Religiosity and Mass Media
Student Debut Paper
A Structural Equation Model of Religiosities Effect on
Mass Media Use and Civic Participation
Greg G. Armfield
University of Missouri – Columbia
Paper submitted to the Annual meeting of the Association for Education in
Journalism and Mass Communication (AEJMC) in Kansas City, MO, July 30 –
August 2, 2003. All correspondence should be directed to 115 Switzler Hall,
Columbia, MO 65211, [log in to unmask], Voicemail:(573) 882-6486 ext.8.
A Structural Equation Model of Religiosities Effect on Mass Media Use and
Civic Participation
Abstract
This research looks at the relationship of mass media, religion, and
secularization theory with regard to civic participation. A structural
equation model was created and successfully tested to test the structural
interactions of religion, secularism, and mass media in relation to civic
participation. Findings show current mass media use by religious
individuals is driven by secularization theory (Buddenbaum, 1986;
Buddenbaum & Stout 1996). Findings also lead the author to believe that
secularization and religiosity may not be viewed on a continuum as
previously believed (see Martin, 1965; Shiner, 1967).
A Structural Equation Model of Religiosities Effect on Mass Media Use and
Civic Participation
The Internet continues to emerge as a significant communication technology
with a vast number of opportunities for scholars to understand the impact
of technology on human behavior through communication sciences (Jackson,
1997). Recent empirical work in this area has looked at the needs gratified
through the use of new media (Flanagin & Metzger, 2001) and calls have been
made to further our understanding of this new media (Ruggiero, 2000).
Current work has looked at Internet use and sociability among such
variables as education, income level, and age (Nie, 2001). However,
research on religion and mass media continues to be under investigated
while the Internet receives even less attention (Buddenbaum & Stout, 1996;
Hoover, 1997; Hoover & Venturelli, 1996).
Previous research has shown religion to be a potential significant
predictor for television use (Hamilton & Rubin, 1992; Jackson-Beeck &
Sobel, 1980; Tankard & Harris, 1980), magazine use (Al-Olayan & Karande,
2000; Swatos, 1988), newspaper use (Buddenbaum, 1986; Buddenbaum & Stout,
1996; Sobal & Jackson-Beeck, 1981; Westley & Severin, 1964a), and radio use
(Westley & Severin, 1964b). However, to date there has been little
(Armfield & Holbert, in press) research that takes into account the
relationship religion plays on Internet use. This study seeks to further
the understanding of potential influences affecting Internet use and
further gain a definite understanding of what factors lead individuals to
use this new form of communication.
In addition to analyzing the relationship of religion on Internet and
television use, this research also seeks to further understand the role of
secularism (Buddenbaum & Stout, 1996) in the use of mass media.
Secularization theory is identified as a dominant paradigm for the study of
mass media and religion (Buddenbaum, 1986; Buddenbaum & Stout, 1996).
Secularization theory states a strong conservative/religious lifestyle has
been under attack from an increasingly more secularist society, and all
forms of mass media are theorized to reflect this movement toward a vast
secularization of society resulting in a strong negative relationship of
religion and mass media. This argument is similar to that of uses and
gratification theory, which states individuals turn to media to reinforce
pre-existing norms or values while shying away from media outlets that tend
not to reinforce their values.
A secondary analysis of 1997-1998 DDB-Chicago national survey data (N =
6812) obtained from Robert Putnam's social capital Web site
(http://www.bowlingalone.com/data.php3) was performed for this study.
First, the zero order relationship between religiosity, Internet use,
television use, secularism, and civic participation was analyzed via OLS
regression. Second, a latent composite structural equation model was
generated using LISREL 8.50 to assess the true structural relationship
between the above endogenous variables.
Literature Review
Religion and Mass Media
Research in the area of mass media and religion can be divided into three
distinct categories: studies on effect, content, and use (Buddenbaum &
Stout, 1996). Effect studies are conducted with respect to individual level
religious views on media use and the incorporation of religious messages in
the media. Studies of media coverage and the persuasiveness of media to
change attitudes find negative news coverage, such as the PTL scandal, can
reduce support for religion (Abelman, 1991). Additionally, the effects of
media when covering religious crusades found factors such as a persons
religious commitment, motive for attending the crusade, and preexisting
beliefs are more likely to predict a persons actions than the media's
coverage of the actual event itself (Swatos, 1991).
Effects. Religious organizations have long been trying to effect the
consumption of mass media by denouncing the mass media and professing
certain television programs have a direct receding effect on wholesome
religious values. The first of many denouncements of the mass media by
religious institutions began with the Protestant and Catholic Churches
during the 19th century. The churches proclaimed that novels should be
considered with the likes of alcohol and tobacco claiming they had a
diminishing effect on Christian values (Douglas, 1988). More recently,
public charges have been made by religious leaders (e.g. Jerry Falwell) to
boycott television shows (e.g. Teletubbies) and movies that have perceived
inappropriate sexual and violent content (Linn, & Poussaint, 1999). While
some religious leaders criticize mass media and proclaim the mass media has
a diminishing effect on societal values, Jorstad (1993) argues that modern
day evangelists have shaped their communicative messages in an attempt to
form to the specific characteristics of television.
While few studies have looked directly at the effects of mass media on
religious beliefs, studies that do exist focus on the persuasive abilities
of media coverage to alter existing individual level religious beliefs
(Buddenbaum & Stout, 1996). Of the studies that do exist, (Abelman, 1991)
found negative news coverage of religious issues and organizations (i.e.
the PTL scandal) can reduce public support of religion. However, studies
looking at the level of pre-existing religious beliefs are stronger than
any form of media effects (Swatos, 1991). In short, little empirical
evidence exists to support a direct effect of traditional media use to
alter pre-existing religious beliefs or activities.
Content. Various conservative religious leaders and organizations have long
linked society's receding Christian values to the consumption of mass
media. Although these accusations lack empirical evidence, their arguments
seem to focus on the secularist messages mass media broadcasts (see Newman,
1996). Recently, public calls have been made by religious leaders to
boycott television shows and movies that contain "inappropriate" content
{Linn, 1999 #1263}. In short, conservative religious beliefs are found to
be negatively related to the viewing of programs with violent and/or sexual
content (Hamilton & Rubin, 1992).
Recent content analyses of newspapers reveal that coverage of religion is
on the increase (Shepard, 1995). Religious stories are longer, broader in
scope, and more issue oriented when compared to nonreligious stories
(Buddenbaum, 1986). In addition, Hynds (1987) reports the coverage of
religious issues in newspapers is growing in terms of quality and quantity.
In fact, the Dallas Morning News has introduced a new religious section to
expand their coverage of religious issues (Hynds, 1999). While newspapers
continue to expand their content to reach an increasing audience of
religious individuals, television channels, and news organizations are
beginning to cover religion in weekly or daily segments (Shepard, 1995).
This recent focus on religious programming and news coverage is not
surprising when you consider the rapidly growing religious and faith based
programming and networks. The Eternal World Network Channel (EWTN) an
International Catholic Cable Channel, Z Music Television, and various
independent religious based channels are expanding across cable systems and
countries as a "religious boom" seems to be taking place (Katz, 1996). With
the vast number of choices currently in media content, researchers in
communication and sociology are focusing a majority of their studies of
religion and mass media on media use.
Use. Research on the relationship of different elements of religiosity
(e.g. strength of religious belief, denominational affiliation, etc.) and
various forms of traditional media use has revealed several findings. The
findings discussed below are focused on four traditional forms of media
research: television, radio, newspaper, and magazine.
In looking at viewership, Tankard and Harris (1980) found heavy viewers of
television report having a weaker relationship to religious preference and
attended worship service less often when compared to nonviewers of
television. These findings support that of Jackson-Beeck, (1977) and
Jackson-Beeck and Sobel (1980), which show heavy television viewers are
less likely to be involved in outside activities such as professional,
academic, or church affiliated groups. In addition to television use,
research has looked at the impact of religion on the use of specific types
of television content. For example, Roberts (1983) found that members of
the religious moral majority were more likely to watch religious programs
than those who were ideologically liberal or a "cross-section group"
between the moral majority and liberal group.
Conservative religious beliefs are negatively linked to viewing sexually
explicit and violent television shows, Hamilton and Rubin (1992) found
religiosity affects churchgoers' motives for using television and their
choice of programs with sexual and/or violent content. They found partial
support for a mediating relationship between religion and television use.
In short, conservative religious beliefs are negatively related to viewing
of programs that are perceived to be morally offensive, violent, or overtly
sexual in nature.
There has been little work to date on the relationship between religion
and individual-level patterns of radio use. Fathi and Heath (1974) found
individuals who are closely aligned with a religious community (e.g.,
attend church and/or religious school) tend to prefer certain radio station
formats (e.g., classical). More recently, works in this area have looked at
specific sub-populations (e.g., African Americans, or senior citizens) to
measure patterns of religious radio stations. Levin and Taylor (1998) found
that increased severity of medical conditions was not linked to
substituting radio programming for church attendance. Additional research
has looked at the African-American community and religious radio
programming, church attendance, church membership, and religiosity. In
short, the study of radio use and religiosity is wide-ranging in terms of
scope, and individual level religious beliefs.
With respect to newspaper, Sobel and Jackson-Beeck (1981) found readers
were more likely to attend church services than nonreaders. Furthermore,
Finnegan and Viswanath (1988) found that those individuals who are members
of a church are more likely to read a major daily newspaper and a local
weekly newsletter. Finally, not surprisingly readers of church newspapers
are more likely to attend worship services than nonreaders (Stamm & Weis,
1986). In short, church attendance is positively associated with higher
levels of all types of newspaper use.
Finally, a recent survey was conducted on members of the Associated
Church Press and the Evangelical Press Association. These two groups are
focused specifically on the production and distribution of religious based
periodicals and magazines. Waters (2001) found readers of religious
magazines are seen as being quite religious, with a large portion of these
readers being clergy or missionaries. While this research does not reveal
exactly who is reading these publications, it does provide some insight
into what groups of individuals the providers of this material believe are
reading their publications. Swatos (1988) provided a more reliable measure
of religious magazine audiences revealing that a large number of
conservative Christians read religious material. But, fascinatingly these
individuals also read some other type of secular publication. In short,
little is know about the audiences of religious periodicals and magazines
or the magazine usage of religious individuals.
Religion and Internet use.
The relationship of religion and Internet use can be divided into the same
three categories as the studies of traditional media. Recent work in media
effects, content, and use are detailed below.
Effects. Several arguments have been posited about possible effects of the
Internet on individual or societal level attitudes on religious beliefs.
Cobb (1998) argues the Internet/World Wide Web, "offers us a new palette on
which to manifest the life-enhancing values that move us toward an ever
greater richness of experience" (p. 233). Additionally, Zaleski (1997)
speculates cyberspace creates a new environment, which can ultimately lead
to a new spiritualism that will cut across the traditional
religion-secularism continuum. Furthermore, Swatos and Christiano (2000)
state, "the more one becomes aware of more and more religions competing in
a marketplace like setting, the harder it becomes to assert that any one
religion contains all truth and that others must be all wrong" (p. 13).
Finally, the University of Pennsylvania's Annenburg School of Communication
hosted prominent media scholars, media executives, and religious leaders
for a discussion of the potential effects of emerging digital technologies
on various aspects of religion. Conclusions were drawn that the
Internet/World Wide Web may be influential in determining the future of
individual and societal level religious practices (see Shayon & Cox, 1994).
However, all of these assumptions are possible and logical predictions,
this area of research is not yet empirical in nature.
Content. Content-based studies in the area of religion and the Internet
tend to investigate the types of religious information available on the
Web. Specifically, Durusau (1998) provides a guide for Web sites devoted to
the study of the Bible. Other researchers have provided summaries of the
types of religious information disseminated on the Internet (e.g., Baker,
1997; Bunt, 2000; Gold, 1999; Zakar & Kaufmann, 1997). Horsfall (2000) has
focused her research on strategies used by religious organizations via the
message they are sending on their Web sites. Findings show that religious
organizations believe the Web provides an information seeking function for
individuals, and the sites attempt to provide guides that will assist
information seekers to navigate through the volumes of information
available on the Web.
Use. Recently there has been some initial work in relation to religion and
Internet use. Bedell (2000) used several non-random convenience samples
from individuals who visited various religious Web sites and focus groups
along the same line of research. Bedell (2000) found religious people are
using the Web, they believe religious information is plentiful on the Web,
and religious groups and individuals expect a great deal from this
technology. In addition, ethnographic studies (e.g., Bainbridge, 2000;
Howard, 2000) have been undertaken; Bainbridge (2000) reveals that
religious individuals use digital technology for a variety of services
(e.g.. prayer, specific religious information), and Howard (2000) finds
individuals use the Internet for "negotiative rhetorical techniques" (p. 242).
More recent work in the area of religiosity and Internet use (Holbert &
Armfield, 2002) found religious individuals use the Internet less than
individuals reporting a lower individual level of religiosity. However,
they found an interaction exists between secularism and religion. Once
individual level secularism was accounted for in high level regression, the
relationship between Internet use and religiosity was not significant. In
short, it is not religiosity that effects how individuals use the Internet,
it is an individuals secularist beliefs (Holbert & Armfield, 2002).
While these works provide some insight into how individuals, primarily
those who are high in individual level religiosity, use religious Web
sites, we cannot generalize to all individuals with regard to the influence
of individual religious beliefs on Internet use. Furthermore, we can
conclude there is a potential and important relationship between religion
and media use however, that relationship is quite ambiguous at this point
(Stout & Buddenbaum, 1996). What is needed is a large national sample of
data to allow for an assessment of the relationship between religion and
the Internet. This study seeks to fill that void with the use of a large
national data set in order to address the following hypothesis:
H1: The relationship between television use and religion is negative.
Secularization Theory
Overview. The general rationale driving the study of secularism is, "in the
face of scientific rationality, religion's influence on all aspects of life
– from personal habits to social institutions – is in dramatic decline"
(Swatos & Christiano, 2000, p. 6). Specifically, traditional religious
beliefs are increasingly being substituted for standards that are more
secular and a new modern lifestyle that reflects these values. While this
may sound like religion and secularization is on a continuum,
secularization theory is not simply the opposite of traditional religious
beliefs (Martin, 1978). In fact, secularization theory is extremely dense,
diverse, and at times very difficult to conceptualize the true meaning of
secularism (see Martin, 1965; Shiner, 1967). Moreover, Dobbelaere (1999)
recognizes three levels of analysis within secularization that exist:
Societal, organizational, and individual. Given this, this study seeks to
improve the understanding of the relationship between religiosity and mass
media use. This research will focus on the micro level analysis of
secularization with two specific concepts being of particular interest to
this line of research, individualization, and pluralism. Both are strongly
linked to secularization theory and are outlined below.
Individualization. With the "privatization" of religion, an increased sense
of the individual being able to function outside the confines of his/her
traditional teachings has evolved (Luckmann, 1967). The role of
individualism within a larger community, which a church functions has been
deeply debated, Bellah, Madsen, Sullivan, Swidler, and Tipton (1985) argue
the church must function as a means by which an individual is allowed to
express himself or herself. This is due largely to the fact that religious
organizations are recognizing they function within a secularist environment
that values individual freedoms. Moreover, Bellah et al. (1985) argue the
church, and the community that it represents, must be chosen by the
individual and that choice is the most important concept within
individualization (see Beyer, 1990). In short, individualization means
individuals have more freedom of choice in relation to how they choose to
live their private lives. With this, greater individualization brings an
understanding that others should be given this identical freedom.
Pluralism. Another concept associated with secularization is pluralism or
individual-level knowledge that many religious beliefs and faiths exist in
the world. With this understanding comes the awareness that no one religion
can truly claim to have the only pure understanding of God. In a secular
world, one where many faiths exist and function along side each other,
members of other religions are going to learn about other religious
traditions over time. As Swatos and Cristiano (2000) state, "pluralism
clearly creates a market-place of ideas where absolute claims for ultimacy
are always at some degree of risk" (p. 14). As such, pluralism works in
conjunction with individualization to create a secular environment that
allows individuals to freely exchange ideas and beliefs about their faith
with people of other faiths. People are given the freedom to base their
personal lifestyle choices not only on the religious teaching of their
church, but on the faith that best suits their individual worldview. Thus,
those individuals can adhere to a more diverse or pluralistic environment,
which affords more choice in constructing a faith of their own, or a faith
that best suites their personal lives.
Secularism and Mass Media Use. Buddenbaum and Stout (1996) provide an in
depth overview and provocative argument in relation to religion and mass
media use as outlined by secularization theory. They state, "Secularization
theorists expect the relationship to be negative because mass media is
considered to be part of modern, secular society" (p. 20). Furthermore,
secularization theory not only argues that individual-level religiosity
would be negatively related to all forms of mass media use, but stronger
secularist attitudes should be positively associated with all form of mass
media use. Based upon the basic assumptions underlined by secularization
theory, its relationship to religiosity, and media use, the following two
hypotheses are posited:
H2: Religion is negatively related to secularist attitudes.
H3: Secularist attitudes are positively related to Internet use.
Civic Participation
Additional research performed on religion and community integration (e.g.
Janowitz, 1952; Merton, 1950) points to a link with those individuals who
are more religious and attend church are more integrated and involved in
their local communities (Stout & Buddenbaum, 1996). Westley and Severin
(1964b) found support for religious individuals being more sociable and
involved in their community. In addition, those individuals who use mass
media less report higher levels of religiosity and involvement in social
clubs or community activities (Tankard & Harris, 1980). In short, those who
attend church are more integrated into the community than those who do not.
However, media use has been found to foster civic participation (Stamm &
Weis, 1986; Stamm, 1988) as those who subscribe to and use media outlets
are more likely to integrate into and involve themselves into the
community. Based on this research, this study will seek to understand the
relationship of religiosity and civic participation with the following
exploratory hypotheses:
H4: Religiosity is positively related to civic participation.
H5: Internet use is positively related to civic participation.
H6: Television use is negatively related to civic participation.
In addition, this research strives to understand the primary relationship
of civic participation and secularism.
RQ1: What is the relationship between secularism and civic participation?
Methods
Survey Data
The data used for this study was collected by Market Facts and funded by
DDB-Chicago, an international marketing communication company, and made
available by Robert Putnam via his Web site
(http://www.bowlingalone.com/data.php3). In an effort to achieve a balanced
sample, a beginning sample of approximately 5,000 panelists is collected by
Market Facts each year and adjusted within subcategories of race, gender,
and marital status to compensate for expected differences in return rates.
This data has just recently been made available for academic use (see
Armfield & Holbert, in press; Shah, McLeod, & Yoon, 2001), and has proven
to be a worthy sample of the American public (see Shah, McLeod, & Yoon,
2001).
Although the DDB-Chicago data includes the years 1975 through 1998, this
study only utilizes data collected for the two most recent years 1997 and
1998. The final samples collected for these years are as follows: 1997, N =
3,462, response rate = 69.2%; 1998, N = 3,350, response rate = 67.0%. All
question wordings for the items used in this study are exact across the two
years, and all variables were standardized and combined to create a single
data set for the two years (N = 6,812).
Measures
First, a criterion variable for Internet use was created from an additive
index consisting of three dichotomous measures. Respondents were asked each
of the following questions, with possible responses being (Yes) or (No):
Have you yourself used the Internet/World Wide Web in the last 12 months to
participate in a chat room or online forum?; Have you yourself used the
Internet/World Wide Web in the last 12 months to explore an interest or
hobby?; Have you yourself used the Internet/World Wide Web in the last 12
months to send electronic mail? This index proved itself reliable
(Chronbach's = .76).
Second, a criterion variable for television use was created from an
additive index consisting of two measures. Respondents reported the hours
of weekday television use and hours of weekend television use. This index
proved itself reliable (Chronbach's = .74).
Religiosity is a two-item additive index consisting of a behavioral and an
attitudinal component. Respondents were asked how many times they had
"attended church or other place of worship" in the last 12 months, with
possible responses ranging from (None) to (52+ times). The attitudinal
component consists of responses to the following statement, "Religion is an
important part of my life." Responses ranged from (definitely disagree) to
(definitely agree). This index proved itself to be reliable (Chronbach's =
.81).
Secularism is a nine-item additive index operationalized to reflect the
three broad areas of (a) individual-level perceptions of the world becoming
more secular, (b) attitudes about the level of individual choice that
should be afforded to people relative to the various stands taken by the
church, and (c) a proclivity for pluralistic environments. Those surveyed
were asked to respond to each of the following statements, with responses
ranging from (definitely disagree) to (definitely agree); Everything is
changing too fast today (reverse); I have somewhat old fashioned values
(reverse); Couples should live together before getting married; I am in
favor of legalizing same sex marriages; I am in favor of legalizing
doctor-assisted suicide; I think the women's liberation movement is a good
thing; I am interested in the cultures of other countries; I like to visit
places that are totally different from my home; I would rather live in or
near a big city than in a small town. The combination of these nine items
represent an individual's general perceptions about a more secular world
and their place within a changing environment, their beliefs about
individuals making their own lifestyle choices relative to organized
religions dogma, and a desire to be in diverse locations. These items form
a reliable index (Chronbach's = .64).
Civic participation is an additive index consisting of three-items.
Respondents were asked how many times they had: Attended "a club meeting,"
"worked on a community project," or "did volunteer work." Possible
responses ranged from (None) to (52+ times) for all three items. This item
proved itself reliable (Chronbach's = .70).
A series of single-item control variables were also included in this study:
Age, income, sex, education, and political ideology. Each of these
exogenous variables is though to have a potential affect on religion and/or
Internet use. Age was measured in exact years. Income was measured along a
continuum from (under $10,000) to ($100,000 or more). The education item
asked for the highest level of schooling completed by the respondent, with
possible responses ranging from (elementary school) to (post-graduate
college education). Ideology was a single-item measure in response to the
following; "Generally speaking, would you consider yourself to be…," with
possible responses ranging from (very conservative) to (very liberal).
Results
The statistical theory underlying structural equation modeling is grounded
in covariance structural analysis (Holbert & Stephenson, 2002) and a
covariance matrice is the preferred technique for use with LISREL. The
model (figure 1) data was input into LISREL 8.50 and a latent composite
model was created.
Insert figure one about here
Each hypothesized path was left open within the model. Based upon previous
research (Holbert & Armfield, 2002) the path between religion and Internet
use was closed. In addition, paths were opened for all exogenous variables
thought to be of significant findings based upon previous theoretical
research (see Table 1).
Insert table one about here
Results
In the original analysis of the data, an interaction between Religion and
Secularism with relation to Internet use and television use was observed
(see Holbert & Armfield, 2002). This is not surprising given the similarity
that preliminary uses and gratifications research on digital communication
(Angleman, 2000; Ruggiero, 2000) has shown. The path from religion and
secularism was then converged to reflect the interaction before predicting
the path between Internet use and television. The model was found to fit
the data well. Although the 2 (13, N=6812) = 29.4 was significant, previous
research has reported the 2 measure is problematic (Holbert & Stephenson,
2002). Solid fit statistics are found by examining the following:
Comparative fit index (CFI) of 1.00, an incremental fit index (IFI) of
1.00, a root mean square error of approximation (RMSEA) of .015, and a
Standardized RMR of .008.
Exogenous variable paths are reported in Table 2. As anticipated age,
income, sex, education level, and ideology are all significant predictors
at many levels of religion, secularism, mass media use, and civic
participation. All paths reported are significant and pruned of
non-significant paths.
Insert table 2 about here
Religiosity and Television. Hypothesis one predicted the relationship
between television use and religion would be negative. This hypothesis was
supported ( = -.33, p < .05). There was a significant difference between
religiosity of an individual and television use. Specifically, people high
in religiosity are less likely to watch television, but the interaction is
not significant.
Secularism. Hypothesis two predicted the relationship between religion and
secularism would be negative. This hypothesis was supported ( = -.36, p <
.01). Individuals who reported higher levels of religiosity have a lower
secularistic view toward society. Second, hypothesis three predicted the
relationship between secularism and Internet use would be positive. This
hypothesis was also supported ( = .06, p < .01). Individuals who have more
secular views toward society are using the Internet more.
Civic Participation. Hypothesis four predicted religiosity is positively
related to civic participation. This hypothesis was supported ( = .10, p <
.01). Individuals who report higher levels of religiosity are more involved
and participate more often within their community. Hypothesis five
predicted Internet use would be positively related to civic participation.
This hypothesis was also supported ( = .08, p < .01). Individuals who use
the Internet more often are also involved to higher levels within their
community. Hypothesis six predicted television use would be negatively
related to civic participation. This hypothesis was also supported ( =
-.00, p < .01). Finally, the research question was looking to explore the
relationship between secularism and civic participation. This path was
significant ( = .06, p < .01). A positive relationship exists for
individuals with high levels of secularistic views to be more involved
within their community.
Interaction. Post hoc analysis revealed that an interaction between
individual level religiosity and secularistic views existed. This finding
replicates in part previous findings by Holbert and Armfield (2002) that
revealed when accounting for secularism at the individual level, religious
individuals used the Internet less ( = -.02, p < .01) and the television
more ( = .15, p < .05). What this reveals is that individuals who have a
high secularistic view for society tend to use the Internet more then
individuals with low secularistic beliefs. Holbert and Armfield (2002)
found that after accounting for secularism at the individual level, the
negative relationship between religiosity and Internet use was longer
significant. In short, the more religious an individual is, the less
secular his/her attitudes are and the more secular an individual is, the
more likely they are to use the Internet (Holbert & Armfield, 2002).
Discussion
The five exogenous variables included in this research: Age, income, sex,
education, and ideology had multiple significant relationships with the
endogenous variables. A brief summary of Table two follows. The older an
individual is the more religious and less secular they are. Older
individuals use the Internet less and watch television more. As for civic
participation, older individuals are very slightly more likely to be more
active in their community. Lower income individuals were more likely to be
religious while higher income individuals held more secular views towards
society. In addition, higher income was also related to increased Internet
use and decreased television use, while higher income individuals were more
likely to be active in their community. This should not be surprising given
the current debate surrounding the digital divide.
Women were found to be both more religious and more secular than men,
furthering the idea that religiosity and secularism are not on the same
continuum. The findings in relation to sex and Internet use replicated
previous findings by Jackson, Ervin, Gardner, & Schmitt (2001) who found
men spend more time online than women. Lower educational levels were linked
to higher television use, while higher education levels were positively
related to individual level religiosity, secularism, Internet use, and
civic participation. Finally, not surprisingly those individuals with a
more liberal ideology were found to be less religious and more secular in
their world views.
There are several general patterns that emerge from this research with
regard to religion, secularization, mass media use, and civic
participation. Religious individuals were positively related to civic
participation. In short, religious individuals are more involved,
integrated, and active in their community than secularistic individuals.
These findings support previous research of Tankard and Harris (1980) as
well as Westley and Severin (1964b). Furthermore, secularism seems to drive
the mass media use of religious individuals who were found to watch less
television. As previous research (e.g. Stamm & Weis, 1986; Stamm, 1988) has
supported, religious individuals use the Internet as a tool to integrate
themselves into their local community and participate in their community.
Focusing on the findings in relation to mass media, previous research has
shown religious individuals watch less television (Hamilton & Rubin, 1992).
While this research confirmed this finding, a previous finding of (Holbert
& Armfield, 2002) was also confirmed in that religion does not alone
account for a negative level of Internet use among religious individuals.
In fact, individuals who ranked high in religiosity tend to use all forms
of mass media less furthering Stout and Buddenbaum (1996) argument that
mass media use among religious individuals is currently being driven by
secularism theory. In fact, secularism accounts for many of the reasons why
individuals are using the Internet. The Internet can provide a vast link to
a more secular world, a world that many religious individuals may choose
not to seek out, explore, or gain additional knowledge. Secularization
theory posits that religion is negatively related to all forms of mass
media use with the understanding that mass media typically reflects a
anti-religious and increasingly secularist view of society (Swatos &
Christiano, 2000). Mass media use and Internet use in particular, is
currently driven by this model, as discussed by (Buddenbaum & Stout, 1996).
While this does not mean that religious individuals are not using the
Internet, it does support the conclusion that they are less likely than
more secular individuals to use the Internet.
The relationship between religiosity, secularism, television, and Internet
use may alter over time as the Internet emerges as a more mainstream
communication technology. For this reason, research in this realm should
continue to strive for a broader understanding of how individual level
religiosity and secularist attitudes affect the use of all forms of mass
media over time. Future research needs to look at specific uses and
gratification sought and obtained from religious individuals who use this
new technology.
While this research has made important steps toward understanding the
relationship between mass media, religion, and secularization, it is
important to point out that a stronger measure of religiosity should be
used in future research. The scale used in this secondary analysis gets
only to the attitudinal level of religiosity and pales in comparison to
other religiosity scales (see Faulkner & DeJong, 1966).
In addition, the scale for secularization is not a complete representation
of the theory of secularization. Secularization theory is multi-level and
to get at the complete function of the concept within mass communication
and society a more representative scale should be developed. This specific
avenue of research in relation to secularization theory needs to be
furthered. The first step of this process should include a full explication
of the theory (e.g., Chaffee, 1991). However, this is not a simple task
given the volumes of literature written on this subject. Nonetheless, this
task needs to be accomplished to further the line of research concerning
mass media, religiosity, and secularization.
While these two limitations have been recognized, it is important not to
undermine the significance of this research and the findings in relation to
secularization theory driving how religious individuals use mass media.
While empirical research in the area of civic participation has shown a
community integration approach to drive the level of Internet use with
regard to civic-minded individuals (Shah, Kwak, & Holbert, 2001;Shah,
McLeod, & Yoon, 2001) this does not seem to be the case for religious
individuals. Bimber (1999) found many individuals use the Internet to
interact with individuals who are outside of their immediate community and
it is very much the possibility religious individuals are doing the same.
Future research should remain focused on the study of these relationships
in order to gain a broader understanding of how individual-level
religiosity and secularism affects mass media use across time. One
direction this research can take is to gain a deeper understanding of
Internet use by linking uses and gratification theory (e.g., Palmgreen,
1984). Focusing on uses and gratifications theory will allow researchers to
explore the motives and elements of a persons religiosity that leads them
to seek out or shy away from mass media. Finally, as discussed earlier, it
is important studies of this kind begin to measure secularism as a variable
that is distinct from religiosity not simply embodied within the more
established attitudinal and behavioral religion items.
Zakar, S. M., & Kaufmann, D. (1998). Judaism on-line: Confronting
spirituality on the Internet. New York: Aronson.
Zaleski, J. (1997). The soul of cyberspace: How new technology is changing
our spiritual lives. San Francisco: HarperEdge.
Figure 1. Hypothesized Paths
Internet
Use
Religion
Civic
Participation
Interaction
Television Use
Secularism
Table 1
Subscale
1
2
3
4
5
6
7
8
9
10
11
Participants (N = 6,812)
1. Age
—
2. Income
-.036**
—
3. Sex
-.008
-.094***
—
4. Education
-.074***
.365***
-.022
—
5. Ideology
-.092***
.060***
.029*
.082***
—
6. Religiosity
.138***
-.039**
.180***
.069***
-.247***
—
7. Secularism
-.193***
.175***
.021
.226***
.443***
-.393***
—
8. Interaction
.024
-.016
-.025
-.063***
-.045***
.048***
-.058***
—
9. Int. Use
-.257***
.223***
-.070***
.261***
.098***
-.089***
.208***
-.077***
—
10. TV Use
.035**
-.223***
.001
-.217***
.014
-.097***
-.033*
.068***
.144***
—
11. Civic
Participation
.167***
.112***
.095***
.222***
-.001
.287***
-.011
-.011
.084***
-.122***
—
*p < .05. **p < .01. ***p < .001
Table 2
Exogenous Variables on Endogenous Variables
Religion
Secularism
Interaction
Internet
Use
Television
Use
Civic
Participation
Participants (N = 6,812)
Age
.01***
-.01***
-.00***
.01**
.00***
Income
-.01***
.02**
.01***
-.10**
.00***
Sex
.35*
.16*
-.04**
Education
.09**
.13**
-.05**
.04***
-.28*
.02**
Ideology
-.24**
.26**
-.04**
*p < .05. **p < .01. ***p < .001
Figure 2. Significant Paths
Internet
Use
Religion
.10 -.02
.08
Civic
Participation
Interaction
.06
.36
-.33
Television
Use
.15
.06
-.00
Secularism
.15
All paths shown are significant at the p < .05 level
|
