|
IS THE INTERNET SHAPING OUR PERCEPTIONS AND ATTITUDE? A CULTIVATION
ANALYSIS PERSPECTIVE TO INTERNET USE
by
Madhukar Kumar
and
Robert Meeds
A.Q. Miller School of Journalism and Mass Communications
Kansas State University
105 Kedzie Hall
Manhattan, KS 66506-1501
Ph: (785) 532-6890
FAX: (785) 532-5484
e-mail: [log in to unmask]
Paper submitted to the Mass Communication and Society Division
for peer review for the 2003 Association for Education in Journalism and
Mass Communication national conference
Running head: Internet and Cultural Indicators
ABSTRACT
A secondary analysis was conducted to investigate possible relationships
between the amount of time a respondent spends on the Internet responses to
cultural indicator questions from the perspective of cultivation theory.
Data from the General Social Survey (2000) were analyzed and results showed
that some of the cultural indicator variables had significant relationships
with the amount of time a respondent spent on the Internet even when
demographic control variables were taken into consideration.
The growth of the Internet over the last few years cannot be stressed by
statistics alone. It has been phenomenal, to say the least. Many
researchers, scholars and technologists believe that changes and growth in
the use of the Internet are transforming people's economic and social life
(e.g., Kraut, Lundmark, Patterson, Kiesler, Mukopadhyay, & Scherlis, 1998).
This study looks at the current Internet use habits of people across the
United States and investigates questions primarily based on cultivation
analysis (Gerbner, 1969).
This study addresses questions to find out possible effects of the
Internet on society. The overarching question this study addresses is
whether there is a relationship between Internet usage and the respondents'
views, attitudes and their feelings towards issues which have been used as
cultivation indicators in the past.
The basic difference between the present study and most of the previous
studies on cultivation analysis is that instead of exposure to television,
this study investigates possible relationships between the cultivation
indicators and exposure to the Internet. Similar to numerous previous
studies on cultivation theory based on television viewing, this study does
not make assumptions of macro or micro societal effects of the media in
question (in this case, the Internet) but argues that if there are small
and persistent cultivation effects over a period of time then it may be
indicative of a greater effect, positive or negative, on the society.
This study specifically investigates possible relationships between
attitudes and views of Internet users and the amount of time spent on the
Internet from a cultivation analysis perspective. To investigate possible
relationships, the study analyzed the data collected by the General Social
Survey 2000.
It might be argued that the Internet is not comparable to television, hence
a study on cultivation analysis may be relevant with television viewers but
quite out of place with Internet users. It is our view, however, that the
Internet, though not directly comparable to television, can be viewed as an
amalgamation of several traditional mass media. We believe that the
Internet can be viewed as a superset mass medium with capabilities of
delivering contents of more than one traditional medium.
This view is supported DiMaggio, Hargittai, Neuman, & Robinson (2001), who
state that the Internet is a highly pliant medium and can act as several
traditional mass media at once. For example, the Internet can act as a
telephone by way of online chat programs, a broadcast medium, by way of
bulk email and chat rooms and it can also act as a collection of libraries
by way of online databases and search engines. In addition, some of the
same content carried in other media, including television, is directly
available on the Internet. For example, people have a choice of listening
to radio on the Internet from all parts of the world, download and watch
news programs, documentaries and even movies from all across the world and
also access information about television programs and additional content
related to television programs though web sites as well as through online
discussion groups and peer-to-peer communication.
The Internet can thus be viewed as a superset of the television medium.
However, the study makes no assumption that the Internet mirrors the same
homogeneity of content that cultivation theorist believe characterizes
television, but because of similarities between the two media, including
some content, it would be worthwhile to investigate possible cultivation
effects of the Internet and compare it with television usage results.
Cultivation theory
The main postulate of cultivation analysis as developed by George Gerbner
and his colleagues at the Annenberg School of Communication states that
television cultivates certain views, perceptions and preferences among its
viewers. They argue that since the masses use television as an
entertainment and information gathering medium, they are influenced by the
repetitive messages generated by television. Over a period of time, these
messages seep into the society and form a mainstream culture (Gerbner,
Gross, Morgan & Signorielli, 1986).
The central proposition of cultivation theory is that television viewers
who report watching greater amounts of television are more likely to
exhibit perceptions and beliefs that reflect television world views
(Potter, 1994). The theory asserts that heavy television viewers'
perceptions of society differ considerably from light viewers' perceptions.
For example, heavy viewers are more likely than light viewers to perceive a
higher incidence of racial problems (Volgy & Schwarz, 1980). Heavy
television viewers also have a tendency to have a dim view of the
trustworthiness of people (Carlson, 1993).
According to Morgan and Signorielli (1990), the methods and assumptions
behind cultivation analysis are different from those traditionally employed
in mass media studies. Although early and traditional mass communication
research investigating media effects focused mainly on individual messages,
programs, episodes or series and their "immediate effects" on the
audiences, cultivation analysis is concerned with more general and
pervasive effects of a medium on the audience in general, due to a
cumulative exposure to a particular medium and its messages, over a period
of time. Morgan and Signorielli (1990) claim that over the years much of
cultivation analysis research focused on television because of television's
dominance as a mass medium in the United States but that the same theory
can also be used to study any other communication medium.
Cultivation theorists argue that cultivation theory does not imply any sort
of a simple, linear relationship or stimulus-response model between
audience and media content. It, in fact, implies a long-term and a
cumulative consequence of repetitive and stable messages over a period of
time. According to Morgan and Signorielli (1990), it is not important that
there is a small change in individual's perception and attitudes over a
period of time. They argue that when taken cumulatively, small changes in
individual perceptions and attitudes could mean substantial changes in the
society at large.
Over the years a number of cultivation theory studies have been conducted
to investigate the relationship between exposure to television and
audience's perceptions and attitudes about certain societal issues. In its
simplest form, cultivation analysis tries to ascertain if those who spend
more time watching television are more likely to perceive the real world in
ways that reflect the most common and repetitive messages of the television
world compared to the people who spend less time watching television but
are comparable in important demographic characteristics (Morgan &
Signorielli, 1990).
Cultivation analysis studies over the years have shown that heavy and
light television viewers are different from each other in many ways, for
example, in terms of the extent to which television dominates a viewer's
source of consciousness. Cultivation theory, however, assumes that light
television viewers tend to be exposed to more varied and diverse
information sources, while heavy viewers tend to rely more on television.
In any case, the goal of cultivation analysis is to determine whether
differences in attitudes, perceptions, actions and beliefs of light and
heavy television viewers reflect their viewing patterns and habits,
independent of the social, cultural or other factors that differentiate
light and heavy viewers. Thus, cultivation theory tries to investigate the
independent contributions of television viewing to viewers' conceptions of
social reality (Morgan & Signorielli, 1990).
Cultivation theory also argues that television viewing gives rise to two
kinds of beliefs, termed first order and second order beliefs. First order
beliefs refer to misperceptions about social reality such as estimates of
violence rates and occupational roles of women. Second order beliefs, on
the other hand, refer to the opinions concerning such matters like meanness
of society, or sex role stereotypes. Second order cultivation analyses
usually focus on the development of values systems (Carlson, 1993).
Two other important postulates of the cultivation theory are resonance and
mainstreaming. According to Morgan and Signorielli (1990), resonance occurs
when certain views get emphasized among the television viewers who are
similar to each other in some ways. For example, a person who has gone
through a crime related experience in the past will have a more pronounced
view of crime rate being higher due to his world-view. Mainstreaming, on
the other hand, occurs when the larger mass tends to converge towards a
certain view or opinion due to long term exposure to television. Morgan and
Shanahan (1991) describe mainstreaming as a phenomenon that cultivates
homogeneity in an otherwise diverse group of television viewers.
Criticisms of cultivation theory
Some critics have argued that cultivation theory may no longer be
relevant because of technology developments that have dramatically changed
mass media. Gerbner (1990), in a defense to this argument, said that though
the media may change or the delivery system for the content may change, the
audience in general remains exposed to the same kind of content. He argued
that content itself had remained largely the same.
It may be argued that the time when Gerbner wrote the above defense to the
criticism, the Internet was not part of everyday life and though the
argument may well hold for the cable television and VCRs, it may not be
appropriate all to extrapolate it to the Internet. Indeed, it would be too
simplistic to say that the Internet is a "fancy new boutique in the same
old cafeteria" (Gerbner, 1990). However, though it is a fact the Internet
offers far more content from different "information wholesalers," at the
same time, it also offers the same content or supplement for the content
that continues to be broadcast in the television medium.
Cultivation theory has been criticized regarding the strength of the
relationships discovered between exposure to television and perceptions
about social reality and opinions. Although a large number of studies have
confirmed the cultivation hypotheses, the correlation coefficients between
television viewing and perceptions and attitudes have been in the range of
.10 to .20. In several studies when control variables were introduced, the
correlation coefficients dropped below statistical significance (Potter, 1994).
Gerbner and his associates responded to this argument and said that the
discovery of consistent systematic differences between heavy and light
viewers has far-reaching consequences. Gerbner argued that even a slight
but consistent change in cultivation perspective could change the entire
political culture of the society (Gerbner, 1986). He also said that it is
unreasonable to argue that television viewing affects everyone equally and
further research conducted by Gerbner and his associates suggested that
even in the absence of an overall relationship under controls from
demographic variables, evidence of strong relationships existed between
specific sub-groups (Morgan & Signorielli, 1990).
Potter (1994), in a seminal critique of the cultivation theory, said that
researchers must consider three important methodological questions before
designing a cultivation theory study: How should television exposure be
measured? How should cultivation perceptions be measured and what is the
appropriate test for the relationship between exposure and perceptions?
In measuring exposure to television there is no standard number of groups
in cultivation studies. While in one study, the respondents were placed in
groups of high, medium and low viewers (Gerbner, Gross, Morgan &
Signorielli, 1982; Volgy & Schwarz, 1980), in another study, the viewers
were placed into groups of high and low (Gerbner Gross, Morgan &
Signorielli, 1978) and in yet another viewers were placed into four groups
(Ogels & Sparks, 1989). A common criticism for measuring media exposure has
been that cut points were assigned arbitrarily to put the respondents in
different groups. Gerbner and his associates replied that the terms heavy
and light television viewers are not absolute but relative to aid in
investigating the relationship between medium exposure and views and
perceptions (Gerbner, Gross, Morgan & Signorielli, 1981).
Regarding the measurement of cultivation indicators, Potter claimed that
the central problem lies in determining what specific indicators to look
at. There have been a wide variety of measurements for cultivation
indicators. According to Potter (1994), though some researchers ask the
respondents to make estimates, others ask for their perceptions, beliefs or
attitude.
When asking the respondents about estimates, cultivation analysis
researchers have used a wide range of topics. These range from asking the
respondents to estimate rates of crime and violence to estimates about
personal victimization, number of people employed in law enforcement, rate
of divorce, etc. For perceptions, the range of topics includes the
respondents' perceptions about a mean world, perceptions about doctors,
perceptions about traditional sex roles, perceptions about American
stereotypes and perceptions about sexism (Potter, 1994).
For attitude, several previous cultivation analysis studies have looked at
the respondents' attitude towards blacks, personal conduct, communism, free
speech, federal spending and taxes (Gerbner, Gross, Morgan & Signorielli,
1982). Other topics of attitudinal measures used by cultivation analysis
researchers include police brutality and bias against civil liberties,
attitudes about free speech restrictions, racism, federal spending, sexual
tolerance, sexism, faith in others and political efficacy (Potter, 1994).
Potter's (1994) criticism about the measurement of the cultivation
indicator is that the cultivation theory provides no direction about what
specific topics to examine and also in justifying what should the
television world answer be. He also talks about "problematic" scaling
issues while measuring the cultivation indicators.
Although a number of cultivation analysis studies have also used various
methods to measure respondents' feelings and values, this study focuses
mainly on some of the attitude variables mentioned above.
Finally, regarding the measures used to test relationships between
television exposure and cultivation indicators, Potter claimed that the
evidence of the relationship may be spurious on account of two
methodological reasons. First, the effect is dependent on a variety of
"third" variables, and when the influence of these third variables is
controlled, cultivation effects virtually disappear. Second, the
relationship between television viewing and perceptions and attitudes may
not be linear and hence the use of statistics like gammas or Pearson
correlation may not be appropriate (Potter, 1994).
Based on the above discussions, this study poses three main research
questions.
RQ1: Is there a relationship between Internet usage and the respondents'
attitudes and perceptions?
RQ2: If there is a relationship then does that relationship hold when
income, age, education and party affiliation are taken into account?
RQ3: How do the relationships between respondents' attitudes and
perceptions and Internet usage, if any, compare to the relationships
between television usage and perceptions and attitudes?
The present study in relation to the above criticisms
In response to the criticisms about cultivation theory studies in which cut
points were arbitrarily introduced to separate high and low television
viewers, this study does not attempt to define any arbitrary cut points in
Internet exposure but merely investigates whether there is any relationship
or correlation between total Internet exposure and perceptions and
attitudes of respondents towards some issues by running statistical tests.
These statistical tests are described later in this study.
Potter's criticism about defining what exactly is a television world answer
is not relevant to this study because the aim here is not to find out
whether heavy Internet users tend to give more Internet-world answers but
merely to see if a relationship exists between Internet use and the views,
perceptions and attitudes of the Internet users. Since the study does not
assume that the Internet is homogeneous in its content like television, it
does not make any assumptions about any Internet-world answers. The study
investigates possible relationships between Internet usage and cultivation
indicator variables and then compares those with the relationships between
television and some cultural indicator variables.
The criticism on scaling issues was not addressed in this study.
Method
Data from the General Social Survey (GSS) of year 2000 (Internet module)
were used to investigate the research questions. The GSS survey is
conducted every one or two years by the National Opinion Research Center
(NORC) at the University of Chicago. The GSS was first conducted in 1972 to
observe social life and trends in the country and is highly regarded as a
valuable nationwide survey to investigate social issues.
Method of analysis
The study used the Survey Data Analysis (SDA) program available on the GSS
2000 module at the website http://webuse.umd.edu. The SDA allows a
researcher to conduct several tests on the GSS dataset through its web
site. A limitation to this method is that since raw data were not
available, the statistical measures were limited to the one available in
the online SDA program. It is also worth pointing out here that the SDA
does not allow factor analysis or principal component analysis that could
be used for data reduction, scaling and exploring structural relationships
among multiple variables.
According to Potter (1994), cultivation analysis researchers use
correlation coefficients to measure the relationship between the
perceptions and attitudes of the respondents and the respondent's exposure
to the medium. We investigated possible relationships between some
variables that define a respondent's attitude and/or views (dependent
variable) and the total amount of time spent on the Internet (independent
variable) by conducting logistic regressions. When significant
relationships were found between a dependent variable and Internet usage,
then additional logistic regressions were run with education, age, income
and political party affiliation as control variables.
For comparison purposes, we ran a similar set of tests with the same set
of dependent variables and the amount of time spent watching television as
the independent variable. Significant relationships were tested further
with the same set of control variables used in the first set of tests.
The variable measuring respondents' exposure to the Internet, television
and other dependent variables that were used for the tests are described
below. These variables are from the complete list of variables from the GSS
2000 data set.
Independent Variables
Internet usage - Total time spent on the Internet, combining all of the
e-mail and time spent on surfing websites on the World Wide Web (hours, 1
decimal place).
TV Hours - On the average day, about how many hours do you personally
watch television (hours, 1 decimal place)?
Dependent Variables
Racist speaker - Or consider a person who believes that Blacks are
genetically inferior. If such a person wanted to make a speech in you
community claiming that Blacks are inferior, should he be allowed to speak,
or not?
Racist teacher - Should such a person be allowed to teach in a college or
university, or not?
Racist Book - If some people in your community suggested a book he wrote
which said that Blacks are inferior should be taken out of your public
library, would you favor removing this book, or not?
Communist Speaker - Now, I should like to ask you some questions about a
man who admits he is a Communist. Suppose this admitted Communist wanted to
make a speech in your community. Should he be allowed to speak, or not?
Communist Teacher - Suppose he is teaching in a college. Should he be
fired, or not?
Communist Book - Suppose he wrote a book which is in your public library.
Somebody in your community suggests that the book should be removed from
the library. Would you favor removing it, or not?
Military Book - Suppose he wrote a book advocating doing away with
elections and letting the military run the country. Somebody in your
community suggests that the book be removed from the public library. Would
you favor removing it, or not?
Homosexual speaker - And what about a man who admits that he is a
homosexual. Suppose this admitted homosexual wanted to make a speech in
your community. Should he be allowed to speak, or not?
Homosexual teacher - Should such a person be allowed to teach in a college
or university, or not?
Death Penalty - Do you favor or oppose the death penalty for persons
convicted of murder?
Legalize Marijuana - Do you think the use of marijuana should be made legal
or not?
Anti-religion speaker - There are always some people whose ideas are
considered bad or dangerous by other people. For instance, somebody who is
against all churches and religion... If such a person wanted to make a
speech in your (city/town/community) against churches and religion, should
he be allowed to speak, or not?
Anti-religion teacher - Should such a person be allowed to teach in a
college or university, or not?
Women unsuited - Tell me if you agree or disagree with this statement: Most
men are better suited emotionally for politics than are most women.
Happy - Taken all together, how would you say things are these days--would
you say that you are very happy, pretty happy, or not too happy?
Exciting Life - In general, do you find life exciting, pretty routine, or dull?
Helpful People - Would you say that most of the time people try to be
helpful, or that they are mostly just looking out for themselves?
Trustworthy People - Generally speaking, would you say that most people can
be trusted or that you can't be too careful in dealing with people?
Space Spending - We are faced with many problems in this country, none of
which can be solved easily or inexpensively. I'm going to name some of
these problems, and for each one I'd like you to tell me whether you think
we're spending too much money on it, too little money, or about the right
amount. First the space exploration program...are we spending too much, too
little, or about the right amount on the space exploration program?
Crime Spending - Are we spending too much, too little, or about the right
amount on halting the rising crime rate?
Drug Spending - Are we spending too much, too little, or about the right
amount on dealing with drug addiction?
Affirmative Spending - Are we spending too much, too little, or about the
right amount on improving the conditions of Blacks?
Arms Spending - Are we spending too much, too little, or about the right
amount on the military, armaments, and defense?
Law Spending - Are we spending too much, too little, or about the right
amount on law enforcement?
Affirmative Assistance Spending - Are we spending too much, too little, or
about the right amount on assistance to blacks?
Results
The first set of tests with Internet usage as the independent variable
(see Table 1) showed that there were significant relationships between the
amount of time a respondent spent browsing the World Wide Web and emailing
and the following variables – Racist Book, Communist Speaker, Communist
Teacher, Communist Book, Military Book, Homosexual Speaker, Homosexual
Teacher, Anti-Religion Speaker, Anti-Religion Teacher, Women Unsuited,
Happy , Exciting Life, Trustworthy People, Space Spending, Drug Spending
and Arms Spending.
The variables that did not show any significant relationships were Racist
Speaker, Racist Teacher, Death Penalty, Legalize Marijuana, Helpful People,
Crime Spending, Affirmative Spending, Law Spending and Affirmative
Assistance Spending.
All the variables were tested after creating dummy variables with values 1
assigned to a particular response and 0 for all other responses. For
example, variable Racist Book asked respondents whether a racist book
should be removed from a library. A dummy variable was created and a value
of 1 assigned to all responses which said the racist books should be
removed and 0 for all other responses, such as "not removed" and "don't
know." Since the value of beta in this test was -.002 at a probability of
.005, it can be inferred that respondents who spend more time on the
Internet are more likely to say that a racist book should not be removed
from a library.
Similarly, respondents who spent more time on the Internet are more likely
to say that a communist should be allowed to speak in a community
(Communist Speaker, _ =.038, p = .000), a communist teacher should not be
fired from a school (Communist Teacher, _ =-.041, p =.000) and a communist
book should not be removed from a community library (Communist Book, _
=-.053, p = 0.000).
A respondent who spendt more time on the Internet was also more likely to
say that a militarist book should not be removed from a library (Military
Book, _ = -.025, p = .002), an admitted homosexual should be allowed to
make a speech in the community (Homosexual Speaker, _ =.090, p = .000) and
that the person should also be allowed to teach in a college or university
(Homosexual Teacher, _ =.068, p =.000).
A positive relationship was also found between Internet usage and
Anti-Religion Speaker (_ =.033, p = .001), which means that a respondent
who spents more time on the Internet was more likely to say that an anti
religionist should be allowed to make a speech in the community and also
more likely to disagree with the statement that women are not suited for
politics.
A person who spends more time on the Internet was also more likely to say
that an anti-religionist should be allowed to make a speech in a community
(Anti-Religion Teacher, _ =.019, p = .009), the government is spending too
little on fighting drug addiction (Drug Spending, _ =-.021, p = .008), not
spending too little on arms and the military (Arms Spending, _ =-.033, p =
.005) and spending too little on space exploration (Space Spending, _
=.028, p = .002).
Such a person was also more likely to disagree that women are not suited
for politics (Women Unsuited, _ =-.031, p = .000), was generally happy with
his/her life (Happy, _ =.044, p = .001), more likely to say that people can
be trusted and also more likely to say that their lives were exciting
(Exciting Life, _ =.013, p = .043).
Table 1 – Indicators with Internet usage
_ (Probability)
Racist Speaker
.006 (.331)
Racist Teacher
.000 (.990)
Racist Book
-.002 (.005)***
Communist Speaker
.038 (.000)***
Communist Teacher
-.041 (.000)***
Communist Book
-.053 (.000)***
Military Book
-.025 (.002)**
Homosexual Speaker
.090 (.000)***
Homosexual Teacher
.068 (.000)***
Death Penalty
0.010 (.083)
Legalize Marijuana
.012 (.068)
Anti-Religion Speaker
.033 (.001)***
Anti-Religion Teacher
.019 (.009)**
Women Unsuited
-.031 (.000)***
Happy
.044 (.001)***
Exciting Life
.013 (.043)*
Helpful People
.005 (.435)
Trustworthy People
.019 (.002)**
Space Spending
.028 (.002)**
Crime Spending
-.015 (.053)
Drug Spending
-.021 (.008)**
Affirmative Spending
.012 (.149)
Arms Spending
-.033 (.005)**
Law Spending
-.005 (.448)
Affirmative Assistance Spending
-.013 (.115)
When control variables of age, income, education and party affiliation were
introduced, the only variables that still had significant relationships
were Racist Book (_ = -.019, p = .040), Communist Teacher (_ =-.021, p =
.019), Communist Book (_ =-.028, p = .006), Homosexual Speaker (_ =.046, p
= .007), Homosexual Teacher (_ =.043, p = .004) Women Unsuited (_ =-.018, p
= .035) and Arms Spending (_ = -.024, p = .050).
Indicators with control variables and Internet usage (only significant
relationships)
Table 2
Internet Usage
Education
Age
Income
Democrat
Independent
Republican
DV1
Racist Book
-.019*
-.074**
.010**
-.096***
.638
.258
.372
DV 2
Communist Teacher
-.021*
-.074**
.024
-.073*
.125
-.041
-.081
DV 3
Communist Book
-.029**
-.136***
.016***
-.087**
.898
.604
.846
DV 4
Homosexual Speaker
.046**
.174***
-.013**
.117***
.211
.112
-.009
DV 5
Homosexual Teacher
.045**
.182***
-.023***
.088**
-.138
-.230
-.696
DV 6
Women Unsuited
-.019*
-.066**
.010**
-.031
.328
.602
.886
DV 7
Arms Spending
-.026*
-.033
.028***
.027
-1.066*
-.713
-.065
* = p < .05 , ** = p < .01, *** = p < .001
The variables that showed a significant relationship with the amount of
time spent on watching television were Racist Book (_ =.052, p = .046),
Communist Book (_ =.050, p = .056), Military Book (_ =.052, p = .047),
Women Unsuited (_ =.056, p = .012), Happy (_ =-.051, p = .018), Exciting
Life (_ =.091, p = .002), Trustworthy People (_ =-.070, .026), Crime
Spending (_ =-.034, p = .001), Drug Spending (_ =.096, p = .002),
Affirmative Spending (_ =.062, p = .019) and Affirmative Assistance
Spending (_ =.072, p = .015).
Logistic regressions between the above mentioned variables and the variable
TV Hours (Table 3) showed that respondents who spent more time watching
television were more likely to say that a communist book should be removed
from a community library (Communist Book), a militarist book should be
removed from a community library (Military Book) and also more likely to
agree with the statement that women are not suited for politics (Women
Unsuited).
Similarly, a respondent who spent more time watching television was more
likely to say that he/she is not very happy in life (Happy ), life is
exciting (Exciting Life), people in general cannot be trusted (Trustworthy
People) and that too little is being spent to halt the rising crime rate
(Crime Spending). Respondents who watch more television were also more
likely to say that we are spending too little to improve the condition of
Blacks (Affirmative Spending and Affirmative Assistance Spending) and that
the government is not spending too little in fighting drug addiction.
Table 3 – Indicators with TV Hours
_ (Probability)
Racist Speaker
-.009 (.721)
Racist Teacher
-.039 (.140)
Racist Book
.052 (.046)*
Communist Speaker
-.001 (.983)
Communist Teacher
.046 (.085)
Communist Book
.050 (.056)*
Military Book
.052 (.047)*
Homosexual Speaker
-.023 (.455)
Homosexual Teacher
-.004 (.905)
Death Penalty
-.021 (.304)
Legalize Marijuana
-.020 (.515)
Anti-Religion Speaker
.022 (.476)
Anti-Religion Teacher
-.046 (.082)
Women Unsuited
.056 (.012)*
Happy
-.051 (.018)*
Exciting Life
.091 (.002)**
Helpful People
-.007 (.796)
Trustworthy People
-.070 (.026)*
Space Spending
.006 (.868)
Crime Spending
-.034 (.001)*
Drug Spending
.096 (.002)**
Affirmative Spending
.062 (.019)*
Arms Spending
.050 (.079)
Law Spending
.023 (.420)
Affirmative Assistance Spending
.072 (.015)*
The variables that still had a significant relationship even when control
variables of age, education, income and political party affiliation and
were introduced were Crime Spending (_ = .090, p = .007), Drug Spending (_
=.086, p = .010), Affirmative Spending (_ =.040, p = .170) and Affirmative
Assistance Spending (_ =.064, p = .052).
Table 4 - Indicators with control variables and TV Hours
TV Hours
Education
Age
Income
Democrat
Independent
Republican
DV1
Crime Spending
.089**
-.089**
-.005
.033
-.324
-.712
-.610
DV 2
Drug Spending
.088**
-.081**
-.001
-.001
-.177
-.315
-.446
DV 3
Affirmative Assistance
.065*
.067*
-.004
-.135***
.180
-.171
-1.290
* = p < .05 , ** = p < .01, *** = p < .001
Table 5 - TV hours and control variables (Multiple Regression)
Regression Coefficients
Test That Each Coefficient = 0
B
SE(B)
Beta
SE(Beta)
T-statistic
Probability
educ
-.185
.022
-.204
.024
-8.400
.000
age
.012
.004
.078
.024
3.292
.001
income
-.124
.029
-.102
.024
-4.217
.000
partyid(d: 0-1) Democat
.633
.537
.116
.098
1.178
.239
partyid(d: 2-4)
.435
.535
.084
.103
.813
.417
partyid(d: 5-6)
.317
.539
.054
.092
.587
.557
Constant
5.786
.690
8.386
.000
Multiple R =
.280
R-Squared =
.078
Std Error of Estimate =
2.441
Table 6 – Correlation between TV hours and control variables
Correlation Matrix
educ
age
income
partyid(d: 0-1)
partyid(d: 2-4)
partyid(d: 5-6)
tvhours
educ
1.00
-.11
.26
-.03
-.09
.11
-.24
age
-.11
1.00
.02
.11
-.15
.05
.10
income
.26
.02
1.00
-.05
-.05
.11
-.16
partyid(d: 0-1)
-.03
.11
-.05
1.00
-.58
-.40
.07
partyid(d: 2-4)
-.09
-.15
-.05
-.58
1.00
-.49
.00
partyid(d: 5-6)
.11
.05
.11
-.40
-.49
1.00
-.06
tvhours
-.24
.10
-.16
.07
.00
-.06
1.00
Table 7 - Internet usage and control variables (Multiple Regression)
Regression Coefficients
Test That Each Coefficient = 0
B
SE(B)
Beta
SE(Beta)
T-statistic
Probability
educ
.767
.065
.252
.021
11.747
.000
age
-.062
.011
-.116
.021
-5.565
.000
income
.146
.089
.035
.021
1.655
.099
partyid(d: 0-1)
.814
1.418
.043
.076
.574
.566
partyid(d: 2-4)
.749
1.409
.042
.080
.531
.595
partyid(d: 5-6)
2.028
1.424
.102
.071
1.425
.155
Constant
-5.852
1.907
-3.069
.002
Multiple R =
.310
R-Squared =
.096
Std Error of Estimate =
8.275
Table 8 – Correlation matrix between internet usage and control variables
Correlation Matrix
educ
age
income
partyid(d: 0-1)
partyid(d: 2-4)
partyid(d: 5-6)
netime
educ
1.00
-.11
.24
-.04
-.09
.12
.28
age
-.11
1.00
.03
.11
-.14
.04
-.14
income
.24
.03
1.00
-.06
-.03
.10
.10
partyid(d: 0-1)
-.04
.11
-.06
1.00
-.57
-.40
-.05
partyid(d: 2-4)
-.09
-.14
-.03
-.57
1.00
-.49
-.04
partyid(d: 5-6)
.12
.04
.10
-.40
-.49
1.00
.09
netime
.28
-.14
.10
-.05
-.04
.09
1.00
Discussion
The main postulate of cultivation theory is that television cultivates
certain views, perceptions and attitudes among its viewers. These views and
attitudes are the result of repetitive messages generated by television
that slowly and gradually seep into the masses over a period of time
(Gerbner, Gross, Morgan & Signorielli, 1986).
Cultivation theory studies in the past have shown that there are
significant relationships between the amount of television exposure and
views, perceptions and attitudes (Rubin, Perse & Taylor, 1988) of the
respondents. This study was trying to investigate whether any similar
relationships exist between the same cultural indicator variables and
exposure to the Internet. The second research question that the study
addressed was that if such relationships are remained significant even
after control variables are taken into account.
With respect to the first research question, the tests showed that
significant relationships existed between Internet usage and some of the
cultural Indicators (see Table 1). However, some critics of cultivation
research have argued in the past that existence of significant
relationships may not be indicative of any effects of the medium and that
the relationships could be because of intervening variables which, if taken
into account, usually result in no significant relationships at all (Rubin,
Perse & Taylor, 1988). The analyses reported here showed that some of these
relationships remained significant even when control variables were
introduced.
The results suggest that cultivation effects, though small, may exist
between the amount of time people spend on the Internet and their opinions
and attitudes about certain issues.
When the same set of variables were used to test relationships with
television exposure, it was found that there were significant relationships
among several variables and the amount of time a respondent spends watching
television. The second set of tests is consistent with the main postulate
of the cultivation theory – that there is a relationship between the amount
of time respondents spend watching television and their opinions and
attitudes on certain issues.
Some of the relationships with television viewing were consistent with
other cultivation research studies (like having less trust in other people,
viewing most of the people as not being very helpful, having a view that
life is not very exciting and agreeing that women are not suited for
politics) (c.f, Rubin, Perse & Taylor, 1988).
However, few of the variables had significant relationships with television
exposure when control variables of education, age, income and political
party affiliation were introduced. The variables that still had significant
relationships after control variables were introduced were Crime Spending,
Drug Spending and Affirmative Assistance Spending.
An important finding in this study is that many of these relationships that
were observed with television exposure were in the opposite direction as
compared to Internet usage. For example, although respondents who spend
more time on the Internet were more likely to say that they were happy with
their lives and people in general can be trusted, those who spent more time
watching television were more likely to say that they were not happy with
their lives and that people in general cannot be trusted.
This suggests that, though cultivation theory may be still relevant to the
Internet, the messages generated by this new medium are quite different
from television. And as such, the views of respondents who spend more time
on the Internet are in some cases the opposite of those who watch more
television. However, when control variables were accounted for, only one
variable, Crime Spending, had significant beta values in opposite directions.
The variables that had significant relationships in exactly opposite
direction without taking the control variables into account were Racist
Book, Communist Book, Military Book, Women Unsuited, Happy, Trustworthy
People and Drug Spending. The opposing views of those who watch more
television and those who use the Internet more are explicated below.
Respondents who spent more time on the Internet were more likely to say
that racist, communist and militarist books should not be removed from a
community library. Respondents who spend more time watching television are
more likely to say that these books should be removed from the library.
Respondents who spent more time on the Internet were more likely to say
that the government is spending too little on fighting drug addiction.
Those who watch more television were more likely to say that the government
is not spending too little on the fighting drug addiction.
Similarly, respondents who use the Internet more were also more likely to
say that they are happy with their lives, and that people in general can be
trusted. Those who watch more television were more likely to say that
people in general cannot be trusted and that they are not happy with their
lives.
The results show that those who spend more time on the Internet were more
liberal and different in views as compared to those who watch television.
This study does not make assertions that the Internet, as compared to
television, has more positive or negative effects on its users at large.
However, the results in this study do show that it would be worthwhile to
conduct more studies to investigate more differences in the Internet users
and television viewers.
It would be also worthwhile to investigate the Internet usage behavior of
the respondents and then build them in studies to find out their
differences in views and perceptions about certain issues. For example just
as respondents who watch more day time television serials are more likely
to have perceptions about higher divorce rate, it may be possible that the
more liberal views of a high Internet user may be related to the kind of
web sites he/she may be frequenting.
Constraints and future work
The present study looked at data for only one year, so it does not show
whether the relationships discovered between Internet usage and cultivation
indicators persist over a period of time. It would be worthwhile to look at
the GPS data for subsequent years and see if the same relationships persist.
If the relationships do persist over the years, it may be indicative of a
causal effect between Internet usage and certain views, perceptions and
attitudes of the Internet users. These relationships should also be
compared with relationships between television exposure and cultivation
indicators to investigate if there are any changing patterns and to
ascertain whether the two mass media, television and the Internet, are
actually generating completely different repetitive messages.
The other constraint in this study is that unlike some previous cultivation
analysis studies, this study did not do any content analysis of the web
sites visited by the respondents. As discussed earlier, future studies
should also account for the breakdown of the respondents' Internet usage
habits to investigate whether there are any direct relationships with the
cultivation indicators and the content of the web sites that the
respondents' frequent.
Finally, because this study used the online SDA program to do data
analysis, it suffered from the constraints of doing more thorough
statistical tests like factor analysis. If the raw data is made available
to the researchers in the future or more statistical tests are added to the
SDA program then some constraints related to the statistical tests could be
eliminated from future studies on the GSS data set.
References
Bryant, J. (1986). The road most traveled: Yet another cultivation
critique. Journal of broadcasting and electronic media, 30(2), 231-235.
Carlson, M.J. (1993). Television viewing: Cultivating perceptions of
affluence and support for capitalistic views. Political Communication, 10,
243-257.
DiMaggio, Paul, Eszter Hargittai, W. Russell Neuman, & John P.
Robinson. (2001). Social Implications of the Internet. Annual Review of
Sociology. 27, 307-336.
Gerbner, G. (1969). Dimensions of violence in television drama. In
R.K.Baker & S.J.Ball (Eds), Violence in the media. Staff report to the
National Commission on the Causes and Prevention of Violence (pp. 311-340).
Washington, DC: U.S. Government Printing Office.
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1978). Cultural
Indicators: Violence profile no. 9. Journal of Communication, 28(3), 176-207.
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1981). A curious
journey into the scary world of Paul Hirsch. Communication Research, 8, 39-72.
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1982). Charting the
mainstreaming: Television's contributions to political orientations.
Journal of Communication, 32(2), 100-127.
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1986). Living with
television: The dynamics of the cultivation process. In J.Bryant and
D.Zillman (Eds), Perspective of media effects. Hillsdale, NJ: Lawrence
Erlbaum.
Gerbner, G. (1990). Advancing on the path of righteousness. In N.
Signorielli & M. Morgan (Eds), Cultivation analysis: New directions in
media effects research (pp. 13-34). Newbury Park, CA: Sage.
Kraut, R., Lundmark, V., Patterson, M., Kiesler, S., Mukopadhyay, T., &
Scherlis, W. (1998). Internet paradox: A Social Technology That Reduces
Social Involvement and Psychological Well-Being? American Psychologist,
53(9), 1017-1031.
Morgan, M & Shanahan, J. (1991). Television and the cultivation of
political attitudes in Argentina. Journal of Communication, 41(1), 88-102.
Ogles, R.M., & Sparks, G.G. (1989). Television violence and viewers'
perceptions of criminal victimization. Mass Communication Review, 16, 2-11.
Morgan, M. & Signorielli, N. (1990). Cultivation analysis:
Conceptualization and methodology. In N. Signorielli & M. Morgan (Eds),
Cultivation analysis: New directions in media effects research (pp. 13-34).
Newbury Park, CA: Sage.
Perse, E.M., Ferguson, D.A. & McLeod, D.M. (1994). Cultivation in the newer
media environment. Communication Research, 21(1), 70-104.
Potter, James W. (1993). Cultivation theory and research: A conceptual
critique. Human Communication Research, 19(4), 564-601.
Potter, James W. (1994). Cultivation theory and research: A methodological
critique. Journalism Monographs 147.
Robinson, John P., Meyer Kestnbaum, Alan Neustadtl, & Anthony
Alvarez. (2000). Mass Media Use and Social Life Among Internet
Users. Social Science Computer Review, 18 (4), 490-501.
Rubin, A.M., Perse, E.M. & Taylor, D.S. (1988). A methodological
examination of cultivation. Communication Research, 15(2), 107-134.
Volgy, T.J., & Schwarz, J.E. (1980). TV entertaining programming and
sociopolitical attitudes. Journalism Quarterly, 57, 150-155.
|
