine 
digital literacy as
a possible indicator for Livingstone's (2005b) suggested literacy 
topics of online political
participation and online news consumption. The results will be 
addressed in the context
of the age demographic findings of Eshet's (2004) experiment on 
digital literacy.
10
Internet Use, Digital Literacy and Purposive Web Use
Given the broad framing of digital literacy as a concept, and the findings of
Livingstone (2005b), Hargittai (2005) and Eshet (2005), the following 
hypotheses are
presented for a survey conducted with a population of high-access 
internet users from the
field of mass communications.
H1. The more respondents use the internet, the more digitally 
literate they will
be.
H2. The older the respondents, a) the less time they will spend 
online, and b) the
lower their digital literacy will be.
H3. The older the respondents, the fewer internet tasks they will know how to
perform.
These hypotheses are supported by findings that indicate that younger 
users take
part in a wider diversity of internet forms (Livingstone, et al. 
2005a). Livingstone et al.
(2005a) found that age has a double effect on active participation 
through its negative
relationship with longer average times on internet use and higher 
internet self-efficacy.
Additionally, this hypothesis is supported by her findings that 
across multiple activities,
young people who participate in one activity in a category are more 
likely to participate
in another, and by the finding that the more social support, the 
greater the internet use
(Bonfadelli, 2002). Results of this hypothesis test use Hargittai's 
(2005) digital literacy
index, and put forward new findings as to the relationship between 
age and digital
literacy among users with high levels of access.
The digital literacy index includes familiarity with advanced 
searching, but this
study will address information-seeking as an additional variable 
distinct from "main
digital literacy" which will denote the index. Information-seeking 
behavior is of interest
11
Internet Use, Digital Literacy and Purposive Web Use
due to its possible conceptual relationship to digital literacy. 
However, given the
population, it is suggested that while older participants may be less 
digitally-literate than
younger users overall, they likely do use the internet, and they 
likely use it more for
information seeking than for the range of activities the index 
represents. Therefore, the
following hypothesis is presented.
H4. The older the respondents, the more information-seeking task familiarity
they will have in comparison to younger users.
Information-seeking behavior is defined in this study by two sub-concepts,
internet search frequency, and mass communication online research 
database familiarity.
H5: When controlling for age, internet use, information-seeking 
behavior, there
will still be a relationship between basic digital literacy and 
advanced digital literacy.
For this hypothesis, advanced digital literacy is a dependent 
variable representing
an index of tasks that are more difficult than those presented in 
Hargittai's (2005) digital
literacy index. Advanced digital literacy addressed in a regression 
model presenting the
fit of the stated independent variables taken together. Taken in 
construct, these variables
are hypothesized to add up within the overall picture of factors 
impacting an individual's
task-oriented digital literacy skills.
Livingstone et al. (2005b) suggest that online political 
participation and online
news consumption are two elements of purposive web use that can lead 
to media literacy
overall, and contribute to the societal purpose of democratic 
participation, and lifelong
learning. Hargittai's (2005) task-oriented digital literacy skill 
concept may be
prerequisites to Livingstone's purposive web use and media literacy 
goals. Therefore the
following hypotheses are presented.
12
Internet Use, Digital Literacy and Purposive Web Use
H6. When controlling for age and internet use, there will still be a 
relationship
between basic digital literacy and online political participation 
variables as well as online
political news consumption.
H6 suggests that internet use and task-oriented digital literacy are 
prerequisites of
online political participation because per Livingstone (2005b) online 
political
participation is an advanced digital activity that may require the 
skills that Hargittai's
(2005) digital literacy variable represents. Test of this hypothesis 
begins to build upon
some of Livingstone's recommendations. All results will be 
contextualized in relation to
Hargittai's (2005), Eshet's (2004) and Livingstone et al.'s (2005b) 
recent work.
H7: When controlling for age, internet use, and task-oriented digital 
literacy,
there will still be a relationship between online political 
participation and offline political
participation.
The impact of political internet use on one's "real world" political 
activity is a
growing topic of interest spanning many fields (for instance, Putnam, 2000 and
surrounding literature on declining social capital). However, in 
surveys of the general
population such as the GSS, it is difficult to find effects of online 
political activity on
individuals' offline political behavior due in part to changes in 
topic modules emphasized
from year-to-year. This hypothesis is presented in line with 
recommendations of Hargittai
(2002), DiMaggio (2001) and others regarding gaining a more detailed 
understanding of
heavy internet users' level of sophistication of use. Given the 
presumed advanced level of
internet access of the population being sampled, it is possible that 
digital literacy and
online political activity may present as indicators of offline 
political activity for this
advanced sample.
13
Internet Use, Digital Literacy and Purposive Web Use
Methods
The population for this online survey is international journalism and mass
communications educators, graduate students, and undergraduate 
students. The sampling
methods used included a random systematic sample taken from several 
association
member directories in the field of mass communications, a convenience 
sample taken
from web searches for students and faculty of international 
university journalism and
mass communication departments using Google, Yahoo, Metacrawler and 
Dogpile, and
snowball samples of students and faculty deriving from forwarded 
emails from the
original lists. Additionally, convenience samples were taken from the 
personal address
books of the sixteen researchers involved in the WIMU project which 
was conducted
within a graduate survey research methods class. The total N of 
association directory
email addresses collected for invitation was 3050. The total N of the 
international
university online email addresses collected for invitation was 1560. 
The number of
forwarded email invitations to others deriving from our invitation 
emails that resulted in a
snowball sample is unknown.
Survey procedure.
This survey was administered using the online service, SurveyMonkey.com. All
participants were emailed an invitation to participate containing a 
link to the survey. The
survey was available 24-hours a day for two weeks in March and April, 2005.
Participants were sent three waves of invitation emails, one every 
five days to increase
the response rate, with prior respondents de-duped. The overall 
response rate is less than
16%, which represents the total N of survey responses divided by the 
total N of known
14
Internet Use, Digital Literacy and Purposive Web Use
invitations sent. Many of these responses however could have resulted 
from the snowball
sample.
Demographic independent variables.
Age and education were the two demographic independent variables used in this
study, supporting past work which has indicated that the two are 
predictors for internet
use and digital literacy. Age was operationalized using the survey 
item, "How old were
you on your last birthday" and measured as a ratio variable in years. 
Education was
operationalized using the survey item, "How many years of schooling 
do you have?" and
measured as a ratio variable in years.
Internet use independent variables.
The concept, "internet use," is represented by the following two independent
variables: "time use," and "history with the internet." The time use 
independent variable
is operationalized using a multiplicative combination of the survey 
items, "On average,
how many days a week do you use the internet?" (measured as a 
ratio-level variable as 1-
7) and "About how much time a day do you spend on the Internet?" (measured in
minutes). History with the internet was operationalized using an 
additive index of the
three items (each measured as a ratio-level variable in years), "When 
did you first start
using e-mail?" and "When did you first start using the internet for 
purposes other than email?"
and "If you have created a web site, in what year did you create your 
first web
site?". Respondents who had not created a website were considered to 
have zero years of
experience with that added item.
Digital literacy independent variables.
15
Internet Use, Digital Literacy and Purposive Web Use
The concept, "digital literacy," is represented by five independent variables:
"main digital literacy," "advanced digital literacy," "self-perceived 
skill," "created
website," and "information-seeking."
The "main digital literacy" independent variable draws from the work 
of Hargittai
(2005). She compares survey questions on respondents' self-reported 
familiarity with a
range of specific online tasks, as they relate to respondents' actual 
tested skill at those
online tasks measured by skill test questions, in order to assess the 
validity of the selfreport
measures as proxies. Hargittai's (2005) study finds high correlations 
between skillbased
test questions and seven proxy variables that she offers up as an 
effective survey
index for measuring adult digital literacy. She operationalizes this 
proxy index by the
question, "How familiar are you with the following Internet-related 
items? Please choose
a number between 1 and 5 where 1 represents having 'no understanding' 
and 5 represents
having 'a full understanding' of the item. (none, little, some, good, 
full)," where she
provides a list of many computer-related terms. She found that the 
optimal index
resulting in highest representation of actual skill includes the 
following items: MP3,
Downloading, Preference setting, Refresh/Reload, Newsgroup, PDF, 
Advanced search.
The index she created out of her "seven best item" list for digital 
literacy achieved a
Cronbach's alpha of .89, and has a predictive power (adjusted R2) of 
.321 for actual
internet skill, the highest among all quantitative indices of digital 
literacy published.
This study will use Hargittai's (2005) digital literacy proxy variables as a
replacement for actual skill test questions. In this study's 
operationalization, the same
seven items were used, but the question wording and categories were 
changed to a more
traditional symmetrical Likert scale, stated as follows: "How 
familiar are you with each
16
Internet Use, Digital Literacy and Purposive Web Use
of the following…" (very familiar = 5, familiar = 4, neutral = 3, 
unfamiliar = 2, very
unfamiliar = 1, don't know = 99). Additionally, because of the 
assumed advanced level
of the population, this study has added the following variables, 
which represent more
advanced tasks, to increase the variance on the digital literacy 
measure: JPG, weblog or
"blog," instant messaging, and shareware. An additive index was 
created for all of the
digital literacy terms, called "advanced digital literacy."
The independent variable, "created website," was operationalized using the
following survey items: "Have you ever created a website?" (0 = no, 1 
= yes), and "If yes,
how many years ago did you create a website" (measured as a ratio 
variable in years, with
"never" represented as 0). These items were multiplied as a combined 
variable and are
presented as "created website" to represent a distinct facet of 
digital literacy.
Information-seeking independent variables.
We also add the following "information-seeking" items to the list 
provided in the
main digital literacy question, and analyze the items as a separate 
index representing this
concept. Information-seeking was operationalized based on 
respondents' familiarity with
the following terms, which represent online search databases widely 
used for research in
the field of mass communications: library database, LexisNexis, OCLC 
FirstSearch,
ProQuest, Communication Abstracts. Additionally, we included the question, "On
average, how many times per day you perform a search using a search 
engine," measured
as a ratio-level variable. Information-seeking is represented by an 
additive index of the
familiarity items, and by the search engine use variable separately.
Purposive web use dependent variables.
17
Internet Use, Digital Literacy and Purposive Web Use
In this study, the concept, "purposive web use" is comprised of the 
four dependent
variables: online political participation, online opinion leadership, 
online alternative
view-seeking behavior, and online political news consumption. The independent
variable, online political participation, was operationalized using 
an index of the
following five items, each measured on a Likert scale (strongly 
disagree = 1, disagree =
2, neutral = 3, agree = 4, strongly agree = 5): "On the Internet, I 
participate in
discussions about political topics," "In my online interactions, I 
provide other people with
information about current events," "I express my opinions even when 
others disagree
with me online," "I talk with people who support different political 
parties than me
online," and "Talking with people who have different political views 
from me can help
build mutual respect online."
The independent variable, online opinion leadership, was 
operationalized using an
index of the following four items, each measured on a Likert scale 
(strongly disagree = 1,
disagree = 2, neutral = 3, agree = 4, strongly agree = 5): "People 
ask my opinion about
current events online," "People consider me a good source of advice 
about current events
online," "I try to convince others to change their political views 
online" and "I express
my opinions even when others disagree with me online."
The independent variable, online alternative view-seeking behavior, was
operationalized using an index of the following five items, each 
measured on a Likert
scale (strongly disagree = 1, disagree = 2, neutral = 3, agree = 4, 
strongly agree = 5):
"News from a variety of viewpoints are important to me when accessing 
online news
sites," "On the Internet, I seek diverse news media with different 
political views," "On the
Internet, I try to find diverse political views different than my 
own," "Points of view that
18
Internet Use, Digital Literacy and Purposive Web Use
are different from my own are important to me when accessing online 
news sites," and "I
can understand the perspectives of those who disagree with me online."
Online political news consumption was operationalized using one item, "When
accessing online news sites, I look for news about politics," 
measured on a Likert scale
(strongly disagree = 1, disagree = 2, neutral = 3, agree = 4, 
strongly agree = 5).
Offline political participation dependent variables.
The dependent variable, offline political participation, was measured as an
additive index of the ratio-level responses to the following 
question, "How many times in
the last twelve months have you… a.) Voted in local elections, b) 
Volunteered to work
for a political party in an election campaign, c) Contributed money 
to a political party or
candidate, d) Attended a public meeting dealing with political or 
social issues, e) Spoken
in such a meeting, f) Personally gone to see, or spoken to, or 
written to government
officials beyond the local level, g) Signed a petition, h) Collected 
signatures, i) Given a
political speech, j) Attended a legal demonstration, k) Attended an 
illegal demonstration,
l) Written to an editor about political news coverage or an 
editorial, m) Called a phone-in
radio show regarding political, societal or public issues. The 
resulting value for each
respondent is a count of their total number of offline political 
actions taken in the last
twelve months, from among this set of thirteen possible actions.
Models.
In a first set of four-variable models, the independent variables of 
demographics,
internet use, and digital literacy are measured as composite variable 
blocks in three
separate hierarchical regression models examining their effects on 
the three dependent
variables representing purposive internet use (online political 
participation, online
19
Internet Use, Digital Literacy and Purposive Web Use
opinion leadership, and online alternative-view-seeking behavior). In 
a five-variable
model, the variable "online political participation" becomes an 
independent variable
following the original three (demographics, internet use, digital 
literacy), and the four
variables are analyzed for their effect on the fifth dependent 
variable, "offline political
participation."
Results
Descriptive statistics. See tables 1-6 for means and standard deviations of
demographic, internet use, digital literacy, online political 
participation, online opinion
leadership, online alternative view-seeking, online political news 
consumption, and
offline political participation variables. The mean age for 
respondents was high (39.89),
as was education level. Education was not considered as a demographic 
in analysis due to
the relatively low variation for this sample in comparison to the 
general population. The
mean for ability to complete specific web tasks, and the "Web task 
index" was very high
(see Table 3), except on the variable "know how to create a website" 
which appears quite
a bit lower (M=1.47, SD=.50, coded as no=1, yes=2). Compared to 
Hargittai (2005)'s
random systematic sample of the general population, the respondents 
in this study's
population of mass communication students and educators on average 
appear to have
higher digital literacy, although further analysis would be needed to 
confirm this.
Digital literacy variables. Table 7 presents the relationships 
between the basic
digital literacy index (Hargittai, 2005), advanced digital literacy 
index, ability to perform
web task index, information-seeking index, ability to create a 
website (distinct), and
number of searches performed per day (distinct). All variables are 
statistically significant
in their relationship to each other. Basic and advanced digital 
literacy appear to have the
20
Internet Use, Digital Literacy and Purposive Web Use
highest strength relationship among the bivariate data (Pearson 
r=.75, p<.01). The
strength of the relationship between basic digital literacy and 
ability to complete web
tasks also appears to be high (Pearson r=.59, p<.01).
Tests of hypotheses.
H1. This study confirms the hypothesis that the more respondents use 
the internet,
the more digitally literate they will be, for both basic and advanced 
digital literacy (see
Table 8).
H2. This study confirms the hypothesis that the older the 
respondents, a) the less
time they will spend online (Table 8), and b) the lower their digital 
literacy will be, for
both basic and advanced digital literacy (Table 7).
H3. This study confirms the hypothesis that the older the 
respondents, the fewer
internet tasks they will know how to perform, using the web task 
index variable (see
Table 6).
H4. The hypothesis that the older the respondents, the more 
information-seeking
task familiarity they will have is unconfirmed. However, a post-hoc 
ANOVA with F
(5,404)=12.55 and p<.001 demonstrates that educators had a 
significantly different,
higher mean for information-seeking (M=3.83, SD=.911, see Table 4 for 
scale) than
undergraduates (M=3.11, SD=1.05, see Table 4 for scale), as did Ph.D. 
students (M=4.06,
SD=.86, see Table 4 for scale) in comparison with undergraduates.
H5. The regression models for advanced digital literacy (Table 9) 
suggest that the
variables age, internet time use, history with the internet, 
information-seeking, website
creation, searches per day and basic digital literacy contribute in 
Model 4 to 59% of the
variance in advanced digital literacy (p<.001), supporting this hypothesis.
21
Internet Use, Digital Literacy and Purposive Web Use
H6. The regression models for online political participation (Table 
10) suggest
that the variables age, internet time use, history with the internet, 
and basic digital
literacy contribute in Model 3 to 8% of the variation in online 
political participation
(p<.01).
H7. The regression models for online opinion leadership (Table 11) 
suggest that
the variables age, internet time use, history with the internet, and 
basic digital literacy
contribute in Model 3 to 7% of the variation in online opinion 
leadership (p<.05).
H8. The regression models for online alternative view-seeking (Table 
12) suggest
that the variables age, internet time use, history with the internet, 
and basic digital
literacy contribute in Model 3 to 10% of the variation in online 
alternative view-seeking
(p<.001).
H9. The regression models for online political news consumption (Table 13)
suggest that the variables age, internet time use, history with the 
internet, and basic
digital literacy do not contribute in Model 3 to the variation in 
online political news
consumption. However, in Model 1 we see that age on its own 
contributes to about 2%
of the variance in this dependent variable (p<.01).
H10. The regression models for offline political participation (Table 
14) suggest
that the variables age, internet time use, history with the internet, 
basic digital literacy
and online political participation contribute in Model 3 to 8% of the 
variation in offline
political participation (p<.001). Post hoc analysis (Table 15) 
switching offline political
participation to an independent and online political participation to 
a dependent variable
suggests that this model contributes to 12% of the variation in 
online political
participation (p<.001).
22
Internet Use, Digital Literacy and Purposive Web Use
Discussion
The findings of this study support past work suggesting that age 
continues to be a
barrier to internet time use and digital literacy, even among a group of mass
communications educators and students whose partial focus is to have, 
and teach, media
literacy overall. The straightforward negative relationship for age 
indicated using
Hargittai's basic digital literacy measure lies in contrast with 
Eshet's (2005) experimental
findings for age and digital literacy as she conceives it, which 
suggest that older users'
comparative digital literacy levels vary based on task. Indeed, while 
age was not found to
hold a significant relationship with information-seeking, another 
type of digital literacy,
educators were found to hold higher levels of information-seeking skills than
undergraduates. It is possible that Ph.D. students highest mean score 
for informationseeking
mitigated the final age results for information-seeking.
The models for advanced digital literacy demonstrate that age, internet use,
information-seeking / website creation / number of searches performed 
per day and basic
digital literacy all build upon each other in succession as 
indicators of advanced digital
literacy. Surprisingly, education was not found to hold a 
relationship to digital literacy in
this sample. This may be due to the low variation for education in 
this narrow population,
and due to the fact that the population as a whole had a very high 
level of digital literacy
overall. The within the basic digital literacy index, MP3, preference 
setting, and
newsgroup held the highest standard deviations, and the advanced 
digital literacy items
JPG, blog, instant messaging and shareware all held standard 
deviations higher than 1.
More work should be done to explore how people come to learn these 
tasks, and what
these tasks mean to their purposes for using the internet. The 
digital literacy findings and
23
Internet Use, Digital Literacy and Purposive Web Use
basic and advanced operationalizations of variables may be helpful 
for future research on
more in-depth internet use given what we now know of their predictive 
strength.
Age was not found to be correlated with online political 
participation, and this
may be due to the fact that age has been found to be positively 
correlated with political
participation in general, while negatively correlated with internet 
use. Nonetheless, the
model for online political participation does appear to be impacted 
by internet use and
basic digital literacy. These results confirm the hypothesis that 
digital literacy is a
possible pre-requisite for online political participation. The same 
models are a bit less
predictive for online opinion leadership, and a bit more predictive 
for online alternative
view-seeking than online political participation. The activity 
represented by "online
opinion leadership" can be seen as requiring more purpose than both 
online political
participation, and online alternative view-seeking (which is the most 
passive of the three
dependent variables). The differences in predictive strength for 
these models may be due
to the fact that more factors are needed to explain more purposive 
action. The role of
digital literacy as a predictive for all three is an interesting and 
new result.
More work needs to be done to corroborate the time order and causality for the
claim which is theoretically implied in this paper, that digital 
literacy is a prerequisite for
online political participation. It is likely that the directions of 
influence between these
two variables are complex.
Surprisingly, post hoc analysis does not appear to indicate any 
interaction for age
and digital literacy on online political participation to the extent 
that old, high digital
literacy respondents might have a higher mean for online political 
participation than
24
Internet Use, Digital Literacy and Purposive Web Use
young high digital literacy respondents. This hypothesis will be 
further explored in
future research.
This study makes the leap to also explore the impact of a model of 
online political
participation on offline political participation in the hypotheses. 
While the data was
being analyzed, post hoc analysis was also performed to address the 
extent to which a
model including offline political participation indicates online 
political participation. The
higher predictive strength of the model for offline to online is not 
surprising, given that in
order to be active in online politics, one is likely to be active in 
offline politics as well.
The surprise was in the finding suggesting that for this sample, the 
model for online
political participation is indicative for subsequent participation in 
the real world. It may
be that online political participation is spurring offline activity 
for this group of
respondents. The results are also indicative that offline political 
participation is indicative
of online. More, possible qualitative, work could help establish the 
validity of these
claims by exploring time order for the variables, and by exploring 
how the process of
online political participation takes form.
25
Internet Use, Digital Literacy and Purposive Web Use
References
Bawden, D. (2001). Information and digital literacies: A review of 
concepts. Journal of
Documentation, 57 (2), 218-259.
Bikson, T. K, & Panis, Constantijn W. A. (1999). Citizens, computers, 
and connectivity:
Review of trends. Rand Corporation. Retrieved February 17, 2005 from
http://www.rand.org/publications/MR/MR1109.
Bonfadelli, H. (2002). The internet and knowledge gaps: A theoretical 
and empirical
investigation. European Journal of Communication, 17, 65-84.
Brown, J. A. (1998). Media literacy perspectives. Journal of 
Communication, 48 (1), 44-
58.
DiMaggio, P. & Hargittai, E. (2001). From the digital divide to 
digital inequality:
Studying internet use as penetration increases. Princeton: Center for Arts and
Cultural Policy Studies, Woodrow Wilson School, Princeton University.
DiMaggio, P., Hargittai, E., Celeste, C., & Shafer, S. (2004). 
Digital inequality: From
unequal access to differentiated use. In K. Neckerman (Ed.), Social Inequality
(pp. 355-400). NY: Russell Sage Foundation.
Eshet-Alkalai, Y. (2004). Digital literacy: A conceptual framework 
for survival skills in
the digital era. Journal of Educational Multimedia and Hypermedia, 13 (1), 93-
106.
Freire, P. & Macedo, D. P. (1987). Reading the Word and the World. 
New York: Bergin
& Garvey.
Gilster, P. (1997). Digital literacy. New York NY: Wiley.
Hargittai, E. (2005). Survey measures of web-oriented digital 
literacy. Social Science
Computer Review. Pre-print; article forthcoming.
26
Internet Use, Digital Literacy and Purposive Web Use
Hargittai, E. (2004). Internet access and use in context. New Media 
and Society, 6(1),
137-143.
Hargittai, E. (2003a). Informed web surfing: The social context of 
user sophistication.
In P. Howard and S. Jones (Eds.), Society Online: The Internet in Context (pp.
257-274). Thousand Oaks, CA: Sage Publications.
Hargittai, E. (2003b). The digital divide and what to do about it. In 
Jones, D.C (Ed.),
New Economy Handbook. San Diego, CA: Academic Press.
Hargittai, E. (2002). Second-level digital divide: Differences in 
people's online skills.
First Monday, 7 (4). Retrieved 1/26/05 from
http://firstmonday.org/issues/issue7_4/hargittai/.
Isbell, D. & Hammond, C. (1993) "Information literacy competencies" 
College and
Research Libraries News, June volume, 325-327.
Kim, Y. & Jung, J. (2002). Digital divide In 90% access: Multi-dimensional
examination of adolescents' internet connectedness in Seoul, Singapore and
Taipei. In Hwang, J., Yu, J., and Lee, J (Eds.), A Study on the 
Digital Divide of
the Youth, (pp. 219-247). Seoul: Korea Institute for Youth Development.
Kraut , R., Patterson, M. Lundmark, V., Kiesler, S., Mukophadhyay, 
T., & Scherlis, W.
(1998). Internet paradox: A social technology that reduces social involvement
and psychological well-being?. American Psychologist. 53(9), 1011-1031.
Livingstone, S., Bober, M., & Helsper, E. (2004). Active 
participation or just more
information? Young people's take up of opportunities to act and 
interact on the
internet. UK Children Go Online Project. Retrieved February 17, 2005 from
www.children-go-online.net.
27
Internet Use, Digital Literacy and Purposive Web Use
Livingstone, S., Bober, M., & Helsper, E. (2005a). Internet literacy 
among children and
young people: Findings from the UK Children Go Online Project. UK Children
Go Online Project. Retrieved February 17, 2005 from 
http://www.children-goonline.
net.
Livingstone, S., Van Couvering, E. & Thumim, N. (2005b). Adult media 
literacy: A
review of the research literature. London, UK: Office of Communications
(Ofcom). Retrieved February 1, 2005 from http://www.ofcom.org.uk/
consumer_guides/media_literacy/Medlitpub/aml.pdf.
Putnam, Robert D. (2000). Bowling Alone: The Collapse and Revival of American
Community. New York: Touchstone.
Rubin, A. M. (1998). Media literacy. Journal of Communication, 48 (1), 3-4.
28
Internet Use, Digital Literacy and Purposive Web Use
Table 1. Means and standard deviations for demographic variables
Variables
Age (in years)
Education (in years)
Table 2. Means and standard deviations for internet use variables
Variables
Internet time use1
History with internet 2
1 Time use was recoded as a multiplied combination of the items, "On 
average, how many days a week do
you use the internet?" and "About how much time a day do you spend on 
the Internet (minutes)?", and
standardized by dividing by 60 to reflect approximate hours per week 
spent with the internet
2 History with the internet represents an index of the two variables, 
"When did you first start using e-mail?
(in years)," and "When did you first start using the internet for 
purposes other than e-mail? (in years)"
which has a Cronbach's alpha of .86, and standardized to reflect the 
average for when an individual
began using both email and the Web.
SD Mean N
15.27 39.89 324
5.60 17.02 324
SD Mean N
324 23.92 28.19
324 3.82 10.80
29
Internet Use, Digital Literacy and Purposive Web Use
Table 3. Means and standard deviations for web task ability
Variables
Web task ability 1
1 Additive index of yes/no variables listed above. Mean for composite 
was standardized
by dividing mean by the total number of variables (8).
Do you know how to… (1=no, 2=yes)
Download a file from the internet to
your computer?
Send a file that is on your computer's
hard drive to someone using another
computer?
Open an attachment someone sent
you via e-mail?
Listen to audio CDs on your computer?
View video CDs/DVDs on your
computer?
Listen to audio files from the internet
(such as streaming audio)?
View video files from the internet?
Create a website?
Additive index2
N SD Mean
.17 324 1.97
.25 324 1.93
.096 324 1.99
.34 324 1.87
.39 324 1.82
.367 324 1.85
.39 324 1.81
.50 324 1.47
.20 324 1.84
30
Internet Use, Digital Literacy and Purposive Web Use
Table 4. Means and standard deviations for digital literacy variables
Variables
Basic digital literacy
(Hargittai, 2005) 1
Advanced digital
literacy
(Familiarity with the
following) 1
Information-seeking1
Searches per day
Created website
1 Responses were coded 5 = very familiar, 4 = familiar, 3 = neutral,
2 = unfamiliar, 1 = very unfamiliar.
2
3
4
5
Cronbach's alpha = .84
Cronbach's alpha = .81
Cronbach's alpha = .82
"Have you ever created a website?" (0 = no, 1 = yes), and "If yes, 
how many years ago
did you create a website" (measured as a ratio variable in years, 
with "never" represented
as 0) were multiplied as a combined variable
Familiarity with the following:
MP3
Downloading
Preference setting
Refresh/Reload
Newsgroup
PDF
Advanced search
Additive index2
JPG
Weblog, or "blog"
Instant messaging
Shareware
Additive index3
Library database
LexisNexis
OCLC FirstSearch
ProQuest
Communication Abstracts online
database
Additive index4
On average, how many times per day
you perform a search using a search
engine?
Composite variable5
N Mean SD
324
324
324
324
324
324
324
324
1.38
.65
1.14
.89
1.17
.88
.78
.72
3.51
4.54
4.00
4.48
3.66
4.42
4.46
4.15
324
324
324
324
324
1.08
1.23
1.25
1.32
.95
4.22
3.72
3.87
3.45
3.82
324
324
324
324
.88
1.34
1.47
1.48
4.37
3.80
2.70
2.99
324
324
1.40
1.02
3.50
3.48
12.00 324 8.15
324 3.96 2.47
31
1
Internet Use, Digital Literacy and Purposive Web Use
Table 5. Means and standard deviations for purposive use of the 
internet (online political participation, online
seeking of alternative views, and perceiving oneself as an opinion 
leader online)
Variables
Online Political participation 1
Perceiving oneself as an
opinion leader online 1
Seeking of alternative views
online1
Online news consumption1
Responses were coded 5 = strongly agree, 4 = agree, 3 = neutral, 2 = 
disagree, 1 = strongly
disagree
2 Cronbach's alpha = .784
3 Cronbach's alpha = .787
4 Cronbach's alpha = .798
On the Internet, I participate in discussions about
political topics
In my online interactions, I provide other people with
information about current events.
I express my opinions even when others disagree with
me online
I talk with people who support different political parties
than me online
Additive index2
People ask my opinion about current events online.
People consider me a good source of advice about
current events online.
I try to convince others to change their political views
online
I express my opinions even when others disagree with
me online
Additive index3
News from a variety of viewpoints are important to me
when accessing online news sites
On the Internet, I seek diverse news media with
different political views
On the Internet, I try to find diverse political views
different than my own
Points of view that are different from my own are
important to me when accessing online news sites
Additive index4
When accessing online news sites, I look for news
about politics.
N
324
Mean SD
1.14 3.87
324 1.26 2.86
324 1.29 3.30
324 1.26 3.07
324
324
.96
1.18
2.86
2.67
324 1.14 2.69
324 1.09 2.44
324 1.29 3.30
324
324
.92
.93
2.80
3.87
324 1.20 3.43
324 1.16 3.00
324 .90 3.43
324
324
.84
.87
3.34
4.14
32
Internet Use, Digital Literacy and Purposive Web Use
Table 6. Means and standard deviations for offline political participation
Variables
Offline political
participation 1
(Items included)
1Composite represents additive count of total number of actions taken 
by respondent in
the last twelve months from among this list of 13 activities
Composite
Voted in local elections
Volunteered to work for a political party in an election campaign
Contributed money to a political party or candidate
Attended a public meeting dealing with political or social issues
Spoken in such a meeting
Personally gone to see, or spoken to, or written to government
officials beyond the local level
Signed a petition
Collected signatures
Given a political speech
Attended a legal demonstration
Attended an illegal demonstration
Written to an editor about political news coverage or an editorial
Called a phone-in radio show regarding political, societal or
public issues.
N Mean SD
19.09 324 6.06
33
Internet Use, Digital Literacy and Purposive Web Use
Table 7. Pearson correlation coefficients for digital literacy variables a
1
2
3
4
5
6
7
a
Variables
Basic digital
literacy
(index, see
Table 4)
Advanced
digital literacy
(index, see
Table 4)
Information
seeking
(index, see
Table 4)
Web Tasks
(index, see
Table 3)
Website
creation
(composite,
see Table 4)
Avg.
searches per
day
Age
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Listwise N=383
4
.59**
3
.46**
2
.75**
.47** .39**
.37**
7
-.27**
6
.20*
5
.39**
-.18** .19* .39**
.013 .11* .30**
-.29** .17* .30**
.18* .028
-.074
34
1
2
3
4
5
6
7
8
9
a
Internet Use, Digital Literacy and Purposive Web Use
Table 8. Pearson correlation coefficients for demographic, internet 
use, digital literacy, and purposive web use
variablesa
Variables
Age (years)
Internet use
time (index, see
Table 2)
Internet use
history (index,
see Table 2)
Basic digital
literacy (index,
see Table 3)
Advanced digital
literacy (index,
see Table 3)
Online political
participation
online (index,
see Table 4)
Online opinion
leader (index,
see Table 4)
Online
alternative viewseeking
(index,
see Table 4)
Online internet
discussion
10 Offline political
participation
11 Education (in
years)
** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Listwise N=324
3 2
.460** -.149**
.032 @
@
@
@
@
@
@
6 5 4
.007 -.251** -
.135*
.226** .255** .259**
.106 .148* .126*
.195** .744**
.207**
8 7
.063 .005
.115* .202**
.078 .055
.227** .183**
.154** .198**
.316** .873**
.251**
11 10 9
.48** -.153** .051
-.066 -.052 .035
.321** .144** .003
-.022 .027 .122*
-.043 .065 .078
-.017 .219** .299**
-.041 .134* .272**
.047 .137* .059
.221** .021
.066
35
Internet Use, Digital Literacy and Purposive Web Use
Table 9. Hierarchical regression analysis of age, internet use 
variables, basic digital
literacy and information-seeking variables on advanced digital literacy index6
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3. Information-seeking and website
creation variables
- Information-seeking (index)3
- Created website (composite)4
- Searches per day
4. Basic digital literacy (index)5
a
b
p<.05
p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 History with the internet is an index of the two variables, "When 
did you first start
using e-mail? (in years)," and "When did you first start using the 
internet for purposes
other than e-mail? (in years)" (Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .82
4 Composite of "Have you ever created a website?" (0 = no, 1 = yes), 
and "If yes, how
many years ago did you create a website" (measured in years), which 
were multiplied
as a combined variable with 0 = "never"
5
6
Additive index, see Table 3. Cronbach's alpha = .84
Additive index, see Table 3. Cronbach's alpha = .81
Std.
beta
.008
.074 a
-.006
.056
.096 a
.012
.672 c
R-square Total
changes R-square
.038 c
.104 c
.176 c
.268 c
.038 c
.142 c
.318 c
.586 c
Adjusted
R-square
.035 c
.135 c
.307 c
.578 c
36
Internet Use, Digital Literacy and Purposive Web Use
Table 10. Hierarchical regression analysis of age, internet use 
variables, and basic
digital literacy on online political participation4
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3 Basic digital literacy (index)3
a p<.05
b p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 Index of the two variables, "When did you first start using e-mail? 
(in years)," and
"When did you first start using the internet for purposes other than 
e-mail? (in years)"
(Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .84
4 Index, see Table 4, Cronbach's alpha = .78
Std.
beta
.042
.186 c
.050
.169 b
R-square Total
changes R-square
.000
.057c
.023 b
.000
.057 c
.081 b
Adjusted
R-square
-.003
.049 c
.070 b
37
Internet Use, Digital Literacy and Purposive Web Use
Table 11. Hierarchical regression analysis of age, internet use 
variables, and basic
digital literacy on online opinion leadership4
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3. Basic digital literacy (index)3
a p<.05
b p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 Index of the two variables, "When did you first start using e-mail? 
(in years)," and
"When did you first start using the internet for purposes other than 
e-mail? (in years)"
(Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .84
4 Index, see Table 4, Cronbach's alpha = .79
Std.
beta
.080
.190 c
-.004
.155 b
R-square Total
changes R-square
.000
.049c
.020 b
.000
.049 c
.069 b
Adjusted
R-square
-.003
.041 c
.059 b
38
Internet Use, Digital Literacy and Purposive Web Use
Table 12. Hierarchical regression analysis of age, internet use 
variables, and basic
digital literacy on online alternative view-seeking4
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3. Basic digital literacy (index)3
a p<.05
b p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 Index of the two variables, "When did you first start using e-mail? 
(in years)," and
"When did you first start using the internet for purposes other than 
e-mail? (in years)"
(Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .84
4 Index, see Table 4, Cronbach's alpha = .80
Std.
beta
.167 b
.068
-.032
.312 c
R-square Total
changes R-square
.003
.022b
.077 c
.003
.025 b
.102 c
Adjusted
R-square
.001
.017 b
.092 c
39
Internet Use, Digital Literacy and Purposive Web Use
Table 13. Hierarchical regression analysis of age, internet use 
variables, and basic
digital literacy on online political news consumption4
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3. Basic digital literacy (index)3
a p<.05
b p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 Index of the two variables, "When did you first start using e-mail? 
(in years)," and
"When did you first start using the internet for purposes other than 
e-mail? (in years)"
(Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .84
4 "When accessing online news sites, I look for news about politics." 
Responses coded
5 = strongly agree, 4 = agree, 3 = neutral, 2 = disagree, 1 = 
strongly disagree
Std.
beta
.169b
.016
-.032
.053
R-square Total
changes R-square
.019 b
.001
.002
.019 b
.020
.022
Adjusted
R-square
.016 b
.012
.012
40
Internet Use, Digital Literacy and Purposive Web Use
Table 14. Hierarchical regression analysis of age, internet use 
variables, basic digital
literacy and online political participation on offline political 
participation5
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3. Basic digital literacy (index)3
a
b
4. Online political participation4
p<.05
p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 Index of the two variables, "When did you first start using e-mail? 
(in years)," and
"When did you first start using the internet for purposes other than 
e-mail? (in years)"
(Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .84
4 Index, see Table 4, Cronbach's alpha = .78
5 Composite variable, see Table 5
Std.
beta
.095
-.104 a
.057
.027
.231c
R-square Total
changes R-square
.016 a
.008
.004
.049 c
.016 a
.024
.028
.076 c
Adjusted
R-square
.014 a
.016
.017
.063 c
41
Internet Use, Digital Literacy and Purposive Web Use
Table 15. Hierarchical regression analysis of age, internet use 
variables, basic digital
literacy and offline political participation5 on online political 
participation4
Blocks of independent variables
1. Age (years)
2. Internet use variables
- Time spent using internet overall1
- History with the internet (index)2
3. Basic digital literacy (index)3
a
b
4. Offline political participation4
p<.05
p<.01
c p<.001
1 Recoded from hours and minutes into total minutes
2 Index of the two variables, "When did you first start using e-mail? 
(in years)," and
"When did you first start using the internet for purposes other than 
e-mail? (in years)"
(Cronbach's alpha = .86).
3 Additive index, see Table 3. Cronbach's alpha = .84
4 Index, see Table 4, Cronbach's alpha = .78
5 Composite variable, see Table 5
Std.
beta
.020
.199 a
.035
.154 b
.218c
R-square Total
changes R-square
.000
-.057 c
.023 b
.046 c
.000
.057 c
.081 b
.127 c
Adjusted
R-square
-.003
.049 c
.070 b
.115 c
42