This paper was presented at the Association for Education in Journalism and
Mass Communication in Toronto, Canada, August 2004.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
[log in to unmask] 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 "").
(Oct 2004)
Thank you.
Elliott Parker
************************************************************************
The Internet and Its Impact on Traditional News Media Use:
A Test of the Principle of Relative Constancy
By Mugur Geana
Doctoral Student
School of Journalism
University of Missouri
Columbia, MO 65211-1200
[log in to unmask]
573-771-0314
And Wayne Wanta
Professor
School of Journalism
University of Missouri
Columbia, MO 65211-1200
[log in to unmask]
573-884-9689
**Manuscript submitted to the Communication Theory & Methodology Division
for consideration of presentation at the annual convention of the
Association for Education in Journalism and Mass Communication, Toronto,
August 2004.
**Geana is a doctoral student and Wanta a professor in the School of
Journalism at the University of Missouri.
The Internet and Its Impact on Traditional News Media Use:
A Test of the Principle of Relative Constancy
Abstract
Using data from a national survey (N=3002), this study examines how time
spent using the Internet is impacting traditional media use. Results go
directly counter to the notion of the Principle of Relative
Constancy. Internet use did not substitute for traditional media use but
instead was positively related to traditional media use. While being
online searching for news contributes to information overload, Internet
users nonetheless perceived this information overload positively.
The Internet and Its Impact on Traditional News Media Use:
A Test of the Principle of Relative Constancy
As a developed society, we greatly depend upon the mass media. From the
moment we wake up in the morning until we turn off the lights at night, we
are constantly surrounded by radio, television and newspapers, magazines
and the Internet. We have a multitude of options for our daily dose of
information. We can turn on the television set in the morning or open the
newspaper to get a first glimpse of the early news and, as the day goes by,
we can continue our quest of exploiting the media, both as a means of
keeping a close contact with reality or to avoid it altogether.
Advances in science and technology over the last century have continuously
shaped and diversified our media outlets, made news more interesting, even
provided us with "condensed" news for the busy consumer, but technology has
not been able to supply us with more available time in which to search and
absorb this information. Today, like 100 years ago when newspapers were the
only main news outlet, the day has the same 24 hours.
Building on the Principle of Relative Constancy (McCombs, 1972), the
present study proposes to take a snapshot at the impact of using the
Internet as a news outlet on the time allocated for traditional media
sources. Furthermore, by building upon a psychological trait described as
"need for cognition" (Cacioppo and Petty, 1982), the study will explore the
perception of information overload by Internet users compared with
traditional media users.
As a relatively recent media outlet, the Internet has had an enormous
impact on society. Many newspapers now include their web addresses on their
front pages, URLs where you can get supplementary information about a
product or service are displayed with advertisements, and email addresses
are as common as telephone numbers on business cards. All this for a medium
that only 10 years ago was out of reach for the majority of the public.
Previous studies on the impact of new media on traditional media sources
used longitudinal approaches that extended over a time period of 10 years
or more (McCombs & Son, 1986, Noh, 1994), and focused on the economic
component of McCombs' principle.
As opposed to the family revenue quota allocated for communication issues,
which varies in close relationship to the GDP and the economy (McCombs &
Eyal, 1980), the daily disposable time available for media interaction has
far more strict variation limits and could be the critical determinant of
media usage (Linder, 1970). The present research looks towards the actual
users of online news and how their interaction with the Internet influences
their time allocation and usage of traditional media sources, trying to
offer insights to the question of whether new media are a replacement for
traditional media or only a new complementary source.
By comparing informational overload perceptions between Internet news users
and non-users, the study will try to provide information regarding eventual
differences in baseline overload perceptions between the two media user
categories and why people who are already exposed to information
bombardment through traditional media channels choose to go online in
search of more information. Data come from the 2002 media consumption
survey conducted by the Princeton Survey Research Associates.
The Principle of Relative Constancy
Developed by Maxwell McCombs in 1972, the Principle of Relative Constancy
builds upon Charles E. Scripps's hypothesis on patterns of economic support
of mass media (Scripps, 1965):
"...the amount of economic support provided for mass communication media
consistently follows the ebb and flow of the general economy.[...] The
important point is that this ebb and flow follows the general economy, not
the competition and technological changes within mass communication
industries." (McCombs, 1972).
By performing an extensive analysis of media expenditures from 1929 to
1968, McCombs acknowledged support for Scripps' Constancy hypothesis in
regard to the media as a whole, but identified very distinct trends among
individual media categories. Disparate studies by previous researchers have
identified correlations among personal income and newspaper circulation
(Peterson, 1950), personal income and early diffusion of radio sets (Willey
& Rice, 1933), and national income, business activity and newspaper
circulation (Levin, 1954). Using as a reference previous research and
DeFleur's comparative analysis of the diffusion of mass media in the United
States (DeFleur, 1970), McCombs stated that, while consumer expenditure on
mass media has remained relatively constant in regard to household income
between 1929 and 1968, the media-mix that was addressed by these
expenditures has changed across time, influenced by competition,
technological development, and introduction of new media The Principle of
Relative Constancy.
Subsequent empirical studies (McCombs & Eyal, 1980; Werner, 1986; Son 1990;
Wood & O'Hare, 1991; Caroll, 2002) have supported the theory that consumer
expenditures on mass media vary in relation with the fluctuations of
household income and economic indicators. Other studies (Dupagne, 1997)
have identified variations in consumer spending among different media
categories, mainly with the introduction of new media communication
technologies.
More recent research has criticized this principle, especially relating to
the time interval selected for the study (Glascock, 1993), and regarding
applying the same methodology to time periods with advent of new media
technologies that show a faster rate of adoption by the consumer. Dupagne
(1997) criticized the Principle of Relative Constancy as "not being
validated in economic terms and research did not draw any connections
between this principle and standard economic theory." Son (1990) found that
mass media expenditures between 1929 and 1987 did not correlate with
increases in income, and she suggests that introduction of new media
technologies on the market (cable TV and VCR) has caused increased consumer
spending on the mass media after 1975, compared with earlier periods.
The majority of the articles that criticize McCombs' initial findings
include in the analyzed data the period after 1970, when new media
technologies with a steeper adoption curve made their debut on the
mass-media panoply, compared to the period between 1929 and 1968 (analyzed
by McCombs in the initial study), when television was the only
technological novelty (with an adoption curve that spanned more than 25
years Rogers, 1986).
Even though some researchers criticized the principle regarding the
correlated link between mass media expenditures and income as not
conforming to economic theory, McCombs' assumption on competition and
partial or total replacement among media components received wide support
from the academic world. The concept of functional equivalence (McCombs,
1972), an integrant part of the principle of relative constancy, has
demonstrated economic validity. According to this concept, consumers have
to trade the amount of resources available (time and money) towards the
different media components. Preference will be given to the media channel
with the greatest degree of utility in comparison with other similar
sources (Noh, 1994, Dupagne, 1997). This could be a valid explanation for
the evolution of media over the last 100 years in regard to the rise and
fall of some media channels when technological advances allowed for
complementary alternatives (Rogers, 1986).
Although the main body of research regarding the Principle of Relative
Constancy addressed its economic implications, there is another dimension
to McCombs' postulate that poses the same amount of challenge to any
researcher. In his Mass Media in the Marketplace, McCombs cites Staffan
Linder that:
"Economic growth has brought a proliferation of goods, gadgets and
services. The consumption of these costs time. As consumption continues to
increase time may become increasingly scarce." (McCombs, 1972)
There is a daily time limit towards which we can increase the time-slot
dedicated to media consumption. Although this limit is individual-related
and dependent on a personal balance with other daily activities, media
consumption has become an integrant part of our lives that we not only
cannot ignore anymore, but we must account for. If we extrapolate McCombs'
principle to time spent consuming media, we could say that "Time spent
consuming mass media is limited; as a consequence, any new media should
compete with traditional media for time slots in our busy schedules."
Available data regarding mass-media consumption over the years and
allocated consumption time shows that, at a societal level, we have not yet
reached a media-consumption saturation point. We are spending more and more
time each year consuming media. A report published in Communication
Industry Forecast in 2002 by Veronis Suhler Stevenson shows that the
average time spent consuming mass media has increased from 9.032 hours/day
in 1996 to 9.64 hours/day in 2001, with a forecasted time-spending of 9.997
hours/day in 2005. Of all the components of the media mix, the Internet is
by far the most dynamic medium, with increases of over 19.4 times of
allocated consumption time in 2005 compared to 1996. At the same time,
television consumption will account for an increase of only 1.076 times in
2005 vs. 1996, while newspapers consumption time will decrease by an
estimated 12% in 2005 compared with 1996.
The possible online news competition versus newspaper news is treated very
seriously by the print industry. On July 1st, 2003, the International
Newspaper Marketing Association (INMA) emailed more than 700 of its most
prominent members from all over the world, asking them to name a research
topic that would be of interest to them. Although the answers were diverse,
one of the three topics that emerged as common interests from the multitude
of responses received was "The effects of media proliferation and increased
competition, particularly from online news" (Wilkinson, 2003). If the
current trend will prevail, it is estimated (Veronis Suhler Stevenson,
2002) that by 2005 the Internet will have 35% more time allocated for media
consumption than newspapers (compared to only 6% of newspapers time in 1996
an increase by 26.5 times).
Thus, based on the above scenarios, we propose to test the following
research questions and hypotheses:
RQ1: How does being active online influence consumption of traditional media?
RQ2: According to the concept of functional equivalence, is online news
displacing newspapers in the consumer's daily media consumption basket?
H1: Users of online news will spend less time consuming general traditional
media than users who do not read online news.
H2: Users of online news will spend less time consuming traditional news
media than users who do not read online news.
H3: Users of online news will spend less time consuming traditional
entertainment media than users who do not read online news.
The first three hypotheses are based on the Principle of Relative
Constancy. Individuals have limited time available to them. If
individuals use the Internet for purposes previously devoted to traditional
media, the time allotted to traditional media should decrease. Individuals
not using the Internet, then, should continue to use traditional media at
the same rates as before. Thus, non-Internet users should use traditional
media more than Internet users. This trend will be found for general media
(H1), news media (H2) and entertainment media (H3).
Need for Cognition
First postulated and empirically studied in 1955 by Cohen, Stotland and
Wolfe, the Need for Cognition is a "need to structure relevant situations
in meaningful, integrated ways. It is a need to understand and make
reasonable the experiential world" (as cited in Cacioppo and Petty, 116).
Failure to achieve this cognitive balance could generate feelings of
"tension and deprivation" and incite to take action towards restoring the
structural equilibrium (Cohen, Stotland and Wolfe, 1955). Overall, the need
for cognition explains the availability of an individual to engage in and
enjoy thinking (Petty and Cacioppo, 1982).
After subsequent testing of a scale to evaluate the need for cognition,
Petty and Cacioppo (1986) included the instrument as a major construct of
the Elaboration Likelihood Model (ELM), a milestone in persuasion research.
Perloff (1993) categorizes the influence of need for cognition on the
effectiveness of persuasion:
1. For high-need of cognition subjects, messages that encourage
issue-relevant thinking should be maximally influential,
2. For low-need of cognition subjects, peripheral cues should be more
effective, and
3. Attitudes based on central route processing of an issue should persist
over a longer period of time, be more resistant to counterattacks, and show
more attitude-behavior consistency than attitude changes caused by
peripheral cues.
Recent research (Ariely, 2000; Das, 2001) emphasized the role of the
Internet as a stimulating and challenging informational medium for
high-need of cognition consumers, due to its interactivity and the
possibility of the consumer to be in control of both the informational flow
and the level of detail of the retrieved information. The amount of
information available on the web will not be perceived as a burden, but as
an opportunity to stimulate cognition and exercise control over the
information gateway.
H4: The perceived information overload of online news users will not differ
from those who only acquire news from traditional media sources.
H5: Users of online news will have a more positive perception of the
information overload compared to traditional news users.
Methodology
The model for this study is presented in Appendix A. As the model shows,
we are mainly concerned with differences between individuals who use only
traditional media (the right side of the model) and individuals who use
both the Internet and traditional media (the left side of the
model). Three main variables are examined: A general media use index
(GMIndex), a news use index (NIndex) and an entertainment use index
(EIndex). If the Principle of Relative Constancy is true here, time
allotment across the two groups will be similar. Thus, if one group uses
traditional media only, individuals in that group will show higher
traditional media use than individuals in the group that uses both
traditional media and the Internet. In other words, time spent online
should take away from time spent with traditional media.
This relationship also should be found when looking at specific types of
media use media use for news and media use for entertainment as well as
overall for all general media. This is depicted in the model by the broken
line at the bottom left representing the impact of Internet usage on news
and entertainment media use.
A final concern here is whether either group feels especially overloaded by
information, of if they enjoy the information that they have access to.
Data
The Princeton Survey Research Associates in Princeton, New Jersey, conducts
a biannual media consumption survey of US consumers. The present study uses
the dataset from the latest available survey, conducted in 2002. The
dataset, questionnaire and coding book are publicly available at the Pew
Research Center for the People and the Press website:
http://people-press.org/dataarchive/
The reasons for using this dataset is because the telephone survey that
generated the data explored uses of both traditional media and new media,
was administered to a representative sample of the US population (N=3,002)
and was conducted by a certified research firm.
The survey consisted of a total of 72 research questions and 26 demographic
questions. For the purpose of the present study we used the following
questions: 8 18, 25, 38, 56-58. A list of the questions used in the study
is presented in Appendix B.
The research questions were divided as follows:
38 Questions that were asked to all the participants 3,002 participants
17 Questions pertinent to form A (administered to 1,551 participants
51.7% of sample)
17 Questions pertinent to form B (administered to 1,451 participants
48.3% of sample)
SPSS was used to run all recoding of variables and statistical calculations.
Since the questions regarding time spent using a specific medium were asked
only to those who received form A of the survey, the dataset was filtered
for form A only, which produced a working sample of 1,551 cases.
Question 25 (How frequently do you go online to get NEWS... would you say
every day, 3 to 5 days per week, 1 or 2 days per week, once every few
weeks, or less often?) was used as criterion for dividing the dataset into
Internet news users and traditional media users. In order to eliminate any
ambiguity regarding participant classification, all cases that answered
Don't know/Refused on question 25 were deleted from the filtered dataset.
The remaining cases (N=981) constituted the final working sample.
The next step was to recode question 25 into two categories: Internet news
users (NetNUsers) and traditional news users (TrdNUsers). We considered as
Internet news users all the participants who went online to get news at
least 1 or 2 days per week (recoded 1, 2, and 3 into 1) and as traditional
media users those who used the Internet less frequently to get news
(recoded 4, 5, and 6 into 2). The reason this criteria for recoding of
variables was selected is that, for the purpose of the present research, we
are interested in those who have a habit of searching news online (at least
1 or 2 days per week).
A frequency analysis of the Internet Usage (INUsage) variable, obtained by
computing the sum of the previous two recoded variables revealed that the
working dataset consisted of 574 cases of Internet news users and 407 cases
of traditional news users.
In the survey, the questions regarding time spent using diverse media used
different scales of measurement, which varied from a 5 items scale for
newspaper use to an 8 items scale used for magazines. In order to ensure
consistency of measurement over all the variables, a 3 items scale was used
to recode all the time variables. The 3 items scale was congruent with each
of the previous scales, hence no interpolation was necessary. The 3 items were:
1. Media usage less than 30 minutes daily value 1
2. Media usage between 30 minutes to an hour daily value 2
3. Media usage for more than an hour daily value 3
Cases previously coded Don't know/Refused were coded with the value 0 (zero).
The recoded variables were checked for the assumptions of normality,
linearity and homoscedascity.
In order to evaluate the amount of time used by the participants with the
media, a General Media Index (GMIndex3) is defined as the sum, for each
participant, of the corresponding values of recoded variables for questions
9, 11, 13, 14, 16 and 18. The GMIndex3 could be any integer between 0 (no
media usage) and 18 (usage of each and all media more than 1 hour/day). The
GMIndex3 (N=981, Mean=5.9419, Skewness=.221, Kurtosis=-.425) provided a
scale wide enough to discern measurable variations of media usage between
the researched groups.
The traditional media measured by the survey can be divided into news and
entertainment media. In order to be able to compare the Internet effect on
time allocated for each of the categories, a News Index (NIndex3) was
computed as the sum, for each participant, of questions 9, 11 and 14 and an
Entertainment Index (EntrIdx3) was computed as the sum, for each
participant, of questions 13, 16 and 18. Each of the indexes could have
values between 0 and 9.
For daily radio and book usage, separate variables were recoded, since the
initial dataset did not include a yes/no question for these media
categories. RadioYN and BookYN are the recoded variables.
In order to operationalize the research questions and hypotheses, initial
correlations were run between the two Internet news usage groups
(Internet/traditional) and the selection of a specific media (media
yes/no); a second analysis tested for any correlations between the amount
of time spent on the Internet and the amount of time using a specific medium.
One-way analysis of variance (ANOVA) was used to test for differences
between general media usage, traditional news media usage and entertainment
media usage between the studied groups.
The survey included a set of questions regarding how the respondents'
experience with online news influenced their usage of traditional media
outlets (Q56 Q58: asked only to those who go online for news at least 1
or 2 days/week). To conduct an analysis of how usage of traditional media
was influenced by the frequency of searching for online news, the whole
initial dataset (N=3,002) was filtered to select only those who go online
at least 1 or 2 days/week (Q25). A sample of 1,087 cases was constructed.
Crosstabulation and Chi-square were used to test for significant
relationships between the usage frequency of online news and the usage of
traditional media outlets, both as overall usage and then structured for
individual media categories (newspapers, television, magazines, radio and
other).
Since the question regarding information overload was only asked to those
participants who answered form B of the survey, in order to test for the
need of cognition and information overload between the study groups, the
original dataset was filtered for form B (1,451 participants) of which,
after eliminating the cases with Don't know/ Refused responses for question
25, furnished a working sample of 931 cases. The same recoding as for form
A regarding frequency of Internet news usage was applied for the second
working dataset.
The new variable (NetUsage2) consisted of 513 cases of News Internet Users
and 418 cases of Traditional Media Users for the second working sample.
Crosstabulation and Chi-square were used to test relationships among
studied variables.
Results
The assumption is made that the observations are independent. SPSS was used
to run the statistical analysis. Every time a working dataset was compiled
or a variable recoded, it was checked for the assumptions of normality,
linearity and homoscedascity. P-P plots were used to check for outliers.
Descriptive statistics of the first working sample used (N=981) regarding
usage of media categories within the sample are presented in Table 1.
Not everybody uses the same media all the time, and media usage varies
according to the predominant media outlet used. Overall,
70.6% (N=381) of those who watched TV news also watched entertainment
programs on TV and 43.3% (N=190) of those who did not watch the news
actually used the TV for entertainment purposes;
63.2% (N=270) of those who read a newspaper yesterday also watched the news
on TV yesterday, like 48.5% of those who did not read a newspaper yesterday;
66.0% (N=282) of those who read a newspaper yesterday also watched
entertainment TV, like 52.3% (N=287) of those who did not read the newspaper;
50.5% of those who read a daily newspaper yesterday also listened to radio
news, as did 41.9% of those who did not read a newspaper yesterday.
30.0% (N=128) of those who read the newspaper also read a magazine
yesterday, and only 21.9% (N=120) of those who did not read newspapers read
magazines.
The analysis of the initial compiled working sample (N=981) was used to
test if overall traditional media usage remained the same between the two
researched sampling categories (Net news users / Traditional news users).
The General Media Index, News Index and Entertainment Index were compared.
Drawing upon the Principle of Relative Constancy, our first hypothesis (H1)
stated that "users of online news will have the same amount of consumption
of traditional media as users who do not read online news." One-way ANOVA
was run in order to test the assumption of equivalence between groups. The
test of homogeneity of variance result was .270, infirming the null
hypothesis. The results are presented in Table 2.
The result shows that a statistical significant (p=.001) difference exists
between the two groups. The observed results are opposite to the prediction
in our first hypothesis that non-Internet users will devote more time
consuming traditional media than will Internet users. The mean score for
Internet users (6.242) was actually larger than the mean for non-Internet
users (5.518) on the general media use index.
After dividing the media components into news outlets and entertainment
outlets, the next two hypotheses tested for equivalence of
news/entertainment media consumption between the studied groups.
Our second hypotheses (H2) posited that users of online news will have less
consumption of traditional news sources as users who do not read online
news. One-way ANOVA was run in order to test differences between groups.
The test of homogeneity of variance result was .219, infirming the null
hypothesis. The results are also presented in Table 2.
The result shows that a statistical significant (p=.000) difference exists
between the two groups. The observed results again run opposite to the
prediction in our hypothesis on the differences in time spent consuming
traditional news between those who search for news online and those who
only resort to traditional news media outlets. Internet users (3.054)
reported using traditional news media more than non-Internet users (2.526).
Our third hypotheses (H3) stated that "users of online news will have less
consumption of traditional entertainment media as users who do not read
online news." One-way ANOVA was run in order to test the assumption of
equivalence between groups. The test of homogeneity of variance result was
.184, infirming the null hypothesis. The results are presented in Table 2.
The result shows that a statistical significant difference does not exist
between the two groups. The observed results did not support our third
hypothesis. The two groups showed an equivalent amount of time spent
consuming traditional entertainment outlets.
It is important to note that these results were obtained on controlling
strictly for those who get news online and that the results are not
influenced by any major discrepancy in technological availability of
hardware that would prevent any of the non-online news users to access
online news content. This is confirmed by responses for question 23 "Do you
use a computer at your workplace, at school, at home or anywhere else on at
least an occasional basis?" The recorded frequency of computer usage for
our analyzed sample was 100%.
One of the research questions posted by this study addressed an en-vogue
issue in the media industry today: How does being active online in search
of news influence consumption of traditional media?
Phi correlations, which test relationships between two nominal level
variables, were computed for the option to use or not a specific media
(media yes/no) and the use of the Internet to gather online news (online
news yes/no). The findings are presented in Table 3.
The results show that statistically significant positive correlations exist
between the use of the Internet to access online news and the choice of
media outlets like newspapers, TV news and magazines and a negative
significant correlation with radio as a chosen media outlet.
To further try to identify the influence that searching for news online has
on the usage of traditional media outlets, correlations between usage of
online news and the time allocated to traditional media outlets were
tested. The results are presented in Table 4. The only significant
correlation regarding time spent with traditional media and Internet news
usage was identified with radio.
The last test performed regarding how searching for news online influences
the usage of traditional media was to test the correlations between the
frequency of searching for online news and the amount of daily usage of
traditional media. The results are presented in Table 5.
The only significant correlation regarding time spent with traditional
media and Internet news usage was identified with TV news, a negative
correlation.
Analyzing the data presented earlier, it is clearly that Internet usage to
get online news has an influence on the traditional media usage, mainly on
the choice of traditional news outlets. Interesting is the fact that,
except for TV news usage time, which showed a negative correlation vs. both
usage on the Internet to get online news and the weekly frequency of online
news use, the rest of the traditional media outlets daily usage seems not
to be influenced by resorting to online news.
As an answer to our first research question, and referring to the Principle
of Relative Constancy, we could say that usage of online news influences
our selection of traditional media components for our daily media mix
consumption, but doesn't have much influence (except for TV news) on how we
allocate time between chosen outlets.
The second research question addressed the influence of being a user of
online news on the usage of newspapers as a traditional media outlet. The
third constructed working sample was used for this analysis, constituted
from those who use online news at least 1 or 2 days/week (N=1087).
Question 56 of the survey (Q56) asked: "since you started getting news
online, are you using other sources of news MORE often, LESS often, or
about the same as you used to?" A descriptive of the answers to Q56 is
provided in Table 6.
The vast majority of online news users did not express a change in their
behavior towards traditional media news sources (72.1%), while the
percentage of those who did show an evident pattern change was
approximately equal (10.3% more usage vs. 11.9% less usage).
Of those who declared they use more often the traditional media since
they've starting using online news (N=112), 24.1% declared they use more
newspapers whereas 26.8% used television more often.
Of those who declared they use traditional news outlets less since they've
started to go online for news (N=186), 38.2% of them use newspapers less
while 28.5% watch television less. Table 7 shows the descriptive statistics
for this question.
In order to evaluate how frequency of searching for news online influences
the decision to use traditional media news sources, correlation and
likelihood ratio Chi-squares were run between Q25 (frequency of going
online) and Q56 (usage of traditional news sources after starting to get
online news), after filtering the dataset for Don't know/Refused in Q25.
(N=1082). The results are presented in Table 8.
The results of the likelihood ratio Chi-square tests confirm a
statistically significant difference among the observed values of
traditional news outlets in respect to the frequency of Internet usage for
online news.
Running the same test for the individual traditional media outlets that are
read more often or less often did not carry any significant statistical
results; it did not show any influence of the frequency of online surfing
for news on any specific traditional news outlet.
As a result, a definitive answer to our second research question cannot be
given. The analyzed data showed that as more time is spent online searching
for news, the more influence it will have on the time allocated to gather
news from traditional media sources. This influence works both ways, with a
minor predominance of those who renounce traditional media sources over
those who actually use more traditional media sources since they've started
to get online news (11.9% vs. 10.4%). An interesting observation is that,
for those who declared that they use MORE traditional news sources since
they've been getting news online, the largest percentage (31.3%) orient
themselves towards alternative sources, while television (26.8%) and
newspapers (24.1%) only come as second and third choices. On the other
hand, newspapers (38.2%) clearly dominate in the options of those who use
LESS traditional media, followed by television (28.5%) and radio (12.4%).
It can be said that the frequency of online news usage has a definitive
influence on the usage of traditional news, with newspapers being the first
choice of the consumers when they have to decide towards which traditional
news outlet to use LESS.
The result also provides indirect support for the concept of functional
equivalence (McCombs, 1972). As more time is spent searching for online
news (thus getting closer to the saturation point regarding the maximum
time that can be allocated for media consumption), the more influence it
has on the replacement of traditional news outlets with new media, in order
to keep the time balanced and a relative equilibrium.
The present study also proposed to analyze the overload perception of those
who use online news compared to the perception of those who use only
traditional media outlets. A working sample (N= 931) was prepared for this
analysis. In order to facilitate further analysis and normalize the data,
it was decided to eliminate from the working sample all cases that answered
with Don't know/Refused or Other to the overload perception question (Q38).
The final working sample numbered 866 cases. Distribution of their answers
to Q38 is presented in Table 9.
A crosstabulation was run between the recoded Q38 and Internet news usage.
A Chi-square test was applied to the observed distribution. The results of
the crosstabulation and the Chi-square test are presented in Table 10.
The results show that is a statistical significant difference between
online news users and traditional news users in respect to their perception
of information overload, online users feeling more overloaded. As a result,
our fourth hypothesis (H4) that states that "the perceived information
overload of online news users will not differ from those that only acquire
news from traditional media sources." was not supported by the results of
our study, the supplemental information acquired through the Internet
exposure contributing to the perceived informational overload.
Our fifth hypothesis (H5) stated that "users of online news will have a
more positive perception of the information overload compared to
traditional news users." The computed data presented in table 15 and table
16 provide strong support to our hypothesis, 72.1% of the users of online
news declaring that "they like" the information overload, compared with
only 27.9% of those who limit themselves to traditional news sources.
Our results are consistent with recent research (Ariely 2000, Das 2001)
that emphasized the role of the Internet as a stimulating and challenging
informational medium for high-need of cognition consumers, due to its
interactivity and the possibility of the consumer to be in control of both
the informational flow and the level of detail of the retrieved information.
Discussion
The present study attempted to examine differences in time allocated to
traditional media use for Internet news users and non-Internet news
users. Based on the Principle of Relative Constancy, we expected
traditional news media use to be inversely related to Internet news
use. Since time spent reading about news is finite, time devoted to
reading news online should detract from time devoted to news consumption in
traditional media.
Not only did we not find support for the Principle of Relative Constancy,
but we found the opposite trends: The Internet users also used traditional
news media more than non-Internet users. Internet users, then, did not use
Internet news consumption to replace traditional news media consumption,
but rather used the Internet to supplement other media.
This result here points to a potential next step for future
research. Because Internet use and traditional media use are positively
related, are individuals' using the Internet to supplement traditional
media or are they using traditional media to supplement Internet use. In
other words, either the Internet or traditional media could be the dominant
source of information for individuals. Internet-dominant news consumers
could differ greatly from traditional media-dominant consumers.
While the positive relationship between the Internet and traditional media
use was found for both the news media and all general media, no
relationship was found for entertainment media. This finding could be
explained by the purposes behind usage patterns. Respondents may consider
the Internet to be an excellent source of information especially for news
stories on the issues of the day. However, they may continue to think of
television as the main source of entertainment. This, of course, may
change as individuals grow more accustomed to using the Internet for
purposes beyond information seeking.
While the ANOVA findings point to an association between high Internet news
use and high traditional news media use, the results here are not all good
news for media professionals. Although online news users used more
traditional news media than non-online news users, they nonetheless
reported using certain media less than in the past. Online users perceived
that they are more likely to use LESS traditional media sources than MORE,
and newspapers are on the top of the list of traditional outlets to be
discarded or used less.
An argument towards "why the newspaper," could be provided by the need for
cognition theory. Our results clearly show that users of online news are
content with the informational overload, content of being in control of
their informational search and input. Opposite to television and radio,
which offers a certain degree of input control (changing channels in order
to search for more information regarding a subject of interest), the
newspaper information is limited to that available when edited, offering
little or no possibility of informational flow and input control.
Regardless, further research is needed to clear up this apparent
contradiction. A possible explanation could be that while Internet users
use traditional news sources more than non-users, they subconsciously could
believe that they somehow must be using traditional media less. Possibly,
then, they could perceive that they are lowering their traditional media
use as their Internet use rises as the Principle of Relative Constancy
predicts while in actuality the added time they are allotting to Internet
news use is coming from a different source.
Overall, our findings do not support the Principle of Relative
Constancy. Future studies, however, should continue to test this
concept. Clearly, the respondents here demonstrate that Internet news use
does not necessarily lead to decreases in traditional media use. Whether
this continues to be the case in the future is uncertain. Internet users
may eventually reach a saturation point, where all available time will be
allotted. With no non-committed time left, individuals may then have to
decrease time allotted to traditional media use if they want to continue
using the Internet.
Table 1: Media use within the first working sample (N=981)
Read a daily newspaper yesterday
43.5%
Watched news on TV yesterday
55.0%
Watched TV (no news) yesterday
58.3%
Listen to news on the radio yesterday
45.6%
Spent some time reading magazines yesterday
25.3%
Read a book yesterday
41.1%
Table 2. Mean scores and ANOVA results comparing general media use index
for Internet news users and traditional news users
Internet Traditional F Sign.
News Users News Users
General media use 6.242 5.518 11.323 .001
S.D. 3.256 3.460
News media use 3.054 2.526 15.295 .000
S.D. 2.072 2.101
Entertainment 3.188 2.993 1.924 .166
media use
S.D. 2.120 2.251
Table 3: Results of correlations between the option to use or not a
specific media and the use of the Internet to gather online news
Internet news usage (yes/no)
Media outlet usage (yes/no)
Correlation coefficient
Significance (Phi)
(p=.05)
Newspapers
.073
.032*
TV News
.069
.036*
TV Entertainment
.014
.058
Radio
-.068
.034*
Magazines
.064
.112
Books
-.026
.414
Table 4: Results of correlations between the times spent daily with
traditional media outlets and the use of the Internet to gather online news
Internet news usage (yes/no)
Media outlet usage (time)
Correlation coefficient
Significance (Phi)
(p=.05)
Directional (Eta)
Media usage as dependent
Newspapers
-.019
.251
.033
TV News
-.087
.283
.071
TV Entertainment
.035
.858
.038
Radio
.048
.016*
.029
Magazines
.049
.730
.042
Books
-.030
.441
.003
Table 5: Results of correlations between the times spent daily with
traditional media outlets and the frequency of Internet use to gather
online news
Internet news usage (time)
Media outlet usage (time)
Correlation coefficient
Significance
(p=.05)
Newspapers
-.054
.262
TV News
-.092
.032*
TV Entertainment
.024
.567
Radio
.050
.121
Magazines
.018
.775
Books
-.074
.136
Table 6. Perceived usage of traditional media news outlets by those who
also use online news
Freq. Percent
More often 112 10.3
Less Often 129 11.9
Some more, 57 5.2
others less
About the same 784 73.2
Don't know/Refused 5 .5
Total 1087 100
Table 7. Responses to which traditional sources are now being used less by
Internet news users.
Freq.
Percent
Newspapers
71
38.2
Television
53
28.5
Magazines
19
10.2
Radio
23
12.4
Other
13
7.0
Don't know/Refused
7
3.8
Total
186
100
Table 8. Cross-tabulation and chi-square results comparing how frequently
respondents use online news and their current usage patterns of traditional
media.
How frequently use online news
Every day
3-5 days/week
1-2 days/week
Total
Usage of traditional news outlets since started getting news online.
More often
13.3%
9.2%
6.9%
10.4%
Less Often
16.9%
10.9%
5.3%
11.9%
Some more, others less
5.1%
4.9%
5.9%
5.3%
About the same
64.8%
75.0%
81.9%
72.5%
Chi-square (df 6) = 37.416, p = .000
Table 9. Responses to whether respondents feel overloaded with information
or like all the information.
Freq. Percent
Overloaded 207 23.9
Like it 659 76.1
Total 866 100
Table 10: Crosstabulation and chi-square results comparing information
overload perception and usage of online news.
Internet usage
Online news
Traditional news
Total (count)
Do you feel overloaded by all the information?
Overloaded
58.9%
41.1%
207
Like it
72.1%
27.9%
659
Total (count)
597
269
866
Chi-square (df 1) = 12.704, p = .000
Addendum A Model of the study
Addendum B.
[INTERVIEWER NOTE: FOR QUESTIONS 8 thru 14 PLEASE DETERMINE WHAT DAY OF THE
WEEK IT IS. IF THE DAY OF THE WEEK IS SUNDAY, PLEASE READ "FRIDAY." IF THE
DAY OF THE WEEK IS NOT SUNDAY, READ "YESTERDAY."]
8 (T) Now, on another subject... Did you get a chance to read a daily
newspaper yesterday, or not? {5-00}
1 Yes
2 No
9 Don't know/Refused
[FORM A ONLY:]
[IF "YES" (Q.8=1) ASK:]
9 (A) About how much time did you spend reading a daily newspaper
yesterday? [DO NOT READ] {5-00}
1 Less than 15 minutes
2 15-29 minutes
3 30-59 minutes
4 1 hour or more
9 Don't know/Refused
[ASK ALL:]
10 (T) Did you watch THE NEWS OR A NEWS PROGRAM on television yesterday, or
not? {5-00}
1 Yes
2 No
9 Don't know/Refused
[FORM A ONLY:]
[IF "YES" (Q.10=1) ASK:]
11 (A) About how much time did you spend watching the news or any news
programs on TV yesterday? [DO NOT READ] {5-00}
1 Less than 15 minutes
2 15-29 minutes
3 30-59 minutes
4 1 hour or more
9 Don't know/Refused
[ASK ALL:]
12 (T) Apart from news, did you watch anything else on television
yesterday, or not? {5-00}
1 Yes
2 No
9 Don't know/Refused
[FORM A ONLY:]
[IF "YES" (Q.12=1) ASK:]
13 (A) About how much time did you spend watching TV yesterday, not
including the news? [DO NOT READ] {5-00}
1 Less than a half hour
2 30 minutes to less than one hour
3 About an hour or more
4 More than one hour but less than two hours
5 Two hours to less than three hours
6 Three hours to less than four hours
7 Four hours or more
9 Don't know/Refused
[ASK ALL:]
14 (T) About how much time, if any, did you spend listening to any news on
the radio yesterday, or didn't you happen to listen to the news on the
radio yesterday? [DO NOT READ] {5-00}
1 Less than 15 minutes
2 15-29 minutes
3 30-59 minutes
4 1 hour or more
5 Did not listen
9 Don't know/Refused
[FORM A ONLY:]
[INTERVIEWER: IF THE DAY OF THE WEEK IS SUNDAY, BEGIN THIS SERIES WITH
_Now, thinking about yesterday..._ OTHERWISE BEGIN WITH _Again, thinking
about yesterday..._]
15 (A) Did you spend any time reading magazines? {5-00}
1 Yes
2 No
9 Don't know/Refused
[IF "YES" (Q.15=1) ASK:]
16 (A) About how much time did you spend reading magazines yesterday? [DO
NOT READ] {5-00}
1 Less than five minutes
2 Five to less than ten minutes
3 Ten to less than fifteen minutes
4 Fifteen to less than twenty minutes
5 Twenty to less than thirty minutes
6 Thirty minutes to less than one hour
7 One hour or more
9 Don't know
[FORM A ONLY:]
17 (A) Not including school or work related books, did you spend any time
reading a book yesterday? [IF YES ASK: Was it a work of fiction or
non-fiction?] {10-99 Values}
1 Yes -- Fiction
2 Yes -- Non-fiction
3 Yes -- Both
4 No
5 Yes -- Don't know whether fiction/non-fiction
9 Don't know/Refused
[IF "YES" (Q.17=1,2,3,5) ASK:]
18 (A) About how much time did you spend reading books yesterday? [DO NOT
READ]
1 Less than five minutes
2 Five to less than ten minutes
3 Ten to less than 15 minutes
4 15 to less than 20 minutes
5 20 to less than 30 minutes
6 30 minutes to less than one hour
7 One hour or more
9 Don't know/Refused
25 (T) How frequently do you go online to get NEWS... would you say every
day, 3 to 5 days per week, 1 or 2 days per week, once every few weeks, or
less often? {5-00}
1 Every day
2 3-5 days per week
3 1-2 days per week
4 Once every few weeks
5 Less often
6 No/Never [VOL.]
9 Don't know/Refused
38 (B) Some people say they feel overloaded with information these days,
considering all the TV news shows, magazines, newspapers, and computer
information services. Others say they like having so much information to
choose from. How about you... do you feel overloaded, or do you like
having so much information available? {5-00}
1 Overloaded
2 Like it
3 Other [VOL.]
9 Don't Know/Refused
[IF GOES ONLINE FOR NEWS ONCE-A-WEEK OR MORE (Q.25-1,2,3) ASK:]
Earlier you said you get some news from the Internet
56 (T) Since you started getting news online, are you using other sources
of news MORE often, LESS often, or about the same as you used to? {5-00}
1 More often
2 Less often
3 Some more, others less [VOL.]
4 About the same
9 Don_t know/Refused
[IF "MORE OFTEN (Q.56=1) ASK:]
57 (T) Which sources are you using MORE often? [DO NOT READ LIST; ACCEPT
MULTIPLE RESPONSES.] {5-00}
1 Newspapers
2 Television
3 Magazines
4 Radio
5 Other
9 Don_t know/Refused
[IF "LESS OFTEN" OR "SOME MORE, SOME LESS" (Q.56=2,3) ASK]
58 (T) Which sources are you using LESS often? [DO NOT READ LIST; ACCEPT
MULTIPLE RESPONSES.] {5-00}
1 Newspapers
2 Television
3 Magazines
4 Radio
5 Other
9 Don_t know/Refused
References
Cacioppo, John T. and Richard E. Petty (1982). "The Need for Cognition,"
Journal of Personality and Social Psychology 42 (January), 116-131.
Caroll B., (2002). Newspaper Readership and Email Immediacy: Can a print
publication have its cake and eat it, too? Convergence: The Journal of
Research Into New Technologies v. 8, no. 3 (Autumn 2002): 78-96.
Cohen, Arthur R., Ezra Stotland, and Donald M. Wolfe (1955). "An
experimental investigation of need for cognition," Journal of Abnormal &
Social Psychology, 51, 291-294.
DeFleur, M.L., (1970). Theories of Mass Communication. New York: David
McKay Company.
Dupagne, M. (1997). A theoretical and methodological critique of the
principle of relative constancy. Communication Theory 7(1), 53-76.
Economic Support of Mass Communication Media in the United States,
1929-1963. (1965) Cincinatti: Scripps-Howard Research, 1965. p4. Cited by
McCombs, M. E. (1972). "Mass Media in the Marketplace," Journalism
Monographs 24, 1972.
Glascock, J. (1993). Effect of cable television on advertiser and consumer
spending on mass media, 1970-1990. Journalism Quarterly 70(3), Autumn, 509-517.
Levin J.L., (1954). "Competition Among Mass Media and the Public Interest,"
Public Opinion Quarterly, 13:226
Linder, S.B. (1970). "The Harried Leisure Class," New York: Columbia
University Press.
McCombs, M. E. (1972). "Mass Media in the Marketplace," Journalism
Monographs 24, 1972.
McCombs, M. E., & Eyal, C. H. (1980). Spending on mass media. Journal of
Communication 31(1), 153-158.
McCombs, M. E., & Son, J. (1986, August). Patterns of economic support for
mass media during a decade of electronic innovation, Paper presented at the
annual convention of the Association for Education in Journalism and Mass
Communication, Norman, OK.
Noh, G. Y. (1994, August). New media departure in the principle of relative
constancy: VCRs. Paper presented to the annual convention of the
Association for Education in Journalism and Mass Communication, Atlanta, GA
Perloff, R. M. The Dynamics of Persuasion. Hillsdale, New Jersey: Lawrence
Erlbaum Associates, Inc., 1993.
Peterson, W (1959). "Is Daily Circulation Keeping Pace With the Nation's
Growth?" Journalism Quarterly, 36:12-22
Rogers, E.M. (1986). Communication Technology: The New Media in Society.
New York: The Free Press
Son. J. (1990). The impact of new electronic media on audience support for
mass media. Dissertation Abstracts International, 51, 3263A. (University
Microfilms No. 91-05,666), cited by Dupagne, M. (1997). A theoretical and
methodological critique of the principle of relative constancy.
Communication Theory 7(1), 53-76.
Werner, A. (1986). Mass media expenditures in Norway: The principle of
relative constancy revisited. In M. L. McLaughlin (Ed.), Communication
yearbook 9 (pp. 251-260). Beverly Hills, CA: Sage
Willey M.M. and Rice S.A., (1933). Communication Agencies and Social Life.
New York: McGraw-Hill Book Co.
Wood, W. C., & O'Hare, S. L. (1991). Paying for the video revolution
consumer spending on the mass media. Journal of Communication 41(1),
Winter, 24-30
|
