.3:
I . ‘
. Jmu

x
hr...

.
.3 f .3
«3W; 3.5.
5W:
. ..
uni}:
‘ i
. 93:33.57:
J... ﬂat): :0...
‘33.,ilr‘5?‘ .-
. {tyuri‘I‘ss

“ﬁg?

o. .
. .9 e. u.
. 31.14.!!!
{61951

i
.{......9. 1.:
. . 5‘

 

: $3?

1.1.1:}
J. y t. :

||'

 

"Heme

202%

LIBRARY
Michigan State
University

This is to certify that the
dissertation entitled

ONLINE POSTING ANXIETY AND ITS INFLUENCE ON
BLOGGING:
COMPARING THE U.S. AND CHINA

presented by
Xun Liu

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Telecommunication, Information
Studies and Media

 

 

ﬂag/03M

Major'Proiessor’s Signature

3/) f/g)?

Date

MSU Is an afﬂnnative-actlon, equal-opportunity employer

:—-—_;-.-.-._._

-—-—o-Q_n-n-.—.--_..

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE . DATE DUE

 

up; 3197 25199

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProi/Acc&Pres/ClRC/DateDue.indd

ONLINE POSTING ANXIETY AND ITS INFLUENCE ON BLOGGING:
COMPARING THE U.S. AND CHINA

By

Xun Liu

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Department of Telecommunication, Information Studies and Media

2008

ABSTRACT

ONLINE POSTING ANXIETY AND ITS INFLUENCE ON BLOGGING:
COMPARING THE U.S. AND CHINA

By

Xun Liu

Through a social cognitive theoretical orientation, this study aims to identify and
illustrate processes and elements central to blog maintenance intentions. A structural
model has been proposed to explain American and Chinese bloggers’ blog maintenance
intentions. The structural model is based on the assumptions that cognitive factors
(blogging self-efﬁcacy, positive outcome expectations for blogging, and negative
outcome expectations for blogging), online posting anxiety, and general anxiety
tendencies (writing anxiety and social anxiety) affect one another and therefore help in
predicting blog maintenance intentions. The original contribution of this dissertation
research is in the application of social cognitive explanations of anxiety to the realm of
computer-mediated communication.

A combination of qualitative and quantitative research methodology was
employed to test the structural model. Three studies were conducted. In study 1, in-depth
interviews were used to examine the nature of blogging among American and Chinese
bloggers. In particular, four new concepts are introduced - a) blogging self-efﬁcacy, b)
positive outcome expectations for blogging, c) negative outcome expectations for
blogging and d) online posting anxiety. Ten American bloggers and 16 Chinese bloggers

were interviewed online about their beliefs pertaining to these four concepts.

Study 2 applied a pre-test and post-test survey design to investigate the
reliabilities and validities of the newly developed measures. A snowball sampling method
was used to recruit bloggers in America and China. Eighty-seven American bloggers and
70 Chinese bloggers participated in the pre-test online survey. Among these, 58
American and 53 Chinese bloggers completed the post-test online survey after two
weeks. A psychometric analysis was also conducted. Two reliability measures namely the
Cronbach’s coefﬁcient a and test—retest reliability were calculated. The content validity,
criterion-related validity, and construct validity were examined. The results showed that
these new scales demonstrated acceptable reliability and validity measures.

The purpose of the main study was to test the proposed 16 hypotheses. One
hundred and forty-ﬁve American and 178 Chinese bloggers were recruited through a
purposive sampling method. Structural equation modeling technique was employed to
test the proposed model and four alternative models. The model with blog maintenance

intentions as the dependent variable showed the best ﬁt.

In summary, the structural model demonstrated the process through which various
social cognitive variables contribute to online posting anxiety and blog maintenance

intentions in two cultures.

Copyright by
Xun Liu

2008

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Robert LaRose for his guidance, support,
patience, and insight throughout my time at Michigan State University. Dr. LaRose has
been a mentor, guiding me through my deveIOpment as a researcher, teacher, and

valuable member of the academic community.

I also would like to thank my committee members, Dr. Mark Levy, Dr. Nicole
Ellison, and Dr. Hairong Li. Mark’s early inﬂuence and nurturing during my ﬁrst years in
the program kept me focused in the development of my research agenda. Without Mark
and his wife Diane, I would not even have taken the journal to U.S. to pursuit my Ph.D.
Dr. Hairong Li has helped me to view media use theories from a different perspective.
His insight regarding this work has been extremely valuable. Nicole provided me with

intellectual stimulation, friendship, as well as a beautiful house for my summer writing.

Thanks also go to Dr. Bella Mody. Her guidance and support to my work in the

years I’ve known her has helped me more than I could possibly relate.

I am grateful to the 506 American and Chinese bloggers who participated in this

dissertation research project. This dissertation could not have been completed without

them.

I also wish to acknowledge my friends throughout my studies at Michigan State
University. Special thanks to Lei J ia and her parents for providing me with great help and

support.

I could never thank enough my family. I would not have reached this point
without the intellectual, emotional, and ﬁnancial support of my parents Dongrong and
Rongxiang. I am also deeply appreciative of my husband Joy during this process for

giving me help, support, and love.

vi

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. ix
LIST OF FIGURES ................................................................................. x
CHAPTER 1
The Focus of the Study ...................................................................... 1
Research Approaches ........................................................................ 2
Research Outline ............................................................................. 4
CHAPTER 2
SOCIAL COGNITIVE THEORY .................................................................. 5
Bandura's Social Cognitive Theory Overview ........................................... 5
Main Constructs: Self-efﬁcacy and Outcome Expectations ........................... 5
A Social Cognitive View of Anxiety ...................................................... 7
CHAPTER 3
LITERATURE REVIEW ............................................................................ 9
Blogging ........................................................................................ 9
Blog Deﬁnition and Typologies .................................................... 9
Blogging Motivations .............................................................. 10
Personal Media with Public Attention .......................................... 11
Blogging in Other Cultures ....................................................... 14
Anxiety— an Introduction ................................................................. 15
Anxiety Deﬁnitions ................................................................ 15
Anxiety Typologies ................................................................ l6
Cognitive Perspectives of Anxiety ........................................................ l9
Integrationist Perspectives of Anxiety .................................................... 20
Anxiety in Different Cultures2l
Social Anxiety ............................................................................... 23
Communication Anxiety ................................................................... 25
Writing Anxiety ............................................................................. 26
Computer anxiety ............................................................................ 27
A Social Cognitive Critique of Communication Apprehension, Writing
Apprehension and Computer Anxiety... ................29
Social Anxiety, Communication Anxiety, Writing Anxiety, and New
Media ......................................................................................... 31

vii

Online Posting Anxiety .................................................................... 43

Culture ........................................................................................ 46
CHAPTER 4
HYPOTHESES AND RESEARCH QUESTIONS ............................................ 53
The Role of Culture ....................................................................... 58
The Impact of General Anxiety Variables ............................................... 59
The Impact of Self-efﬁcacy and Outcome Expectations ............................. 60
The Impact of Online Posting Anxiety .................................................. 62
CHAPTER 5
METHODS .......................................................................................... 65
Study 1 ....................................................................................... 65
Method .............................................................................. 65
Brief Results ........................................................................ 66
Study 2 ....................................................................................... 68
Method .............................................................................. 68
Brief Results ........................................................................ 69
Main Study ................................................................................... 70
Sampling Method .................................................................. 70
Evaluations of the Reliability and Validity of the New Scales .............. 71
Structural Equation Modeling Test ....................................................... 74
Operational Measures ............................................................. 74
Analysis ............................................................................. 76
CHAPT ER 6
RESULTS ............................................................................................ 77
CHAPTER 7
DISCUSSION ........................................................................................ 91
The Role of General Anxiety Variables .................................................. 92
The Role of Blogging Self-efﬁcacy and Outcome Expectations for Blogging. ...94
Online Posting Anxiety ................................................................... 103
Social Cognitive Deﬁnition of Online Posting Anxiety .................... 103
Online Posting Anxiety in Lasswell’s Transmission Model of
Communication ................................................................... 104
Online Posting Anxiety as a Process .......................................... 108
Online Posting Anxiety as a Unique Concept ................................. 109
Blog Maintenance Intentions as the Dependant Variable .......................... 111
Cultural Inﬂuences ......................................................................... 114
Implications ............................................................................... 1 17
Future Study Areas ....................................................................... 121
Limitations ................................................................................. 123
REFERENCES .................................................................................... 128

viii

LIST OF TABLES

Table 5.1. American and Chinese Bloggers’ Blogging Beliefs .............................. 67
Table 5.2. Cronbach’s Coefﬁcient Alpha of the New Measures ............................. 69
Table 5.3. Test-Retest Reliability of the New Measures ...................................... 70
Table 5.4. Blogging Self-Efﬁcacy Factor Loadings ........................................... 72
Table 5.5. Positive Outcome Expectations for Blogging Item Loadings ................... 73
Table 5.6. Negative Outcome Expectations for Blogging Item Loadings .................. 73
Table 5.7. Factor Loadings for Online Posting Anxiety Items ............................... 73
Table 6.3. Gender’s Correlations with Other Variables in both Samples ................... 80
Table 6.4. Fit Indexes of Structural Models .................................................... 83
Table 6.1. Means, Standard Deviations, Reliability Estimates, and Intercorrelations for
Study Variables for the American Sample .......................................... 84
Table 6.2. Means, Standard Deviations, Reliability Estimates, and Intercorrelations for
Study Variables for the Chinese Sample ............................................. 85
Table 7.1. VIP and Tolerance Value of Factors ................................................ 97

ix

LIST OF FIGURES

Figure 4.1. The Causal Relationship between Cognition and Anxiety ....................... 54

Figure 4.2. The Causal Relationship between General Cognition and Speciﬁc
Cognition ............................................................................... 55

Figure 4.3. The Causal Relationship between General Anxiety and Speciﬁc Anxiety. ...56

Figure 4.4. The Causal Relationship between General Anxiety and Speciﬁc Cognition..56

Figure 4.5. The Causal Relationship Examined in This Study ................................ 57
Figure 4.6. Theoretical Model of Direct and Indirect Effects of Blog-Maintenance
Intentions ................................................................................ 64
Figure 6.1. Model 6.1. Blog-Maintenance Intentions as the Dependent Variable .......... 86
Figure 6.2. Model 6.2. Social Anxiety and Writing Anxiety Directly Inﬂuence Online
Posting Anxiety ........................................................................ 87
Figure 6.3. Model 6.3. Online Posting Anxiety Precedes Cognition ........................ 88
Figure 6.4. Model 6.4. Online Posting Anxiety Precedes Outcome Expectations ......8.9
Figure 6.5. Model 6.5. Model without Online Posting Anxiety .............................. 9O

CHAPTER 1
INTRODUCTION

1.1 The Focus of the Study

Blogs are arguably one of the most signiﬁcant media revolutions since the arrival
of television (Sullivan, 2002). More than 80,000 blogs are created each day, with a new
blog being created about every second (Perlmutter & McDaniel, 2005). In the ﬁeld of
communication, much of the scholarly research about blogs has focused on online
journalism (Gillmor, 2003; Johnson & Kaye, 2004; Lasica, 2002, 2003; Matheson, 2004;
Singer, 2005; Susan, 2006; Ziomek, 2005), political communication (Kavanaugh et al.,
2006; Keren, 2004; Lawson-Borders & Kirk, 2005; Park, 2003a; Poor, 2005; Schafer,
2006; Singer, 2005), and blogging motivations (Brady, 2006; Kaye, 2005, 2007; Li, 2005;
Nardi et al., 2004a; Nardi et al., 2004b; Trevino, 2005). Qualitative and quantitative
research methods have been used to examine blogging including content analysis
(Herring et al., 2006; Herring et al., 2004b; Papacharissi, 2007; Trammell &
Keshelashvili, 2005), rhetorical analysis (Herring et al., 2004a; Miller & Shepherd, 2004),
network analysis (Herring et al., 2005; Marlow, 2006b), interviews (Milne, 2004; Nardi
et al., 2004b), and surveys (Marlow, 2006a; Viegas, 2005).

This dissertation focuses on American and Chinese bloggers’ blog-maintenance

intentions. Perseus’s report found that two-thirds of blogs have not been updated in more

than two months. The report concluded that blogs, much like icebergs, are composed of a
vast bulk of unseen material, in this case abandoned blogs (Perseus, 2003). Chinese blogs
also have iceberg characteristics. According to the most recent CCNIC (China Internet
Network Information Center) report, there were 17.49 million bloggers and 33 million
blogs in China. More than 70% of blogs were “sleeping blogs”—blogs that had not been
updated for more than one month (CNNIC, 2006). These statistics may imply that two-
thirds of American and Chinese bloggers had low blog-maintenance intentions.

This dissertation investigates what inﬂuences the blog-maintenance intentions of
American and Chinese bloggers. Speciﬁcally, the objects of this research are four-fold.
First, evidence is established for the existence of online posting anxiety in bloggers.
Second, measures of variables such as blogging self-efﬁcacy, positive outcome
expectations for blogging, negative outcome expectations for blogging, and online
posting anxiety are developed and tested. Third, structural models are built to examine
the factors that contribute to blog-maintenance intentions. Fourth, cross-cultural
comparisons are made between American bloggers and Chinese bloggers. The original
contribution of this dissertation is that it extends the social cognitive theoretical
interpretation of communication anxieties to the computer-mediated communication
realm.

1.2 Research Approaches

This current study proposes a structural model, based on social cognitive theory,
the literature of anxiety, and blogs, to explain American and Chinese bloggers’ blog-
maintenance intentions. The structural model is based on the assumptions that cognitive

factors (blogging self—efﬁcacy, positive outcome expectations for blogging, and negative

outcome expectations for blogging), blogging anxiety (online posting anxiety), and
general anxiety tendencies (writing anxiety and social anxiety) will affect one another
and, in turn, will predict blog-maintenance intentions. A combination of qualitative and
quantitative research methodology was employed to test the structural models. Three
studies were conducted.

Study 1 used in-depth interviews to examine the nature of blogging and American
and Chinese bloggers’ beliefs about blogging self-efﬁcacy, positive outcome expectations
for blogging, negative outcome expectations for blogging, and online posting anxiety.
The goal of Study 1 was to generate measures for the four new concepts for later studies.
Ten American bloggers and 16 Chinese bloggers were interviewed online.

Study 2 applied a pretest and posttest survey design to investigate the reliabilities
and validities of the newly developed measures. A snowball sampling method was used
to recruit bloggers in America and in China. Eighty-seven American bloggers and 70
Chinese bloggers participated in the pretest online survey. Fifty—eight American bloggers
and 53 Chinese bloggers completed the posttest online survey after two weeks. A
psychometric analysis was conducted, and the new measures demonstrated acceptable
reliability and validity.

The purpose of Main Study was to test the proposed structural model described
above. One hundred forty-ﬁve American bloggers and 178 Chinese bloggers were
recruited with a purposive sampling method. The structural equation modeling technique
was employed to test the proposed model and four alternative models. The original
proposed model with blog-maintenance intentions as the dependent variable showed the

best ﬁt.

1.3 Research Outline

This dissertation is structured as follows. Chapter 2 introduces social cognitive
theory, the theoretical framework for this research, including a review of Bandura’s
social cognitive theory. The theory’s two key concepts are explored: self-efﬁcacy and
outcome expectations. Also, social cognitive perspectives of anxiety are discussed.
Chapter 3 reviews signiﬁcant prior research that is relevant to this study. Theories and
studies about blogging, general anxiety, ﬁve types of speciﬁc anxiety (social anxiety,
communication apprehension, writing apprehension, computer anxiety, and computer-
mediated communication apprehension), and individualism-collectivism are reviewed.
This literature serves as the bedrock of this dissertation. Chapter 4 presents the
hypotheses and the structural model. Seven variables (writing anxiety, social anxiety,
blogging self-efﬁcacy, positive outcome expectations for blogging, negative outcome
expectations for blogging, online posting anxiety, blog-maintenance intentions), one
research question, and 16 hypotheses are presented. Chapter 5 brieﬂy describes the
methods of Study 1, Study 2, and the Main Study. Chapter 6 presents the results of the
structural equation modeling testing. Chapter 7 provides a summary discussion, report of

practical and theoretical implications, research limitations, and areas for future studies.

CHAPTER 2

SOCIAL COGNITIVE THEORY

2.1 Bandura’s Social Cognitive Theory Overview

Social cognitive theory provides a comprehensive theoretical framework for
understanding human behavior, environment, and social interaction (Bandura, 1986,
1989). Through symbolizing, vicarious experience, forethought, self-regulatory, and self-
reﬂective capabilities, people can give meaning to their experiences, think through the
consequences of a behavior without actually performing the behavior, learn from
observation of others, anticipate outcomes of behaviors, and evaluate the anticipated

outcomes (Bandura, 1986, 1989).

2.2 Main Constructs: Self-Efficacy and Outcome Expectations

Self-efﬁcacy and outcome expectations are the two most important variables in
social cognitive theory. People’s judgments of personal efﬁcacy are the most central and
pervasive self-reference thoughts that inﬂuence human motivation, affect, and action
(Bandura, 1997). Self-efﬁcacy belief is “a person’s judgment of his/her capability to
organize and execute the actions required to attain designated types of performances”
(Bandura, 1986, p. 391). It is a multidimensional construct that varies in strength,
generality, and level (or difﬁculty) (Bandura, 1986). Perceptions of efﬁcacy are

inﬂuenced by four sources of information: performance accomplishments, vicarious

experiences, verbal persuasive messages, and physiological signals (Bandura, 1986,

1997).

Outcome expectations regulate behavior and inﬂuence self-perception. People can
have both positive and negative expectations of the perceived consequences of a behavior
(Bandura, 1997). Positive outcome expectations help to build self-esteem, self-
satisfaction, and pride, while negative outcomes produce self-devaluation (Bandura,
1997). Courses of action that are likely to produce positive outcomes are generally
adopted and used, while those that bring unrewarding or punishing outcomes are
generally discarded (Bandura, 1999). Nevertheless, knowing what outcome will result
from a given course of action is unlikely to spur an action if people doubt they can
actually accomplish the action. Self-efﬁcacy has an impact on outcome expectations.
People’s thoughts about their self-efﬁcacy inﬂuence how well they perform and thus the
outcomes they are likely to experience. Together, self-efﬁcacy beliefs and outcome
expectations inﬂuence people’s thoughts, emotions, motivations, and actions, leading to

speciﬁc behavioral choices and diverse outcomes.

Social cognitive theory has been proven to be a very efﬁcient theory to examine
effects and usage of computers and the Internet (Agarwal & Karahanna, 2000; Bandura,
2001; Bassam, 2006; Brown et al., 2004; Compeau & Higgins, 1995; Compeau et al.,
1999; Dumdell & Haag, 2002; Eastin, 2005; Fagan et al., 2003; Hasan, 2003; Igbaria &
Iivari, 1995; Johnson, 2005; LaRose & Eastin, 2004; LaRose et al., 2001; LaRose et al.,
2005; LaRose et al., 2003; Sam et al., 2005; Thatcher & Perrewe, 2002). General and

speciﬁc computer and Internet self-efﬁcacy and outcome expectations measures have

been developed. Although Chen et a1. promoted a general computer self-efﬁcacy scale
(Chen et al., 2001), the work of Bandura and most other studies supported the notion that
a speciﬁc self-efﬁcacy measure would have better predictive value in a particular context

(Agarwal et al., 2000; Bandura, 1986; Marakas et al., 1998).

2.3 A Social Cognitive View of Anxiety

The relationship between cognition and emotion has long been the subject of a
scholarly debate. At one extreme it is claimed that emotion inﬂuences cognition (Izard,
1993); at the other, that cognitive appraisals are invariably necessary for the production
of emotion (Bandura, 1986; Lazarous, 1991, 1995). Mathews and MacLeod supported the
view that cognition inﬂuences emotion after they reviewed previous research (Mathews
& MacLeod, 1994). The idea that emotion is embedded in and inﬂuenced by social
contexts is also held by researchers (Averill, 1980; Bandura, 1986; Jakobs et al., 1997;
Lazarous, 1991, 1995; Manstead, 1991).

Social cognitive theory suggests that cognition and social environments inﬂuence
emotion (Bandura, 1986). Bandura’s early experiments provided evidence of this
argument. The emotional reactions of two groups of people, those who were told that the
associated painful outcomes had ceased and those who were not told that the threat no
longer existed, were compared. The study found that induced awareness promptly
eliminated fear arousal and avoidance behavior in the informed participants, while the
uninformed lost their fear only gradually (Bandura, 1969). Bandura further showed that
environmental cues can induce emotional reactions, while the impact of these cues on
emotion depends on cognitive processing (Bandura, 1986). The judgment of the

likelihood of aversive outcomes does not rely solely on reading external signs of danger

or safety. Rather, it involves a transaction between personal capabilities and potentially
hurtful aspects of the environment.

Bandura deﬁned anxiety as “a joint effect of two separate factors, visceral arousal
and cognitive self-labeling of the internal state as anxiety” (Bandura, 1986, p. 191). This
deﬁnition indicates that cognition plays a very important role in anxiety. Bandura further
speciﬁed that self-efﬁcacy and outcome expectation beliefs predict how well people cope
with threats and how much anxiety they experience (Bandura, 1986; Bandura et al.,
1982). People experience high anxiety associated with tasks in which they perceive
themselves to be inefﬁcacious and with tasks that they envision as having negative

outcomes (Bandura, 1986).

CHAPTER 3

LITERATURE REVIEW

This chapter reviews prior studies that are relevant to this dissertation. Literature
exploring blogs and different types of anxiety is summarized. In the following section,
the deﬁnition and typologies of blogs will be presented, followed by a discussion of the

speciﬁc characteristics of blogs.

3.1 Blogging
3.1.1 Blog Definition and Typologies

A blog is a kind of frequently modiﬁed web page in which dated entries are listed
in reverse chronological sequence (Herring et al., 2004a). Typologies have been
developed to capture the differences of blogs. Krishnamurthy divided blogs into four
quadrants with two axes: personal versus topical and individual versus community
(Krishnamurthy, 2002). Blood proposed two types of blogs: ﬁlter-style blogs and joumal-
style blogs. Filter-style blogs are usually thematic blogs that link to and comment on
online resources pertaining to a particular interest. Journal-style blogs, on the other hand,
are made up of a blogger’s thoughts or anecdotes from daily life (Blood, 2000). In a later
article, Blood added another type of blogs—notebooks, which may contain either
external or internal contents and are distinguished by longer, focused essays (Blood,
2002). Herring et a1. applied Blood’s typology to analyze 203 randomly sampled blogs.
The study found that 70.4% of the sampled blogs were of the personal journal type, while

the ﬁlter blogs only accounted for 12.6%. Knowledge blogs, the notebook type,

accounted for even fewer, at 3% of the sample (Herring et al., 2004a). Kawaura,
Kawakarrri, and Yamashita (1998) differentiated four types of blogs: a record of fact type,
an expression of sentiment type, a self-oriented type, and a relationship-oriented type.
The study found that the bloggers who wrote reader-oriented blogs made more effort to
have the existence of their diaries widely known than those who wrote self-oriented blogs.
The bloggers predominantly expressing sentiment were more inclined to believe that
readers were interested in people (the author) compared with bloggers predominantly
expressing facts (Kawaura et al., 1998).
3.1.2 Blogging Motivations

Kavanaugh et al. suggested that blogs represent self-organizing social systems
that can help many persons to: (l) interact collaboratively, (2) learn from each other by
exchanging ideas and information, and (3) solve collective problems. Blogs are tools for

greater social interaction, informal discussion, and conversation (Kavanaugh et al., 2006).

Nardi et a1. investigated the motivations behind blogging through ethnographic
interviews with bloggers. The study found ﬁve principal motivations of blogging:
documenting one’s life, providing commentary and opinions, expressing deeply felt
emotions, articulating ideas through writing, and forming and maintaining community
forums (Nardi et al., 2004b). Trevino identiﬁed three blogging motivations by
interviewing college students: the power over the website; the pull medium through
which their audience actively accesses content; and the generally positive feedback from
unknown readers, particularly on personal subjects (Trevino, 2005). Under the uses and
gratiﬁcations approach, Li identiﬁed seven blogging motivations: self-documentation,

improving writing, self-expression, medium appeal, information, passing time, and

10

socialization (Li, 2005). Two types of bloggers emerged from Marlow’s survey of
random bloggers: professional and social, with differing behavior and motivations.
Professional bloggers, in the minority, tended to have larger audiences and weaker social
relations with their audience. Social bloggers, on the other hand, had smaller audiences
but more multiplex relationships (Marlow, 2006b). Eﬁmova differentiated two types of
blogging motivations: motivations to start a blog and motivations to maintain a blog.
Getting hands-on experience with blogs, exploring opportunities for their use in a speciﬁc
context (e.g., business, teaching, and knowledge management), a channel to communicate,
and a need for expressing and publishing ideas are the motivations for starting a blog. For
blog-maintenance motivations, improving knowledge and skills; sharing, evaluating and
developing their ideas; and the social effects of blogging were most frequently identiﬁed.
Comparing these motivations to those of starting a blog, blog-maintenance motivations
highlighted the social effects of blogging, such as ampliﬁed networking, relation-building,

and community-forming (Eﬁmova, 2004).

The motivations identiﬁed in previous studies can be seen as the outcomes
bloggers expect (LaRose & Eastin, 2004). Following this argument, the studies reviewed
in this section showed that bloggers’ outcome expectations vary with blog types. For
personal diary bloggers, their blog outcome expectations are centered on communication,

socialization, self-expression, and self-documentation.

3.1.3 Personal Media with Public Attention

Blogs are personal media with public attention. Blogging is an externalizing of
the private mind in a public way. This new computer mediated communication (CMC)

genre presents an interesting paradoxical combination of private and public sphere, which

11

allows the private domain to become public and privatizes a portion of the public sphere

(MacDougall, 2005 ; Papacharissi, 2007; Rak, 2005; Rosen, 2004).

Blogs are personal media. Papacharissi’s content analysis of a random sample of
260 blogs hosted by Blogger.com suggested that a blog is a self—reﬂective account that
serves the purpose of personal expression and provides the perceived gratiﬁcations of
self-fulﬁllment (Papacharissi, 2004). On the other hand, blogs “invite the audience’s
gaze” (Scheidt, 2006). Bloggers address their potential audiences in a very direct manner

(Papacharissi, 2007).

Audiences can be a source of tension and threats. First of all, blogs’ selectivity is
low. Selectivity is the way communication can be exclusively targeted at one person or a
group of people (Wijnia, 2004). Bloggers do not have the option of revealing posts to
various subsets of individuals in their personal network (Park, 2003a). Second, blogs’
audiences are heterogeneous. Li (2005) identiﬁed three types of audiences: targeted,
expected, and actual readers. The author suggested that a blogger intends to write for
some speciﬁc readers. However, due to the openness of the Internet, the blogger should
expect some untargeted readers to be reached. In addition, the ﬁnally reached audiences
can be another group of people (Li, 2005). Trevino indicated that misjudged or unknown
audiences (parents, employers, future children, future political constituency) can lead to

serious interpersonal and perhaps professional repercussions (Trevino, 2005).

Utilizing blogs, the personal media, bloggers have to maintain two delicate
balances: the balance between satisfying different types of audiences and the balance

between satisfying themselves and their audiences (Gumbrecht, 2004; Lenhart, 2005; Li,

12

2005). Sometimes, the cquilibria might be disturbed (Bortree, 2005). Through
ethnographic interviews, Lenhart found that (Lenhart, 2005):

Most bloggers feel a signiﬁcant tension over their audiences. While on one hand a

blog is a personal space, it is also a public space that is created with an

expectation of an audience. Bloggers must come to grips with how much they
want a readership, how much they want it to increase and how that desire impacts
how they will express themselves online. And frequently, they ﬁnd themselves
conﬂicted, uncertain and uncomfortable, and yet simultaneously supported and
afﬁrmed by the thought of or the actions of virtual (actual or envisioned) visitors.

Bloggers must continually navigate the line between being authentically

themselves (or a version thereof), protecting their privacy and entertaining their

readers (Lenhart, 2005, page 102).

Lack of privacy is the largest drawback of blogs. Wilnia argued that privacy is
hard to maintain on blogs because of high stimuli richness and information complexity.
(Wijnia, 2004). It is one of the reasons that people choose not to blog (Eﬁmova, 2004).
Viegas (2005) conducted an online survey to examine bloggers’ subjective sense of
privacy. The study found that bloggers worry about a number of privacy issues, ranging
from minor embarrassments with family and friends to termination of their employment
(Viegas, 2005).

Thus, blogs are personal media with public attention. Audiences are an important
component of blogging. Audiences can be a reason for people to start blogging, to keep
blogging, and to drop out. Audience is a source of positive outcomes (i.e., interactions
and communications), negative outcomes (i.e., conﬂicts, frustrations, uncertainties,

tensions, and privacy invasion), and anxiety. Therefore, blogs, as personal media with

public attention, bring positive and negative outcome expectations and anxiety.

13

3.1.4 Blogging in Other Cultures

Studies have examined blogging practices in other cultures. Scheidt argued that
blogs are socially and culturally constructed, which reﬂects ideas about how we interact
within a culture and how that a culture inﬂuences us (Scheidt, 2006). Lento et al. (2006)
used longitudinal data taken from Wallop, a blog system designed by Microsoft Research,
to compare Chinese bloggers and non-Chinese bloggers’ use patterns. The study found
that Chinese bloggers contributed more content and remained active in the system longer
than non-Chinese bloggers. The study argued that the reasons for the differences are the
strength of social ties and preexisting social networks. Chinese bloggers had stronger
social ties than non-Chinese bloggers. There were also more overlapping ties among their
blog worlds and their real lives (Lento et al., 2006). Su et al. investigated the inﬂuence of
a regional culture on a blogging community. A multilingual worldwide blogging survey
was conducted with 1,232 participants. The results suggested that overall, bloggers
around the world share quite a bit in common in terms of how they experience
community through their blogs. However, some differences do exist between Eastern and
Western cultures. For example, Japanese bloggers’ score was highly skewed toward not

revealing identities, even with the use of aliases (Su et al., 2005).

To summarize, previous studies discussed the deﬁnition and typologies of blogs.
Blogs have been seen as personal media with public attention. The roles of audiences are
highlighted in recent studies. Privacy is a concern bloggers have about blogging. Culture
inﬂuences blogging practices. Synthesizing this research, the starting point of the present
study are the characteristics of blogs. Blogs are a new Internet application that has unique

formats and features, producing both positive and negative expectations and anxiety

l4

concerns. In the following section, the focus will switch from the medium to the user.
Walther (1997) posited that computer mediated communication research should not focus
mainly on the features of the medium, but on the personal factors of the users.
Speciﬁcally, what needed to be explored in whether certain types of individuals are
inclined, or not inclined, to blog. In the following section, literature of anxiety will be
reviewed because bloggers’ trait and state anxieties are important determinants of their

blog-maintenance intentions.

3.2 Anxiety—an Introduction

In this section, reviews of general anxiety theories and measures will be
introduced. Five types of anxiety that are related to this study—social anxiety,
communication apprehension, writing apprehension, computer anxiety, and computer-
mediated communication apprehension—will then be presented.

We will see that from the social cognitive perspective developed here, anxiety and
blogging are related in that anxiety is a personal factor that predicts blogging, a
comparatively new online practice. To confront the present problem of blog—maintenance
intentions, state and trait anxieties might act together in the following ways: generally
anxious individuals may avoid blogging in the ﬁrst place, and for those who decide to set
up a blog because of curiosity or positive outcome expectations, their situational
congruent anxieties will generate more negative outcome expectations and decrease their
blog-maintenance intentions.

3.2.1 Anxiety Deﬁnitions
Although the topic of anxiety has prompted voluminous research, differing

methodological formulations, and some attempts at comprehensive theories, there has

15

been little consensus as to exactly what anxiety is and how it can be reliably measured.
After summarizing the literature about anxiety, Spielberger and Sarason concluded that
anxiety has been conceptualized as a stimulus for behavior, as a learned drive, as a
personality variable, and as a complex response (Spielberger & Sarason, 1991). Anxiety
has been conceptualized as an independent variable (stimulus), an intervening variable
(drive, personality trait), and a dependent variable (response). Consistent with the social
cognitive perspective, anxiety is deﬁned as a joint effect of two separate factors, visceral
arousal and cognitive self-labeling of the internal state as anxiety (Bandura, 1986, p. 191).
Bandura’s deﬁnition of anxiety suggested that anxiety is a process through which
individuals cognitively label an arousal as anxiety after experiencing visceral arousal.
This deﬁnition does not indicate that anxiety should be measured as a two-factor concept.
In fact, Bandura has never produced a two-factor anxiety measure.
3.2.2 Anxiety Typologies

A trait-state typology has been applied to examine anxiety (Spielberger et al.,
1983). Anxiety as an emotional state (A—State) is characterized by subjective, consciously
perceived feelings of tension, apprehension, and nervousness accompanied by or
associated with activation of the autonomic nervous system (Spielberger, 1972, 1985;
Spielberger & Sarason, 1991). Trait anxiety (A-Trait) refers to relatively stable individual
differences in anxiety proneness and in the disposition or tendency to perceive a wide
range of situations as threatening and to respond to these situations with differential
elevations in state anxiety (Spielberger, 1972, 1985; Spielberger & Sarason, 1991). The
trait approach minimizes the role of situational and environmental factors to maximize

the role of individual internal differences and structures on behavior. On the other hand,

16

the state approach underlines the inﬂuence of situational factors. The state paradigm
suggested that relatively stable and consistent individual attributes exert little inﬂuence
on behavior.

Studies have framed the trait-state distinction as the relative importance of distal
and proximal inﬂuences in explaining behavior (Brown et al., 2004). This perspective
argued that there is a continuum from distal individual differences that are thought to be
trait-like to more proximal individual differences that are thought to be state-like.
Proximal constructs serve as mediators for distal constructs because proximal constructs
are more temporally proximal to behavior (Chen et al., 2000; Kanfer, 1990). Rosenberg
examined the interaction between trait self-esteem and state self—esteem. The researcher
made a distinction between baseline global self-esteem and barometric self-esteem that
represents short-term ﬂuctuations over situations and time. This model suggested that the
barometric state is a function of the baseline trait (Rosenberg, 1986). This model was
supported by Cole et al.’s trait and state linear model. Cole and colleagues conceived a
state as a latent variable that can be represented by one or more manifest variables at time
t. A state can be decomposed into trait (T) and occasion (0t) factors. A trait is a
completely stable person factor. The occasion factor (0t) represents the situational
circumstances impinging on the person at the time of testing plus the person-situation
interaction (Cole et al., 2005).

In addition to the trait and state typology, researchers deﬁned a new type of
anxiety: situation speciﬁc trait anxiety (Horwitz, 2002; Martens et al., 1990; Spielberger,

1972). Spielberger suggested that “in general, situation-speciﬁc trait anxiety measures are

17

better predictors of elevation in A-state for a particular class of stress situations than are
general A-trait measures” (Spielberger, 1972, p. 490).

Studying communication apprehension, McCroskey proposed a new typology.
Based on his typology, there are four types of communication apprehension: trait-like,
generalized-context, audience-based, and situational communication apprehension.
McCroskey deﬁned communication apprehension as “an individual level of fear or
anxiety associated with either real or anticipated communication with another person or
persons” (McCroskey, 1982, p. 148). Trait-like communication apprehension is a
relatively enduring, personality- type orientation toward a given mode of communication
across a wide range of contexts. Generalized-context communication apprehension is a
relatively enduring, personality-type orientation toward communication in a given type of
context. McCroskey and Richmond (1980) further identiﬁed four general communication
contexts: public speaking, meetings, small groups, and dyadic situations (McCroskey,
1984). Audience-based communication apprehension is a relatively enduring orientation
toward communication with a given person or group of people. Situational
communication apprehension is a relatively transitory orientation toward communication
with a given person or group (McCroskey, 1982; Richmond & McCroskey, 1998).

The perspective of the present research is an integration of the trait and state
perspectives and social cognitive theory. Specially, this research suggests that occasion
factors have inﬂuence on both trait and state variables. Situational trait-like variables and
situational state-like variables are more accurate terms to describe these variables than
traits and states. Moreover, following the arguments of Cole et al. (2005), Chen et al.

(2000), and Kanfer (1990), this research posits that situational trait-like variables

18

determine situational state-like variables. For example, social anxiety, a situational trait-
like variable which will be introduced in the following section, will predict online posting
anxiety, a state-like variable for blogging.

3. 3 Cognitive Perspectives of Anxiety

The cognitive perspective of anxiety is the perspective applied in this dissertation.
Cognitive models of anxiety suggested that anxiety is a result of cognitive processes
(Arnold, 1960; Bandura, 1986; Lazarous, 1991, 1995; Lazarus, 1991, 1995; Lazarus et al.,
1970; Lazarus & Folkman, 1984; Mandler, 1984; Reisenzein, 2001; Roseman et al., 1996;
Scherer et al., 2001; Spielberger, 1972; Spielberger & Sarason, 1991). Lazarus and
associates have developed what is perhaps the most systematic and extensive formulation
of the cognitive perspective of anxiety. They employed the concepts of primary appraisal,
secondary appraisal, and reappraisal while arguing that anxiety is a function of these
appraisals. Their model emphasized the relationship between a person and his/her
environment, which took into account characteristics of the person on the one hand, and
the nature of the environmental events on the other. For this person-environment
relationship, cognitive appraisal is a critical element (Lazarus & Folkman, 1984).

Other cognitive models also underscored cognitive inﬂuences of anxiety.
Mandler’s interruption theory posited that anxiety is generated in a process when ongoing
cognitive activities are interrupted. This interruption produces a diffuse autonomic
discharge. The autonomic discharge further results in detailed appraisal of the source of
the interruption, which is then evaluated either positively or negatively, depending on the
results of the appraisal. A negative evaluation and high arousal will generate anxiety

(Mandler, 1984). Similarly, Spielberger argued that the appraisal is the most critical point

19

in the process of anxiety generation (Spielberger, 1985). Beck suggested that “cognitive
schemata,” the automatic thoughts and images which are generated from distorted
information processing, trigger inappropriate motor, physiological, and affective
components of anxiety responses (Beck, 1993; Beck et al., 1985). To summarize,
cognitive perspectives of anxiety suggest that: (1) environmental stimuli lead to
cognitions; (2) cognitions generate anxiety; (3) anxiety brings forth behavior (e.g.,
avoidance) (Brown et al., 2004).

Thus, we can integrate this perspective with social cognitive theory as follows:
environmental cues, together with cognitive appraisal (i.e., assessments about personal
skills, capabilities, and possible outcomes), will induce anxiety, which in turn will
generate avoidant behavior. In the particular case of blogging, these theories might
explain the anxiety bloggers may experience. For example, the novelty and uncertainty of
blogs and the heterogeneity of the audience, together with low self-efﬁcacy and negative
outcome expectations, will induce anxiety about blogging and will create low blog-
maintenance intentions.

3. 4 Integrationist Perspectives of Anxiety

Before further discussing the integrationist perspectives of anxiety, it is important
to differentiate between two approaches: anxiety as a theoretical framework and anxiety
as a pure phenomenon. Anxiety as a phenomenon should be a pure affective arousal
index. Many anxiety indices, however, consist of cognitive, affect, and behavior
components. This approach is problematic because “asking whether anxiety inﬂuences
behavior would be to ask whether behavior inﬂuences itself” (Bandura, 1986, p. 190).

However, anxiety theoretical frameworks can embrace behavior, cognition, and other

20

factors to explain the anxiety phenomenon. The following introductions will focus on
theoretical frameworks that explain the anxiety phenomenon.

Anxiety theorists now see the necessity for an integrationist perspective that
encompasses neurobiological, behavioral, and cognitive subsystems (Barlow, 2002). The
process of anxiety is extremely complex and involves a number of different measurable
components (Spielberger & Sarason, 1991). Asmundson and colleagues suggested that
four components tend to be involved in anxiety: information processing biases,
physiological reactions, safety-seeking behaviors, and affective changes (Asmundson et
al., 2002). Horwitz suggested that anxiety has three components: cognitive, physiological,
and behavioral. The cognitive component is a persistent, chronic sense of uneasiness or
dread from negative perceptions about a present or future event or interchange with a
person. The concurrent physiological process of alarm and activation is marked by
changes in neural chemistry, among other responses. The behavioral response is often
manifested in an emergency ﬁght-or-ﬂight reaction (Horwitz, 2002).

To explain blog-maintenance intentions, the mechanism could work as follows:
cognitive factors and bloggers’ general tendencies to be uneasy can generate online
posting anxiety. This online posting anxiety, together with the cognitive factors, will
inﬂuence blog-maintenance intentions.

3. 5 Anxiety in Different Cultures

Cultural differences are heuristic in the present context because they can inﬂuence

anxiety both directly and indirectly though cognitive components. Examining the

similarities and differences of anxiety across cultures can advance anxiety theories, since

21

the salience and importance of various factors (cognitive, affective, and behavioral
factors) in inﬂuencing anxiety can be discovered.

Anxiety is shaped by culture (Adamopoulos & Kashima, 1999; Barlow, 2002;
Markus & Kitayama, 1991; Wierzbicka, 1999). Individuals in all cultures are
apprehensive, worried, fearful, and aroused. But the objects of apprehension, the sources
of fear, and the speciﬁc attributions these individuals make are culture-speciﬁc (Barlow,
2002). Anxiety should be studied as a subjective feeling as well as a social product that is
inﬂuenced by social environment (Adamopoulos & Kashima, 1999). Wierzbicka
elaborated that “feelings are subjective... they are based on certain recurrent thoughts—
cognitive scenarios shaped by the particular culture...” (Wierzbicka, 1999, p. 306).
Culture offers a set of “cultural scripts” that tell people how to feel, how to express their
feelings, and how to think about their own and other people’s feelings (Wierzbicka,
1999). Markus and Kitayama (1991) suggested that anxiety experiences may differ
fundamentally between cultures. The ways of thinking and talking about anxiety in
different cultures and societies exhibit considerable diversity (Markus & Kitayama, 1991).
Individuals’ representations of a threat and their selection and evaluation of coping
procedures reﬂect the attitudes and beliefs embedded in their social and cultural
environment. This environment includes families, friends, mass media, socially deﬁned
roles, and the linguistic terms used to label and describe anxiety (Marks et al., 2005). For
instance, a given degree of arousal under conditions of potential failure, requiring
another’s assistance, might be diagnostic of “performance anxiety” in an individualist
culture while of a “feeling of indebtedness” in a collectivist culture (Adamopoulos &

Kashima, 1999).

22

In summary, in these sections, the theoretical frameworks that have been applied
to examine anxiety have been reviewed. In the following sections, ﬁve speciﬁc types of
anxiety will be reviewed. The various types are important because they are pertinent to
the topic of the current research—blog-maintenance intentions. Speciﬁcally, in the
context of blogging, each is important because blogging encompasses social,
communication, writing, and computer practices. Social anxiety will be introduced ﬁrst.
By integrating these perspectives, our understanding of blog-maintenance intentions can
be advanced.

3. 6 Social Anxiety

Holt et al. identiﬁed four social situations that may generate anxiety. These
situations are formal speaking/interaction, informal speaking/interaction, assertive
interaction, and observation by others (Holt et al., 1992). What is common about each of
these situations is that an individual is required to do something while knowing that
others will be watching and, to some extent, evaluating his/her actions (Barlow, 2002).
Beck et al. suggested that social anxieties are concerned with one’s exaggerated fear of
being the focus of attention and devaluation by another person or persons (Beck et al.,
1985).

Social anxiety has obvious relevance to blogging because bloggers are observed
and evaluated by others on their blogs. It is especially relevant to one particular type of
blogging discussed earlier, the personal journal type, because the bloggers, instead of
some media links or facts, are the focus of the audience’s attention.

Patterson and Ritts conducted a meta-analysis to investigate the relationship

between social and communicative anxiety and speciﬁc physiological, behavioral, and

23

cognitive reactions. The study found that social and communicative anxiety was more
strongly related to the cognitive measures than to the behavioral and physiological
measures. The effect sizes of self-efﬁcacy and negative thoughts were 1.56 and 1.18,
which can be identiﬁed as large effect sizes based on Cohen’s criteria (Cohen, 1977). The
results supported a cognitive view of social anxiety. A variety of negative self-focused
cognitions may be at the core of social and communicative anxiety. Schneier and
Welkowitz’s study supported this conclusion. Their study found that socially fearful
people have been found to think differently from others. They have more negative
thoughts about their performance, compared to subjects with general anxiety and those
without anxiety, and consistently underestimate the quality of their performance
(Schneier & Welkowitz, 1996). Other studies conﬁrmed the negative relationship
between social anxiety and self-efﬁcacy beliefs (Leary & Kowalski, 1995). People with
social anxiety think of themselves as less skilled or efﬁcacious (High, 2006).

Thus, social anxiety is relevant to blog-maintenance intentions because of
exaggerated fear of being the focus of attention. In the social cognitive view, social
anxiety may be interpreted as visceral arousal caused by negative self—focused cognition,
i.e., low self-efﬁcacy and negative outcome expectations, in social situations. It should be
noted that self-efﬁcacy and outcome expectation vary in the dimension of generality. The
social self-efﬁcacy and outcome expectations that determine social anxiety are more
general than blogging self-efﬁcacy and outcome expectations and more speciﬁc than
general self-efﬁcacy and outcome expectations. Self-efﬁcacy and outcome expectations

directly precede anxiety at the same level of generality.

24

3.7 Communication Anxiety

McCroskey deﬁned communication apprehension as “an individual level of fear
or anxiety associated with either real or anticipated communication with another person
or persons” (McCroskey, 1982, p. 148). Self-efﬁcacy and outcome expectations are
related to communication anxiety (Ayres & Hopf, 1985; Ayres & Schliesman, 2002;
Lucchetti et al., 2003; MCcullough et al., 2006; Miller, 1987). Phillips viewed a reticent
communicator as an unskilled or incompetent communicator (Phillips, 1980). Burgoon
and Hale (1983) showed that people who are “unwilling to communicate” perceive
communication as unproﬁtable (Burgoon & Hale, 1983). Greene and Sparks (1982)
described three characteristics of apprehensive individuals: (1) perceptions of low
personal competency, (2) an inability to identify appropriate social behavior, and (3)
anticipation of negative outcomes of communication (Greene & Sparks, 1982). These
researchers developed a cognitive process model of communication apprehension one
year later. The model suggested that communication apprehension arises when an
individual is unable to identify the behavior that is expected to lead to the
accomplishment of interaction goals (Greene & Sparks, 1983). Ayres (1997) proposed a
component theory of communication apprehension. He suggested that communication
apprehension in a given situation is a product of the interaction between self-perception
of communication competence, negative evaluation, and motivation (Ayres, 1997).

Thus, communication apprehension might be relevant to the study of blog-
maintenance intentions because communication is an important component of blogging.
Communication apprehension can be an obstacle that bloggers must overcome in order to

keep blogging.

25

3.8 Writing Anxiety

Writing anxiety is potentially relevant to the problem of blog-maintenance
intentions because blogging is a form of text communication. Bloggers may experience
writing anxiety when they start to write a post, work on a post, or ﬁnish a post. They may
quit blogging altogether because they worry about the evaluations of their blog writings.

The most studied writing-related anxiety is writing apprehension (Daly, 1978).
Daly (1978) deﬁned writing apprehension as an individual difference characterized by a
general avoidance of writing and of situations perceived by the individual to potentially
require some amount of writing, accompanied by the potential for the evaluation of that
writing. It is obvious that Daly deﬁned writing apprehension as a behavioral tendency
(approach or avoidance). He further explained that a person high in writing apprehension
will perceive writing as unrewarding and will consequently seek to avoid it. If such an
individual cannot avoid such situations, he or she will experience abnormal amounts of
anxiety.

Contexts are highlighted by many writing anxiety researchers. Five situational
characteristics were identiﬁed to affect situational state writing apprehension: evaluation,
novelty, ambiguity, conspicuousness, and previous experience (Daly & Hailey, 1984).

Self-efﬁcacy and outcome expectations are pertinent to writing apprehension.
Mabrito argued that a history of perceived negative evaluations and the prospect of
having one’s writing evaluated in a public forum form the basis of writing apprehension
(Mabrito, 1989). Crumbo’s study conﬁrmed the negative relationship between writing
self-efﬁcacy and writing apprehension (Crumbo, 1999). Corbett-Whittier argued that the

most common causes of writing apprehension are self-efﬁcacy beliefs and past

26

educational experience (Corbett-Whittier, 2004). A review by Boice found that writing
anxiety is induced by the presence of internal censors that act to immobilize the writer,
who has accepted and exaggerated negative comments from teachers and other authority
ﬁgures. These censors cause the writer to reject what he/she might write before and
during the writing process (Boice, 1993). Within social cognitive theory, the internal
censors Boice described can be referred to as efﬁcacy beliefs or outcome beliefs. Pajares
and Johnson grounded their theory of writing anxiety in Bandura’s social cognitive
theory. They found that writing self-efﬁcacy and writing outcome expectations, together

with writing anxiety, inﬂuence students’ writing performance (Pajares & Johnson, 1994).

Through the socio-cognitive lens, writing anxiety can be seen as a visceral arousal
in writing situations. People who experience this arousal label it anxiety because of their
low writing self-efﬁcacy, low positive outcome expectations for writing, and high
negative outcome expectations for writing. Again, the level of generality is important
here. Writing anxiety is directly determined by writing self-efﬁcacy and outcome

expectations.

3.9 Computer Anxiety

Social anxiety, communication apprehension, and writing anxiety are task-
oriented anxieties. In this section, the review will be focused on computer anxiety, a
media-related anxiety.

Computer anxiety is deﬁned as fear of computers or the tendency of a person to
be uneasy, apprehensive, and phobic towards current or future use of computers in

general (Al-Khaldi & Al-Jabri, 1998; Carnbre & Cook, 1985; Igbaria, 1993; Loyd &

27

Loyd, 1985). Rosen, Sears, and Wei] (1993) defined three types of computer-anxious
individuals: (1) anxious computerphobics who exhibit signs of anxiety reactions (e.g.,
sweaty palms, heart palpitations, headaches, and so forth) when dealing with computers;
(2) cognitive computerphobics who, despite a calm exterior look, are internally fearful of
computers and continually think negatively about dealing with computers; and (3)
uncomfortable users who are slightly anxious merely because of lack of information
about computers.

Beckers and Schmidt (2001) proposed a social cognitive model of computer
anxiety. Their model was comprised of six factors: (1) computer literacy, (2) self—efﬁcacy,
(3) physical arousal in the presence of computers, (4) affective feelings towards
computers, (5) positive beliefs about the beneﬁts for society of using computers, and (6)
negative beliefs about the dehumanizing impact of computers.

Computer anxiety research has two limitations: (1) most of the research was
conducted in the late 19805 and early 19908, with the majority of the research following
this period seeking to replicate or extend the original ﬁndings (Hemby, 1998); and (2)
most of the scales used to measure computer anxiety have no explicit psychological or
psycho-sociological theory basis.

The relevance of computer anxiety to blog-maintenance intentions is not
immediately apparent, however. Computers are the exact tool for blogging. Computer
anxiety may make people avoid computers in general and blogging speciﬁcally. Through
social cognitive theory, it is possible to clarify some of the confusion of the computer
anxiety construct by interpreting its varied dimensions in terms of social cognitive

theoretical variables. For example, in the work of Rosen et al. (1993), three types of

28

computer-anxious individuals were summarized: (1) those who experience a visceral
arousal when dealing with computers; (2) those who perceive negative outcomes; (3)
those who have low computer self—efﬁcacy.

3.10 A Social Cognitive Critique of Communication Apprehension, Writing
Apprehension, and Computer Anxiety

Communication apprehension, writing apprehension, and computer anxiety
studies have made signiﬁcant contributions. Based on the theoretical principles of social
cognitive theory, however, these studies have limitations.

McCroskey’s communication apprehension studies omit an important concept that
is particularly relevant to communication—self—efﬁcacy. Communication, no matter
whether in public speaking, meetings, small groups, or dyadic situations, requires skills to
exchange information, encode/decode verbal and nonverbal cues, and interact with
people with different communication goals.

The trait-state conceptual deﬁnitions of communication apprehension, writing
apprehension, and computer anxiety are ambiguous as well. McCroskey’s View of
communication apprehension changed from a state View to a trait view. However,
McCroskey’s PRCA-24 measurement approach betrays the researcher’s trait view.
PRCA-24 looks at public speaking, meetings, small groups, and dyadic situation anxieties
to form a communication apprehension trait measure. This approach was criticized by
Bandura. Using aggression as an example, Bandura suggested that lumping different
kinds of physical aggression, verbal aggression, and oppositional behavior into an
indeﬁnite conglomerate can boost correlations but yields indeterminate or empty

predictions. “To be able to predict through aggregation that individuals will sometime,

29

somewhere, do something within a wide assortment of acts is of no great interest.”
(Bandura, 1986, p. 10)

The same argument can be applied to the PRCA-24. Mixing four different
communication situations obscures the understanding of psychological functioning and
reduces the predictability.

Daly originally posited writing apprehension to be a trait. In later studies, Daly
recognized the important role of individual writing occasions. Daly abandoned the trait
view of writing apprehension, and deﬁned it as a behavioral tendency. This is
problematic, too. As Bandura suggested, using the behavioral tendency to explain
behavior is the problem of circularity.

Research involving computer anxiety did not specify whether computer anxiety is
a trait or a state. However, the most used deﬁnition of computer anxiety is very similar to
McCroskey’s communication apprehension deﬁnition. It is reasonable to argue that these
researchers perceive computer anxiety as a trait. Interestingly, many computer anxiety
measures comprise different areas of computer use [see aforementioned computer and
Internet anxiety scales (Chou, 2003; McInemey et al., 1999; Mcinemey et al., 1994;
Presno, 1998; Rosen & Weil, 1995)]. Although this approach is in line with social
cognitive theory, it is problematic. First, the measures are not mutually exclusive.
Additionally, rapid changes in computers and the technological environment may make
these scales not relevant for future computer and Internet use studies. In fact, the majority
of computer and Internet measurement instruments were developed in the 19803 and
early 1990s. Marcoulides and colleagues (2004) argued that people may be experiencing

lower levels of computer anxiety, or at least different forms of computer anxiety, than

30

they were in the past. It is obvious that certain forms of computer and Internet use (e. g.,
email and Word) probably do not induce intensive anxiety in users anymore. However,
newer forms of computer use (e.g., podcasting and blogging) may generate anxious
experiences.
3.11 Social Anxiety, Communication Anxiety, Writing Anxiety, and New Media
Social anxiety, communication apprehension, and writing apprehension have been
explored in the new media environment (Bline, Lowe, Meixner, Nouri, & Pearce, 2001;
Campbell & Neer, 2001; Carlson & Wright, 1993; Chelley & Larry, 2002; Clarke, 1991;
Fishman, 1997; Hartman et al., 1991; Karahanna, Ahuja, Srite, & Galvin, 2002; Leen &
Ramayah, 2006; Mabrito, 1989; Mabrito, 2000; Mazur, Burns, & Emmers-Sommer, 2000;
McCarson, 2005; McDowell, 1998; Reed, 1990; Reinsch, 1985; Scott & Rockwell, 1997 ;
Scott & Timmerman, 2005; Shaver, 1990; Sugiyarna, 2003; Wright, 2000b). Some
researchers have suggested that there is a negative relationship between anxiety and new
media usage (Campbell 8: Neer, 2001; Kivela, 1996; Madell & Muncer, 2006;
Mazur et al., 2000; Reinsch, 1985; Wright, 2000); other researchers have found no
relationship between them (Davis, 2003; Patterson & Gojdycz, 2000; Phinney, 1991;
Sugiyarna, 2003), and some have even argued for a positive one (Fuller et al., 2006;
Hartman et al., 1991; McDowell, 1998; Patterson & Gojdycz, 2000; Reinsch, 1985;
Wright, 2000). Researchers have also examined various directions of the relationship
between anxiety and new media usage. Anxiety has been applied both as an independent
variable (i.e., the effects of anxiety on new media usage) and a dependent variable (i.e.,

the effects of new media on anxiety).

31

Not all these studies are directly relevant to blog-maintenance intentions; here
only relevant ones are reviewed. Starting with the studies that argue that writing
apprehension and communication apprehension will decrease CMC tools usage, this
section will then introduce the studies suggesting that new media practices can be
employed as a method to research social anxiety, communication apprehension, and
writing apprehension. These conﬂicting views are then integrated to propose the
existence of online communication anxiety.

Harris and Grandgenett surveyed 189 randomly selected teachers with accounts
on Tenet (Texas Education Network) and examined the relationship between writing
apprehension, communication apprehension, computer anxiety, and network use.
Network use was measured by one year of logins and online time on Tenet. These
researchers found that subjects with high writing apprehension tended to use Tenet less
than those with lower writing apprehension. Communication apprehension and computer
anxiety did not correlate with Tenet use. Therefore, the researchers concluded that
writing apprehension was the main anxiety factor that affected Tenet use (Harris &
Grandgenett, 1996).

Fishman’s study of high school students conﬁrmed this argument (Fishman,
1997). The researcher explored the relationship between communication apprehension,
writing apprehension, and CMC tool use. The CMC tools examined in the study were
email, Usenet news groups, collaborative groupware, and videoconferencing. The study
found a signiﬁcant negative correlation between writing apprehension and the use of
Usenet news, and a marginally signiﬁcant negative correlation between writing

apprehension and collaborative groupware and overall CMC tool use. Writing

32

apprehension was not signiﬁcantly correlated with email use and videoconferencing use.
Fishman argued that the nature of the tools decided the strength of the relationships.
Videoconferencing uses video, instead of text, to communicate. It is not relevant to
writing apprehension. For the rest of the CMC tools, email is the most private of the
CMC tools, followed by semiprivate collaborative groupware and then by Usenet news,
the least private tool. Hence, the author suggested that the private levels of the CMC tools
were related to writing apprehension. The students who exhibited a high degree of
writing apprehension were less likely to use more public CMC tools (Fishman, 1997).
Brown and colleagues found that communication apprehensives have negative attitudes
toward using email and avoid using it (Brown et al., 2004).

Studies also have argued that new media can provide a low-anxiety environment
for writing, interaction, and communication (High, 2006; Lea & Spears, 1995; Loch et al.,
2003; Parks & Floyd, 1996; Wright, 2000). Scott and Rockwell interpreted their
nonsigniﬁcant relationship between writing apprehension and text-based technologies use
to indicate that writing-apprehensive students may not experience many difﬁculties in
using new text-based technologies, such as email, online discussion groups, and word
processing, because these devices are less personal and less public than other types of
written communications (Scott & Rockwell, 1997). Campbell et al. compared online chat
users with nonusers on social anxiety and found that, after controlling for time online,
chat users displayed signiﬁcantly less social fearfulness than nonusers of chat. The
researchers suggested that socially anxious individuals may use online chat as a low—risk
social approach and an opportunity to rehearse social behavior and communication skills

(Campbell et al., 2006). Hiltz and Turoff contended that CMC is better organized than

33

natural conversations because users have more time to think about what they are writing
(Hiltz & Turoff, 1993). Walther argued that socially anxious people can have more
control over the interactions and present themselves better online. The controllability and
anonymity of computer-mediated communication can help socially anxious people plan
ahead and feel less anxious (Walther, 1996). Many of the situational factors that can
foster feelings of anxiety in social situations (e.g., having to respond on the spot, feeling
under visual scrutiny) are absent in online interactions. People tend to feel more
comfortable online (Amichai-Hamburger & McKenna, 2006).

Empirical studies conﬁrmed these arguments. Cooper and Sportolari noted that
shy people carefully plan out their communication in computer-mediated communication
contexts, thus gaining more conﬁdence in their online interactions (Cooper & Sportolari,
1997). Kivela found that as many as 72% of the participants felt more conﬁdent about
expressing their ideas in the electronic environment because they had more time to think
and organize their ideas before sharing them with their peers (Kivela, 1996). Mabrito
found that high-apprehensive writers contributed to and initiated more topics of
discussion and felt more comfortable participating in electronic discussions with
unknown audiences than they did when communicating with familiar audiences (Mabrito,
1989; Mabrito, 2000). Karahanna et al. suggested that communication apprehensives
embrace the Internet as a way to avoid face-to-face communication (Karahanna et al.,
2002). Peter and Valkenburg surveyed 687 adolescents to investigate how social anxiety
inﬂuences adolescents’ perceptions of Internet communication. The results suggested that
socially anxious adolescents valued the controllability of Internet communication more

than adolescents who are not socially anxious. Compared with adolescents with low

34

social anxiety, adolescents high in social anxiety perceived Internet communication to be
more reciprocal, broader and deeper than face-to-face communication (Peter &
Valkenburg, 2006). Moreover, socially anxious people do manage their impressions
better online. McKenna et al. compared the ratings that socially anxious people received
from their interaction partners after face-to-face and Internet communication. Socially
anxious participants were viewed as more likeable and extroverted when they interacted
online than in face-to-face communication situations (cited in Amichai-Hamburger &
McKenna, 2006). High’s (2006) study reported similar ﬁndings: highly anxious
individuals were perceived to be signiﬁcantly less anxious after computer-mediated
conversations than following face-to-face interactions (High, 2006).

The literature presented conﬂicting results. One reason for the discrepancies are
the differing foci of the studies. Some studies attended to general Internet characteristics
(Hiltz & Turoff, 1993; Walther, 1996), while others looked at a speciﬁc type of CMC
tools, such as email (Fishman, 1997; Scott & Rockwell, 1997; Brown et al. 2004), online
chatting (Campbell et a1, 2006; High, 2006), online groups (Fishman, 1997; Mabrito,
2000; Amichai-Hamburger & McKenna, 2006), or video conferencing (Fishman, 1997).
The characteristics of the tasks, e. g., synchronicity, modality of the media, publicity,
controllability, and anonymity, are important factors that inﬂuence the relationship
between communication apprehension, writing apprehension, social anxiety, and CMC
use. Additionally, these studies employed various measures of CMC use. For example,
Fishman used the number of messages posted as the measure for email use and online

group use. Scott and Rockwell used a single question for each tool to measure the

35

likelihood of using email and online groups. The different tasks and measures might be a
reason for the conﬂicting ﬁndings.

More important, however, the majority of the studies failed to present a sound
theoretical framework that can explain the causal relationships between anxiety and CMC
use. In these studies, the theoretical base was the literature of communication
apprehension, writing apprehension, and computer anxiety, which has theoretical
shortcomings as mentioned above.

How might the conﬂicting results reviewed above be reconciled? Social cognitive
theory provides a provisional answer that forms the basis of the present research. CMC
use, a type of online behavior, is jointly determined by personal factors (cognitive factors
and anxiety) and environmental cues. Two types of personal factors need to be
distinguished: ofﬂine personal factors (ofﬂine self-efﬁcacy, ofﬂine outcome expectations,
and ofﬂine anxiety) and online personal factors (online self-efﬁcacy, online outcome
expectations, and online anxiety).

Ofﬂine personal factors will precede online personal factors because the online
context is a relatively novel context. Bandura posited that people construct outcome
expectations from observed conditional relations between environmental events in the
world around them and past related outcomes. Moreover, two sources of self-efﬁcacy are
mastery experiences and emotional states. It is proposed that when people face relatively
novel online situations, their previous ofﬂine experiences, beliefs, and anxiety, including
ofﬂine anxiety in social, communication, and writing situations, will inﬂuence their

perceptions and beliefs related to online situations. Ofﬂine social-, communication-, and

36

writing-anxious individuals may anticipate similar anxious experiences when they are in
online correspondent situations.

Environmental factors are the other important determinant of behavior. The
characteristics of the Internet, such as anonymity, interactivity, and asynchronicity, will
help to shape online personal factors and behavior. Individuals may form new systems of
online beliefs and anxiety that are different from their ofﬂine systems. For example, those
who are social apprehensive ofﬂine may develop high online social self-efﬁcacy and
positive outcome expectations because they have better control on the Internet. These
high online social efﬁcacy and positive outcome expectations will decrease their online
social anxiety. Online environment cues can help ofﬂine anxious individuals to recognize
the speciﬁc advantages of online socialization, communication, and writing, and further
form positive online experience, perceptions, and beliefs. These new beliefs decrease
their anxiety online. The Internet becomes a channel to decrease social, communication,
and writing anxiety.

In summary, human functioning should be analyzed as socially interdependent,
richly contextualized, and conditionally orchestrated within the dynamics of various
societal subsystems and their complex interplay (Bandura, 2001). Previous studies
examining CMC use and anxiety represented only part of the picture. They concentrated
either on environmental cues or on personality characteristics. These studies failed to
capture the triadic causations among environmental cues, personal factors, and behavior.
None of the studies pointed out the importance of self-efﬁcacy and outcome expectations
on behavior, nor did they explain the mechanism of the interactions among ofﬂine and

online environmental, behavioral, and personal factors.

37

Another line of research that reﬂects social cognitive views is new online anxiety
construct development. Bandura suggested that a situational-congruent concept is a more
salient predictor of behavior in a given situation (Bandura, 1986). In line with this
principle, scholars have developed new constructs that target apprehension when
communicating and writing online.

In his dissertation, Clarke developed a concept he called computer-mediated
communication apprehension (CMCA). It is deﬁned as an individual’s level of fear,
apprehension, or anxiety associated with using or the anticipation of using computers as a
medium to interact with another person or persons (Clarke, 1991). This deﬁnition is very
similar to McCroskey’s deﬁnition of Communication Apprehension--“an individual level
of fear or anxiety associated with either real or anticipated communication with another
person or persons” (McCroskey, 1982, p. 148)”. Although Clarke deﬁned CMCA as a
trait apprehension, it is a situational apprehension because it is a computer-mediated
communication situational apprehension. Flaherty and colleagues suggested that CMCA
is more state-like because it is computer speciﬁc (Flaherty et al., 1998). Another
limitation of Clarke’s study was that it mainly focused on email use apprehension. The
only correlation reported in the study was the signiﬁcant negative correlation between
CMCA and email use, which served as the evidence of the concurrent validity of CMCA.

Additionally, Clarke’s CMCA measure was confounded with self-efﬁcacy and
outcome expectations. The factor analysis results show that the CMCA scale was not a
pure measure of apprehension. The three dimensions of the CMCA scale were: (1)
conﬁdence in one’s abilities to use a computer to communicate (i.e., self-efﬁcacy); (2)

interest in using a computer to communicate (i.e., outcome expectations), and (3) privacy

38

concerns (i.e., negative outcome expectations) (Clarke, 1991). These three dimensions
are possible sources/causes of CMCA.

Wright examined online communication apprehension (OCA) and its impact on
online support group use. The study failed to ﬁnd a signiﬁcant correlation between OCA
and time spent on online communication. Moreover, the study did not provide a
deﬁnition of OCA. OCA was measured using PRCA-24 items, while asking respondents
to think about online relationships when responding (Wright, 2000).

Brown and colleagues developed a concept called computer-mediated
communication anxiety (CMC anxiety), which is deﬁned as an individual’s level of fear
or apprehension associated with the actual or anticipated use of information technology to
communicate with others (Brown et al., 2004). Again, CMC anxiety mainly focused on
email anxiety. Brown’s CMC anxiety measure consisted of ﬁve items, four of which
concerned email. Only one item measured listserv anxiety [I look forward to using email;
I would like to take courses in which email use is required; I am afraid of sending an
email message to a large group of people; I would be comfortable sending email
messages that I know a lot of people will read; I would be comfortable sending messages
to a listserv (where a large number of people will read them)] (Brown et al., 2004).

Scott and Timmerrnan (2005) introduced another version of CMC apprehension to
examine the use of new communication technologies in the workplace. They did not give
a speciﬁc deﬁnition of this concept. They generally described CMC apprehension as
anxiety about communicating through new communication technology (Scott &
Timmerman, 2005). They employed a broad deﬁnition of CMC. Nineteen technologies,

including landline telephone, cellular telephone, fax, and voice mail, were covered in

39

CMC. Thus, new technology apprehension, not CMC apprehension, is a more accurate
term to reﬂect Scott and Timmerman’s new concept.

Their CMC apprehension measure was modiﬁed from McCroskey 1970’s version
of PRCA (McCroskey, 1970), which was widely criticized. McCroskey abandoned this
scale. In particular, Scott and Timmerman’s 10—item CMC apprehension scale measured
behavior avoidance (“I always avoid communicating via a computer if possible”; “I look
forward to expressing myself during online meetings”), behavioral intention (“I look
forward to the opportunity to interact with others on the computer”), attitude (“I like to
get involved in computer-based group discussion”; “I dislike having to limit my
communication to what is possible on a computer”), and self-efﬁcacy (“I feel that I am
more skilled than most others when interacting with people online”). It may not be a valid
measure of apprehension in that it did not match the conceptual deﬁnition of the construct.

Fuller and colleagues examined the factors inﬂuencing the e-learning experience
of college students and developed a concept called email anxiety, which is deﬁned as fear
or apprehension associated with the real or anticipated use of a computer to communicate
via email (Fuller et al., 2006). Like Brown et al.’s study, the study of Fuller and
colleagues adopted a trait deﬁnition. But it is obvious that email anxiety is an application-
speciﬁc CMC anxiety (Fuller et al., 2006).

Fuller et al.’s measure of email anxiety consisted of six items1 (Fuller et al., 2006).

Four items measure anxiety emotion. However, two items concern intention.

 

' Using electronic email makes me nervous; I got a sinking feeling when I use electronic email; I avoid
taking courses in which electronic mail is used; Using computer-based communication tools like electronic
mail scares me; I look forward to using electronic mail; and I would like to take courses in which electronic
mail use is required.

40

It may seem at ﬁrst glance that CMCA and the other constructs may fully explain
the problem of blog—maintenance intentions. However, on a theoretical level they fail to
do so, as we will now proceed to demonstrate. Fuller et al. (2006), Brown et al. (2004),
and Clarke (1991) examined email. All failed to capture the public attention dimension of
the blogging experience. Blogs are personal media with public attention. Blogs’
selectivity is low. Bloggers do not have the option of revealing posts to various subsets of
individuals. Email users, however, can select the people their email messages intend to
reach. This is an important distinction theoretically because public attention is an
important source of anxiety.

Wright (2000) focused on online forums, reﬂecting the dimension of public
attention. Online forums, however, cannot capture the personal media feature of blogs.
Papacharissi suggested that blogs present a turn to personalized, self-referential, and self-
serving uses of the Internet (Papacharissi, 2007). Schuster compared blogs and earlier
online communication technologies, including bulletin boards, Usenet, and email listservs,
and concluded that the main difference between blogs and other technologies is that other
technologies are focused on speciﬁc topics or organizations. By contrast, blogs are
generally focused on speciﬁc individuals (Schuster, 2006).

Bloggers believe that blogs are their personal media. The blogger is the main
focus of the blog. Bloggers control the layout, the content, and the frequency of posting.
Few online forum users would perceive the forum as their personal medium. Forums
usually do not focus on users, but on certain topics or interests. Online forum anxiety is

distinguished from blogging anxiety. The same argument can be applied to examine Scott

41

and Timmerman’s (2005) workplace-related CMC apprehension. Users would not
perceive the workplace technology as their personal media.

A new variable for blogging, a situation-speciﬁc measure of anxiety, is needed
because blogging is a unique type of online communication behavior that is different
from email, online discussion groups, and other workplace technology practices. In line
with social cognitive theory, a situational interdependent behavior can only be explained
by a type of situational congruent anxiety.

Moreover, CMCA and the other theories have important shortcomings in their
operational deﬁnitions compared to the conceptual deﬁnitions of computer-mediated
communication apprehension/anxiety as types of situational anxiety. All the studies
adapted McCroskey’s trait anxiety measure. This can explain their sometimes
paradoxical ﬁndings because the operational deﬁnitions failed to take situational
inﬂuences into account. An operational measure that truly reﬂects the nature of the
concept in a given situation can resolve the seemingly irreconcilable ﬁndings in the
following way: online communication anxiety concepts should integrate both the nature
of the behavior and the nature of the situation. For example, the operational deﬁnition of
blogging anxiety should reﬂect: (1) the nature of blogging behavior, i.e., a personal
medium with public attention, and (2) the nature of the environment: text-based
communication, accumulative format of blogs, heterogeneity of audiences, and low
selectivity.

To match the conceptual deﬁnitions of anxiety, the operational deﬁnitions should
reﬂect the nature of the concept, i.e., the visceral arousal experience. In asking a question

worded: “I feel that I am more skilled than most others when interacting with people

42

online,” Scott and Timmerrnan failed to match their conceptual deﬁnition with their
operational one. Therefore a more suitable measure would be: “A constant awareness in
the back of my mind of my audience causes me to feel tense,” because it better reﬂects
the nature of the concept, the behavior, and the situation.

3.12 Online Posting Anxiety

In this dissertation, a new concept, online posting anxiety, is developed. Online
posting anxiety is deﬁned as a joint effect of two separate factors, visceral arousal and
cognitive self-labeling of the internal state as anxiety in blogging. This deﬁnition is
adapted from Bandura’s anxiety deﬁnition (Bandura, 1986). Here, blogging is deﬁned as
posting to a blog. Other blog related behaviors, such as read a blog or comment on a blog
post, are excluded from our deﬁnition of blogging.

It should be noted that blogging shares similarities with writing, social activities,
and computer-mediated communication activities. Blogs, however, are a unique
communication genre (Herring et al., 2004a). Blogs are personal media with public
attention. Blogging is a type of online communication that differs from oral
communication types such as public speaking, meetings, small groups, and dyadic
situations. Written texts and other multimedia content are the communication messages
of blogging, while speech is the way of communication for public speaking, meetings,
small groups, and dyadic situations. Blogging is also distinctive from the writing tasks
that most writing anxiety researchers have studied. Although writing is required for the
majority of blogging posts, for many personal diary bloggers blogging is not a task that
users have to ﬁnish in a certain amount of time like many writing tasks. Blogging is a

voluntary activity. Moreover, the format of blog posts is different from previous writing

43

types. There are no fixed rules, standards, or requirements for blog posts. A single word
or a sentence can be a blog post.

Blogging is also different than other CMC applications. Compared to emails, chat
rooms and MUDs, blogs are a comparatively new format of Internet communication tool.
Blogs are the only online genre that combines personal media with public attention.

Thus, blogging differs from the situations in which communication apprehension
arises in that the most common sources of communication apprehension—speeches and
face-to-face evaluations—are not present. It differs from situations in which writing
anxiety is found in that the strict standards or requirements for various writing tasks are
not applied to blog posts. Finally it differs from other forms of CMC as examined by
CMCA because blogs are personal media with public attention.

Blogs are personal media. A blogger performs various roles on his/her blog—
writer, editor, publisher, graphic designer, discussion monitor, and web administrator. For
other CMC applications, it is very rare that a person will take so many responsibilities.
Moreover, the major focus of blogging is the blogger—his/her life stories and
perspectives—while the message is the center of other CMC applications.

This personal medium gains public attention. In many applications of CMC,
information senders communicate with certain receivers. To put it in other words, they
have their target audience; they only deliver their information to the people they want to
reach. For example, individuals only put email addresses of the people they want to reach
in the address bar when they compose emails. Individuals chat with a speciﬁc person or
persons in chat rooms or through instant messengers. In CMC, the audience has a

boundary. In blogging, however, there are usually no ﬁxed boundaries of audience,

unless a blogger only distributes the URL of his or her blog to people he or she knows.
Theoretically, information published on blogs can reach an infinite number of people.
Bloggers cannot selectively target their information to a speciﬁc audience. It is true that
some blogging software allows more control over audience. For example, blogs can be
password protected. However, most bloggers have little control over their audience. This
argument is supported by previous literature (Bortree, 2005; Gumbrecht, 2004; Lenhart,
2005; Li, 2005; Nardi et al., 2004a; Park, 2003a; Takhteyev & Hall, 2005 ; Viegas, 2005;
Wijnia, 2004). This is the reason that scholars suggested that blogging is a “radio show”
(Scheidt, 2006; Sundar et al., 2007), a personal medium with public attention.

The integration of personal media with public attention generates anxiety, which
is named as online posting anxiety. Again, online posting anxiety is the joint effect of two
separate factors, visceral arousal and cognitive self-labeling of the internal state as
anxiety in blogging. It is a situational, state-like variable.

Aside from understanding blog-maintenance intentions, an examination of online
posting anxiety will make an important original scientiﬁc contribution to social cognitive
theory. That one, single most important contribution is applying the social cognitive
explanation of anxiety to CMC situations.

Bandura started to examine the Internet and human agency (Bandura, 2002). His
research is a general description of the importance of self-efﬁcacy and self-regulation on
individual educational and occupational self-development, organizational productivity,
societal digital gaps and health implications, and social changes in the digital age when

new media are ubiquitous. These descriptions could not answer the questions: What are

45

some unique characteristics of the Internet? How do agentic factors function in online
environments?

Internet researchers have advanced our knowledge about how social cognitive
variables function on the Internet. Intemet-related self-efﬁcacy scales were developed
(Compeau & Higgins, 1995; Eastin & LaRose, 2000; Torkzadeh & van Dyke, 2001).
Self-efﬁcacy and outcome expectations were proposed as predictors of various Internet
application usages and effects (Bani Ali, 2005; Chou, 2001; Eastin & LaRose, 2005; Hsu
& Chiu, 2003; LaRose & Eastin, 2004; LaRose et al., 2001; LaRose et al., 2005; LaRose
et al., 2003; Song et al., 2004). Computer/Internet anxiety was connected with
computer/Intemet self-efﬁcacy, but only at a correlation analysis level (Chua et al., 1999;
Durndell & Haag, 2002; Maurer, 1994; Sam et al., 2005). In short, no previous study has
examined the process and dynamics of how various cognitive variables can generate
anxiety in online situations.

3.13 Culture

Hofstede organized cultural differences into overarching patterns. He identiﬁed
ﬁve primary dimensions of culture differences: power distance, individualism,
masculinity, uncertainty avoidance, and long-term versus short-term orientation
(Hofstede & Hofstede, 2005). The individualism/collectivism dimension is the most
widely applied framework to investigate culture issues. Hofstede deﬁned individualism
and collectivism on a societal level. He suggested that individualism “pertains to societies
in which the ties between individuals are loose: everyone is expected to look after himself
or herself and his or her immediate family.” Collectivism, on the other hand (p. 51),

“pertains to societies in which people from birth onwards are integrated into strong,

46

cohesive in-groups, which throughout people’s lifetime continue to protect them in
exchange for unquestioning loyalty” (Hofstede, 1991). Later research modiﬁed these two
constructs to encompass both characteristics of culture and personality attributes. Other
terms have been used to investigate the distinctions between cultures at individual levels.
For example, independence and interdependence (Markus & Kitayama, 1991),
idiocentrism and allocentrism (Triandis, 1995), and the private self and collective self
(Traﬁmow et al., 1991). Oyserman, Coon, and Kemmelmeier (2002) concluded that the
core elements of individualism/independence/idiocentrism/private self are personal
uniqueness and independence, whereas duty to the in-group and maintaining harmony are
the main constituents of collectivism/interdependence/allocentrism/collective self
(Oyserman et al., 2002). Usually, researchers use the individualism/collectivism
framework to contrast Western and Eastern cultures.

The individualism/collectivism dimension inﬂuences individuals’ cognition and
emotion. Triandis found that individuals’ self-presentation strategies are inﬂuenced by
culture. He made a distinction between the private self (the assessment of self by the self),
the public self (the assessment of self by a generalized other), and the collective self (the
assessment of self by a particular reference group). He argued that culture will inﬂuence
self-selection and sampling. In individualistic cultures, the private self tends to be more
complex and more salient than the collective self. In collectivistic cultures, the collective
self tends be more salient than the private self. The self-selection and sampling determine
self-presentation. In individualist cultures, people tend to have more individualist (or
interdependent) self-representations; in collectivist cultures, they have more collectivist

(or interdependent) self-representations (Triandis, 1989). Markus and Kitayama (1991)

47

showed that different types of self-representations are likely to lead to different types of
cognitive, emotional, and motivational consequences (Markus & Kitayama, 1991).
Adamopoulos and Kashima concluded that culture is an antecedent of psychological
processes, in which self-representations act as a mediator of cultural inﬂuences on
cognitive, affective, and motivational processes (Adamopoulos & Kashima, 1999).

Attribution is also a function of culture. Fiske et a1. indicated that Western
individuals perceive behavior primarily as a function of personal attributes and
dispositions, such as emotions. However, East Asian individuals see situational factors,
such as norms, roles, and obligations, as the major determinants of behavior (Fiske et al.,
1998).

The Chinese culture is an interdependent-based culture, which is characterized by
collectivistic social patterns whereby people are linked closely in groups and act in
accordance with each other to a great extent (Bagozzi & Lee, 2002). The Chinese culture
stresses “we” consciousness, collective identity, emotional interdependence, group
decision making, reciprocity, and the maintenance of social harmony (Watson & Ebrey,
1991). In the Western world, an individual is an independent entity with free will,
emotions, and a personality. In the Chinese culture, the self represents a tendency for a
person to act in accordance with external expectations or social norms, rather than with
internal wishes or personal integrity (Yang, 1981; Sun, 1991). In the Confucian paradigm,
the traditional Chinese self is deﬁned by relations with others. A person in this relational
network tends to be sensitive to his or her position above, below, or equal to others (Gao,
1998). In this Confucian social environment, social norms will have great inﬂuence on

people’s behavior, which of course includes media behavior.

48

A number of studies have shown that cultural perceptions about different
computer-related technologies are key factors related to both the initial acceptance of
these technologies as well as future behavior regarding their usage (Albirini, 2005; Collis,
1999; Doostdar, 2004; Karen et al., 2003; Li et al., 2001; Loch et al., 2003; Makrakis,
1992; McCoy, 2002; McCoy et al., 2005; Parboteeah et al., 2005). In fact, the impact of
cultural, social, and environmental factors on technology use and adoption has long been
emphasized in diverse theoretical perspectives, such as social cognitive theory (Bandura,
1986), diffusion of innovation (Rogers, 2003), critical mass (Markus, 1990), and social
inﬂuence (Fulk et al., 1987). In social cognitive theory, normative inﬂuences are covered
by outcome expectancies that focus on the perceived social consequences of behavior
(Norman & Conner, 1996). In his theory of diffusion of innovation, Rogers identiﬁed the
social system, the social context in which the innovation diffuses, and norms of the social
system, the established patterns of behavior that tell members of the system what
behavior is expected, as two important parameters in the process of innovation adoption,
which inﬂuences individuals’ awareness, interest, evaluation, trial, and adoption of
innovations (Rogers, 2003). Thomas believed that cultural suitability, which deﬁned as
how well the proposed innovation ﬁts the existing culture, would inﬂuence how
acceptable a new technology will be in a society (Thomas, 1987). In his theory of
planned behavior, Ajzen raised the concept of the subjective norm, which is deﬁned as
perceived social pressure to perform or not to perform the behavior, as an important
determinant of intention and/or behavior (Ajzen, 1991).

These differences might be expected to evince themselves in blog-maintenance

intentions in the following way. Individualism/collectivism as a dimension of culture will

49

affect individuals’ blogging self-efﬁcacy, outcome expectations, and online posting
anxiety. Speciﬁcally, vicarious experience and social persuasion are two sources of self-
efﬁcacy, while individualism/collectivism dimension of culture is reﬂected in vicarious
experience and strategies and content of persuasion in a given culture. People’s outcome
expectations are decided by self-efﬁcacy and past experience, which is also tightly
connected with the culture dimension. As to online posting anxiety, literature reviewed in
the previous section has provided abundant evidence for the inﬂuence of culture on
anxiety.

Social cognitive theory underemphasized the individualism/collectivism
categorization of cultures—“People’s efforts to manage their everyday lives cannot be
reduced to polarities that arbitrarily partition human agency into individual and collective
forms”(Bandura, 2002, p.270). This does not indicate that the theory ignores cultural
differences. Social cognitive theory indeed recognizes culture differences. Bandura
believed that basic agentic capacities and mechanisms of operations are common across
nature, but how they are used varies between cultures. Using efﬁcacy beliefs as an
example, he suggested that “although efﬁcacy beliefs have generalized functional value,
how they are developed and structured, the ways in which they are exercised, and the
purposes to which they are put, vary cross-culturally” (Bandura, 2002, p. 273).

Integrating the social cognitive view of culture, this research proposes that
blogging self-efﬁcacy, blogging outcome expectations, and online posting anxiety will
have impacts on blog-maintenance intentions for both American bloggers and Chinese
bloggers. However, their differing cultures will produce two types of inﬂuence on blog-

maintenance intentions. First, the content of beliefs about blogging self-efﬁcacy, outcome

50

expectations, and online posting anxiety will vary between cultures. Social related
blogging self-efﬁcacy, outcome expectations and online posting anxiety may be more
salient among Chinese bloggers than American bloggers. Second, the amount of impact
these factors will have on blog-maintenance intentions will be different between cultures.
Negative outcome expectations and online posting anxiety may have a greater impact on
Chinese bloggers than on American bloggers.

To summarize, in this chapter, literature about blogging, anxiety, and culture was
reviewed. The literature review serves as the bedrock for the development of the
hypotheses and a model that I will present in the following chapter. But it is one thing to
summarize literature and another to synthesize literature into a researchable problem. The
speciﬁc null ﬁndings and inconsistencies and paradoxes that might be explained through
the social cognitive theoretical perspective are as follows: (1) Anxiety can be deﬁned in
social cognitive terms. For example, communication apprehension can be explained as a
combination of low communication self-efﬁcacy and negative outcome expectations for
communication. Similarly, a social cognitive view of online posting anxiety suggests that
online posting anxiety is predicted by blogging self-efﬁcacy and outcome expectations.
(2) Blogging is jointly decided by the nature of blogs and the bloggers’ personal factors.
The personal factors that are important for blogging are social anxiety, writing anxiety,
blogging self-efﬁcacy, blogging outcome expectations, and online posting anxiety. Blogs
are a unique online communication genre (Herring et al., 2004a). Blogs form a context
that is different from traditional social, communication, and writing situations and also
from other CMC applications. The unique nature of blogging indicates that online posting

anxiety, blogging self-efﬁcacy, and blogging outcome expectations are unique concepts.

51

(3) Situational factors can be incorporated into trait views to improve the predictability of
trait variables. (4) Depending on the breadth and stability of the environment, variables
can function as a general trait-like variable, a generalized-context one, a speciﬁc-context
one, and a situational state-like variable. The nature of the situation decides the
characteristics of the variables. In this research, social anxiety and writing anxiety
function as generalized—context variables, and online posting anxiety as a situational
state-like variable. (5) Cognition determines anxiety. The direct causal relationship,
however, only presents at the same level of generality. For example, it is writing self-
efﬁcacy and outcome expectations, not general self-efﬁcacy and outcome expectations or
more speciﬁc blogging self-efﬁcacy and outcome expectations, that directly predict
writing anxiety. (5) Culture inﬂuences cognition and anxiety. The triadic causal
relationship between environment, behavioral, and personal factors is a universalized
model across individualistic and collective cultures, but how it is used varies in different

cultural milieus.

52

CHAPTER 4

HY POTHESES AND RESEARCH QUESTIONS

Based on social cognitive theory and the literature of anxiety, blogs, and culture,
the current research proposes a structural model to explain the determinants of blog-
maintenance intentions. This structural model suggests that cognitive factors (blogging
self-efﬁcacy, positive outcome expectations for blogging, negative outcome expectations
for blogging) and anxiety (social anxiety, writing anxiety, and online posting anxiety)
inﬂuence blog-maintenance intentions. The model integrates three different theoretical
perspectives—social cognitive theory and the two conﬂicting views of the causation
between anxiety and cognition. The contribution of this model is that it uses social
cognitive terms to explain anxiety and provides a unique perspective to integrate these
discrepant theories to explain blog-maintenance intentions.

There are paradoxical arguments in the social cognitive view of the relationship
between cognition and anxiety. On one hand, Bandura ﬁrmly posited that cognition
determines anxiety (Bandura, 1986). On the other hand, Bandura mentioned that one
source of self-efﬁcacy is affective arousal (Bandura, 1997). When discussing sources of
self-efﬁcacy, Bandura argued that people rely partly on their somatic and emotional
states in judging their capabilities. A way of modifying self-beliefs of efﬁcacy is to

reduce people’s stress reactions and alter their negative emotional proclivities and

53

interpretations of their physical states (Bandura, 1997). Social cognitive theory failed to
provide an explanation of these conﬂicting arguments.

This research proposes a model that may untangle the paradoxical views. In line
with Bandura’s basic argument, this study suggests that cognition predicts anxiety. The
relationship between cognition and anxiety, however, is not a single equation. It is, as
demonstrated in Figure 4.1, a linear process in which cognition and anxiety interact at
different levels of generality. Generality is deﬁned as the scope of the contexts a variable
can cover. The broader the scope of the coverage, the higher level the generality has. For
example, writing anxiety and writing self-efﬁcacy cover all the writing contexts, while
online posting anxiety and blogging self-efﬁcacy contain only blogging contexts. The
level of generality of writing anxiety and writing self-efﬁcacy is higher than that of
online posting anxiety and blogging self-efﬁcacy.

The causal relationship between cognition and anxiety is decided by the
generality of the cognitive and anxiety variables. In the same level of generality,
cognition inﬂuences anxiety. On different levels, higher-level anxiety directly inﬂuences

lower-level cognition.

54

Figure 4.1. The Causal Relationship between Cognition and Anxiety.

 

General Cognition

 

 

 

 

 

General Anxiety

 

 

 

 

 

Speciﬁc Cognition

 

 

 

 

I

 

Speciﬁc Anxiety

 

 

 

Previous studies provided partial support for this model of cognition and anxiety.
Chen et al.’s (2001) empirical study conﬁrmed Bandura’s argument that general self-
efﬁcacy inﬂuences speciﬁc self-efﬁcacy. Their model can be presented as Figure 4.2.
This model is not in conﬂict with the model proposed in this study. In the current study,
general cognition indirectly predicts speciﬁc cognition through general emotion.
Omitting general emotion in Figure 4.2 does not threaten the validity of our model. It
only decreases the explanatory power of Chen et al.’s model.

Figure 4.2. The Causal Relationship between General Cognition and Speciﬁc Cognition.

 

General Cognition

 

 

 

 

 

Speciﬁc Cognition

 

 

 

Studies also examined the relationship between general anxiety and speciﬁc
anxiety (Figure 4.3). For example, Brown et al. (2004) suggested that computer anxiety

and communication apprehension indirectly determine computer-mediated

55

communication anxiety. Similarly, previous studies examined the relationship between
general anxiety and speciﬁc cognition (See Figure 4.4). For example, Thatcher and
Perrewe (2002) found that trait anxiety inﬂuenced computer self-efﬁcacy. Again, these
models did not falsify the model proposed in this study. Instead, discarding important
cognition/anxiety variables is a limitation of these models.

Figure 4.3. The Causal Relationship between General Anxiety and Speciﬁc Anxiety.

 

General Anxiety

 

 

 

 

 

Speciﬁc Anxiety

 

 

 

Figure 4.4. The Causal Relationship between General Anxiety and Speciﬁc Cognition.

 

General Anxiety

 

 

 

 

 

Speciﬁc Cognition

 

 

 

Thus, previous studies presented part of the picture of the reciprocal process of
cognition and anxiety. The model proposed in this study shows a more comprehensive
picture. It is not a complete picture because the model starts with general anxiety (Figure
4.5). The general proposition of this model is that general anxiety variables will ﬁrst

inﬂuence speciﬁc cognitive variables, which in turn will inﬂuence speciﬁc anxiety

56

variables. These three kinds of variables (general anxiety variables, speciﬁc cognitive
variables, and speciﬁc anxiety variables) will determine intention.

Figure 4.5. The Causal Relationship Examined in This Study

 

Social Anxiety, Writing Anxiety

 

 

 

 

 

Blogging Self-Efﬁcacy, Blogging
Outcome Expectations

 

 

 

 

 

Online Posting Anxiety

 

 

 

This is a unique contribution because, unlike all the research that preceded it, it
speciﬁes the process of how cognition and anxiety interact at different levels of
generality. Social cognitive theory keeps emphasizing the role of situations. But it does
not provide a general solution in terms of how to incorporate situations in the variables.
In this study, the role of situations is reﬂected in the concept of generalit. Speciﬁcally,
this model suggests that within the same level of generality, cognition inﬂuences anxiety.
Nevertheless, before individuals enter a new level, their anxiety at the higher level serves
as a reference point or anchor for their cognitive beliefs in this level. The model hopes to
resolve the following theoretical issues: (1) how cognition and anxiety with different
levels of generality interact for blogging, a unique online communication behavior; (2)
how high-level anxiety inﬂuences speciﬁc cognitions and anxiety; and (3) how culture

demonstrates its inﬂuence on blogging.

57

4.1 The Role of Culture

While a great deal of work has been done in terms of studying blogging in the
U.S., few studies have examined other countries’ blogging practices. Even fewer studies
have conducted cross-cultural research that compares the blogging practices of two
countries. Bandura clearly stated that agentic capabilities are common across cultures
(Bandura, 2002). This research proposes that:

H1: blogging self-efﬁcacy, positive outcome expectations for blogging, negative
outcome expectations for blogging, and online posting anxiety will predict blog-
maintenance intentions for both American and Chinese bloggers.

The social environments for blogging are different when looking at American
bloggers and Chinese bloggers. Blogs are arguably one of the most individualized media,
while Chinese culture is considered one of the cultures that value collectiveness most.
China is also has some of the strictest Internet censorship among countries in the world.
The differences between blogging contexts will elicit inﬂuence on blogging cognition
and anxiety. However, the directions and variances of the inﬂuence are still in doubt. A
research question is proposed

R1: What are signiﬁcant differences between the Chinese model and the
American one?

Previous media use studies used intention as their dependent variable (Agarwal &
Karahanna, 2000; Dabholkar & Bagozzi, 2002; Davis et al., 1989). Intention can be
deﬁned as one’s motivation in the sense of his or her conscious plan to exert effort to
carry out behavior by himself or herself (Eagly & Chaiken, 1993). In this study, blog-

maintenance intention is used as the dependent variable.

58

4.2 The Impact of General Anxiety Variables

Writing anxiety is an important anxiety variable that will inﬂuence blogging-
speciﬁc cognitive factors because bloggers use texts as communication syntax. Social
anxiety should be salient, too. The interactions between bloggers and their audience
imply that blogs can be seen as social stages.

Two sources of self-efﬁcacy are the somatic and emotional states (Bandura, 1986).
People can gauge their degree of self-efﬁcacy by the emotional state they experience as
they contemplate an action (Bandura, 1997). Thus, anxiety, as an affective response,
might have a direct inﬂuence on self-efﬁcacy beliefs (Barbeite & Weiss, 2004). Strong
emotional reactions to a task also provide cues about the anticipated success or failure of
the outcome (Bandura, 1997). Anxiety can inﬂuence outcome expectations.

Computer anxiety may not be important here. Leso and Peck (1992) found that
computer anxiety was more related to computer programming than it was to general
Internet tools and applications. Scott and Timmerman (2005) further suggested that
computer anxiety could not capture the use of new communication technologies. When
bloggers think of blogging, two of the most familiar and relevant anxiety types—writing
anxiety and social anxiety—will be salient because blogging is text-based and the
interaction between bloggers and audience is an important component of blogging.
Communication apprehension was not included in the model based on the results of
Study 1 (see Chapter 5 for detailed discussion).

Writing and social cues are present in blogging behaviors. These two sources of
anxiety will serve as the cues for bloggers to judge their blogging self—efﬁcacy, positive

outcome expectations for blogging, and negative outcome expectations for blogging.

59

H2a: Writing anxiety will decrease blogging self-efﬁcacy.

H2b: Writing anxiety will decrease positive outcome expectations for blogging.
H2c: Writing anxiety will increase negative outcome expectations for blogging.
H3a: Social anxiety will decrease blogging self-efﬁcacy.

H3b: Social anxiety will decrease positive outcome expectations for blogging.
H3c: Social anxiety will increase negative outcome expectations for blogging.
4.3 The Impact of Self-Efficacy and Outcome Expectations

Self-efﬁcacy is important for Internet usage (Barbeite & Weiss, 2004; Compeau
& Higgins, 1995; Eastin & LaRose, 2000; Huang & Liaw, 2005; Igbaria & Iivari, 1995;
McFarland & Hamilton, 2006; Wang et al., 2003). Social cognitive theory proposed that
self-efﬁcacy will inﬂuence outcome expectations since outcomes are driven largely by
judgments of how well one can perform the behavior (Bandura, 1986). Studies found that
computer/Intemet self-efﬁcacy will increase computer/Internet outcome expectations
(Compeau & Higgins, 1995; Compeau et al., 1999; LaRose & Eastin, 2004). Self-
efﬁcacy was also found to make signiﬁcant contributions to intention (Manstead & Van-
Eekelen, 1998).

Bandura deﬁned anxiety as a joint effect of two separate factors, visceral arousal
and cognitive self-labeling of the internal state as anxiety. From its deﬁnition, we can see
how cognition detemrines anxiety. Empirical studies have indeed found negative
correlations between self-efﬁcacy and anxiety (Chua et al., 1999; Dumdell & Haag, 2002;
Dwyer & Fus, 2002; Hopf & Colby, 1992; Maurer, 1994; Prickel, 1994; Sam et al., 2005).

Evidence was also found that self-efﬁcacy decreased anxiety (Crumbo, 1999; Dwyer &

60

Fus, 2002; Greene & Sparks, 1983; Hopf & Colby, 1992; Lucchetti et al., 2003; Prickel,
1994).

For blogging, bloggers who are conﬁdent in their blogging skills and capabilities
will experience more positive outcome expectations while anticipating fewer negative
outcome ones. They will be less anxious when blogging. Their blog-maintenance
intentions will be higher than those of bloggers who are not conﬁdent in their blogging
skills.

H4a: Blogging self-efﬁcacy will increase positive outcome expectations for blogging.
H4b: Blogging self-efﬁcacy will decrease negative outcome expectations for bloggin g.
H4c: Blogging self-efﬁcacy will decrease online posting anxiety.

H4d: Blogging self-efﬁcacy will increase blog-maintenance intentions.

Outcome expectations inﬂuence intention and anxiety (Bandura, 1986; Compeau
et al., 1999; Crable et al., 1994; Greene & Sparks, 1983; LaRose & Eastin, 2004; LaRose
et al., 2001; Littlejohn, 1994). People’s intention to use information technology is
inﬂuenced by their beliefs about the relative advantages and disadvantages of the
technologies (Al—Khaldi & Al-Jabri, 1998; Culpan, 1995; Hebert &_ Benbasat, 1994).
Mikkelsen et al. argued that whether the introduction of a new technology is perceived as
a challenge or triggers anxiety will depend on the individual expectancies of the
outcomes of available strategies in any given situation (Mikkelsen et al., 2002). If
bloggers perceive that blogging will bring positive outcomes, they will increase their
intention to blog and decrease their anxiety. On the other hand, if they believe that
blogging will bring negative outcomes, their intention will be decreased and their anxiety

will be increased.

61

H5a: Positive outcome expectations for blogging will decrease online posting anxiety.
HSb: Positive outcome expectations for blogging will increase blog-maintenance
intentions.
H6a: Negative outcome expectations for blogging will increase online posting anxiety.
H6b: Negative outcome expectations for blogging will decrease blog-maintenance
intentions.
4.4 The Impact of Online Posting Anxiety

The impact of anxiety on intention is well supported (Allen & Bourhis, 1992,
1996; Barbeite & Weiss, 2004; Beatty, 1987; Burroughs et al., 2003; Campbell & Neer,
2001; Chelley & Larry, 2002; Chua et al., 1999; Daly & Miller, 1975b; Richmond &
McCroskey, 1998; Scott & Rockwell, 1997; Smith & Caputi, 2001; Vician & Davis,
2002; Weil & Rosen, 1995; Wheeless et al., 2005). Researchers found that computer
anxiety usually causes computer use avoidance (Chua et al., 1999; Weil & Rosen, 1995).
For blogging, it is assumed that the more anxious people feel about blogging, the more
they will try to avoid blogging.
H7: Online posting anxiety will decrease blog-maintenance intentions

In summary, seven variables, one research question, and 16 hypotheses are
proposed to examine the factors that contribute to blog-maintenance intentions. To
demonstrate the direct and indirect theoretical causal structure as hypothesized, the
following path model is proposed (Figure 4.6).

While most of these hypotheses have been investigated before, they are included
here to provide a comprehensive model of blog-maintenance intentions. The unique,

original contribution of the present research is located in Hypotheses H1, R1, H2a-c,

62

H3a—c, and H4a-d. Speciﬁcally, the contribution is that the present research provides a

social cognitive explanation of online posting anxiety in the CMC realm.

63

        

bursa use;
35

  

 

 

 

 
 

.2thmeme am:
05350 cm:
o>=omoz
om:
av: <
coco—mount}. 30.53 i
-moﬁ mczaon 25.5 A ow: >ooo_tm.=om 9:305 been?

       
  

room

em:
3:

 

pm:

  

mﬁumoﬁ
co.— ocozoaoooxm
oEooSO 03:qu

82325 855638-35 co Soctm «8:65 can H085 .Lo Eco—2 323.505. 6.». Ilmluxx .E

CHAPTER 5

METHODS

5.1 Study 1
5.1.1 Method

In this dissertation, three studies were conducted to test the proposed model.
Study 1 involved qualitative in-depth interviews. Interviewing provides a way for
researchers to understand the experience of other people and the meaning they make of
an experience. The in-depth interview approach uses open-ended questions to build upon
and explore participants’ responses to those questions (Seidman, 1998). The goal of
Study 1 is to examine the nature of blogging and to generate measures for the four new
constructs in the model: blogging self-efﬁcacy, positive outcome expectations for
blogging, negative outcome expectations for blogging, and online posting anxiety.

Participants were recruited by messages posted on the author’s and her friends’
blogs. The interviewees were selected by nonrandom (nonprobability) sampling. To
recruit bloggers who might experience online posting anxiety, nonactive bloggers were
the desired sample. In Herring et al.’s study (2004), all blogs whose latest update was
older than 14 days were eliminated, to examine only active blogs. Perseus’s blog survey
(2003) also discovered that the average updating cycle of active blogs is 14 days. In this

dissertation, the length of the nonactive period was expanded. The recruitment messages

65

speciﬁed that only bloggers who had not updated their blogs for more than three months
were qualiﬁed to participate.

Ten American bloggers and 16 Chinese bloggers were interviewed online using
instant message tools (MSN Messenger, Google Talk, and AOL Messenger). Of the 10
American bloggers, ﬁve were female and ﬁve were male. Of the 16 Chinese bloggers,
nine were female and seven were male. The interviews lasted between 55 to 180 rrrinutes.

Mason suggested that asking personalized questions can generate situated
knowledge with all the interviewees (Mason, 2002). Before each interview, the
interviewee’s blog was examined to design speciﬁc questions related to his/her blogging
experience. The open-ended interviews were conducted under the guidance of an
interview protocol. The sequence of the questions, however, was not ﬁxed. American
participants were given $20 Amazon gift cards for participation. Chinese participants
were paid by mailing RMB40 checks. The interview scripts were analyzed using a
modiﬁed constant comparative technique (Strauss & Corbin, 1998). The data analysis
took place during and after data collection to assist in designing follow-up interviews.
The entire corpus of data was examined using a grounded research approach thoroughly
after the interviews were completed.

5.1.2 Brief Results

Table 5.1 summarizes American and Chinese bloggers’ blogging self-efﬁcacy,
blogging outcome expectations, and online posting anxiety beliefs. It should be noted that
American and Chinese bloggers shared similar beliefs. However, there were salient
cultural differences. Compared to American bloggers, Chinese bloggers believe that

blogging is a more difﬁcult task that requires more skills. They are more afraid of the

66

negative health outcomes and audience-related outcomes, while they value their blogging
status and audience more. Audiences and social concerns are their biggest sources of
online posting anxiety.

Another ﬁnding worth mentioning is that communication-related anxiety was not
identiﬁed by either American or Chinese bloggers. It was hypothesized that
communication-related anxiety exists because blogging is a form of communication.
However, the data indicated that bloggers do not worry about communication anxiety.
This might be explained by the special nature of blogs. Text is the medium of
communication for blogging. Instead of worrying about speeches, bloggers are anxious
about writing. This might be the reason that writing-related anxiety, not communication-
related anxiety, is salient for blogging.

Table 5.1. American and Chinese Bloggers’ Blogging Beliefs

 

 

American Bloggers Chinese Bloggers
Blogging self- 0 Computer and Internet 0 Computer and Internet
efﬁcacy beliefs 0 Writing Writing

0 Observation Observation

OOO

Moral, communication,
English, persistence

Outcome 0 Enjoyable activity 0 Enjoyable activity
expectation for 0 Social 0 Social
blogging beliefs 0 Status 0 Status
0 Self-reactive o Self-reactive
0 Economic
0 Health
0 Government censorship
Online Posting 0 Social 0 Social
Anxiety beliefs 0 Writing 0 Writing
0 Computer and Internet 0 Computer and Internet
0 Safety 0 Safety
0 Time and environment

 

67

5.2 Study 2
5.2.1 Method

Based on the results of the qualitative in-depth interview study, a pretest-posttest
online survey was conducted to develop scales to measure blogging self-efﬁcacy, positive
outcome expectations for blogging, negative outcome expectations for blogging, and
online posting anxiety. A snowball sampling method was used to recruit participants. A
recruitment message was posted on the author’s, her friends’, and some of the interview
participants’ blogs. In the recruitment message, the author invited bloggers to spread the
information and post this message on their own blogs.

For the American sample, there were 87 pretest respondents. The mean time since
their last update of their blogs was 1.59 months (SD = 3.59). They were paid $3 for the
pretest survey and $5 for the posttest survey. The sample was 69% female and 31% male
and the participants’ mean age was 32.16 (S_I_)_ = 10.84) years old. The majority of
respondents were Caucasian (55.2%). Other ethnic groups represented include African
American (9.2%), Asian (29.9%), and Other (10.3%). The majority of respondents had
four-year college degrees (31%), followed by some college (27.6%), master’s degree
(16.1%), two-year college degree (9.2%), high school diploma (8.0%), doctoral degree
(4.6%), and unidentiﬁed (3.4%). Fifty-eight of the 87 pretest respondents participated in
the posttest, a 66.7% response rate.

For the Chinese sample, there were 70 pretest respondents. The mean time since
their last update of their blogs was 1.42 months (ﬂ; = 3.13).They were paid RBMlO for
the pretest survey and RMB20 for the posttest survey. The sample was 49% female and

43% male and 8% unidentiﬁed. Participants’ mean age was 32.36 (SD = 13.07) years old.

68

Of the Chinese participants, 1.4% had ﬁnished middle school, 14.3% had ﬁnished high
school, 51.4% had a college degree, 24.3% had a master’s degree, and 1.4% had a
doctoral degree. Fifty-three of 70 pretest subjects participated in the posttest survey two
weeks later, a 75.7% response rate.
5.2.2 Brief Results

The purpose of this longitudinal study was to test the reliability and validity of the
measures. Based on item evaluation criteria, a six-item blogging self-efﬁcacy scale, a
ﬁve-item positive outcome expectations for blogging scale, a ﬁve-item negative outcome
expectations for blogging scale, and a three-item online posting anxiety scale were
developed and tested.

The internal consistency reliability was estimated using Cronbach’s coefﬁcient
alpha. The coefﬁcient alpha is a commonly quoted statistic that provides support that a
scale is acceptable. Table 5.2 presents the Cronbach’s coefﬁcient alpha for the new
measures. The values of Cronbach’s coefﬁcient alpha ranged from .610 to .757. The new
measures demonstrate acceptable reliability. The test—retest reliability was calculated.
Table 5.3 demonstrates the values of test-retest reliability. The test-retest reliability
values were from .684 to .828, indicating that the measures are stable over time.

Table 5.2. Cronbach’s Coefﬁcient Alpha of the New Measures

 

 

American Chinese Sample
Sample

Blogging self-efﬁcacy .757 .754

Positive outcome expectations .663 .704

for blogging

Negative outcome expectations .745 .610

for blogging

Online posting anxiety .749 .629

 

69

Table 5.3. Test-Retest Reliability of the New Measures

 

American Sample Chinese Sample

 

Blogging self—efﬁcacy .743 .684
Positive outcome .828 .818
expectations for blogging
Negative outcome .688 .783
expectations for blogging
Online posting anxiety .789 .771

 

Thus, the Cronbach’s coefﬁcient alpha and test-retest reliability were calculated
to access reliability. The content validity, criterion-related validity, and construct validity
were examined as well. The majority of the results suggested that these new scales

demonstrated accepted reliabilities and validities.

5.3 Main Study
5.3.1 Sampling Method

Participants were recruited by messages posted on blogging online groups.
Previous blogging studies also employed online postings of the survey URL and
announcements to recruit participants (Johnson & Kaye, 2004; Kaye, 2004, 2007).
Nonprobability sampling is an acceptable method when random sampling is not possible
(Babbie, 2001).

American respondents consisted of 145 bloggers. They were given $5 for their
participation. The sample was 71.3% female and 28.7% male, and the mean age was
34.14 (S_D = 9.57) years old. The majority of respondents were Caucasian (64.8%). Other
ethnic groups represented included African American (6.2%), Asian (16.6%), and Other
(4.8%). When asked about who reads their blogs, 33.1% believed that their audience was
mostly people they personally know; 28.7% thought that their audience was mostly

people they have never met; 29.4% answered both equally, and 8.8% didn’t know who

70

reads their blogs. Regarding to the question “whom they blog for,” 38.2% answered that
their blogs were mostly for themselves, 14% indicated that their blogs were mostly for
their audience, 44.1% believed that their blogs were both for themselves and their
audience, and 2.9% answered that their blogs were not for themselves or their audience.

Chinese respondents consisted of 178 bloggers recruited by messages posted on
blogging online groups. They were given RBM20 for their participation. The sample was
51.2% female and 48.8% male and the mean age was 26.96 (§I_) = 6.34) years old.
Among these bloggers, 44.9% believed that mostly people they personally knew read
their blogs, followed by 16.9% of the participants stated that they didn’t know who read
their blogs. 15.2% indicated a readership of both equally (people they personally know
and people they have never met), and 10.7% who believed the blogs were read mostly by
people they have never met. When asked about “overall, would you say you blog...”
55.6% answered “mostly for yourself,” followed by 16.3% “mostly for your audience,”
2.8% “both equally,” and 8.4% “others.”
5.3.2 Evaluations of the Reliability and Validity of the New Scales

Criterion validity and reliability were investigated again to evaluate the new
scales. Individual items for the blogging self-efﬁcacy scale, the positive outcome
expectations for blogging scale, the negative outcome expectations for blogging scale,
and the online posting anxiety scale were correlated with the dependent variable of this
study: blog-maintenance intentions. Based on social cognitive theory, blogging self-
efﬁcacy items and positive outcome expectations for blogging items should be positively
correlated with blog-maintenance intentions, while negative outcome expectations for

blogging and online posting anxiety items should be negatively correlated with blog-

71

maintenance intentions. Based on the criterion validity evaluation, a few items were
deleted from the four scales. The remaining scales demonstrated criterion validity and
reliability. See Table 6.1 and Table 6.2 in the next chapter.

To further investigate the reliability of the new scales and the equivalence of the
scales across cultures, two principle factor analyses were conducted. Tables 5.4-5.7
present the results of the factor analyses for the American sample and Chinese sample.
The blogging self-efﬁcacy, positive outcome expectations for blogging, negative
outcome expectations for blogging, and online posting anxiety scales demonstrated
similar factor structures for the American and the Chinese samples. The equivalence of
these four scales across culture was supported.

Table 5.4. Blogging Self-Efﬁcacy Factor Loadings

 

 

American Sample Chinese Sample
Items Factor Loading Factor Loadigg
Write a well-organized post with good
. . . .858 .706
introduction, body, and conclusron.
Using multimedia to decorate my blog. .807 .749
Usrng a blog to present my unique .803 .667
personality.
Collect information that attracts my .793 .745
audience 1nterests and attentron.
Make friends with people I don’t know

.779 .786

at all on my blog successfully.
Be persrstent and write at least one post .636 .544

every week.

 

72

Table 5.5. Positive Outcome Expectations for Blogging Item Loadings

 

American Sample Chinese Sample

Items Factor Loading Factor Loading
Examine, rnterpret, and remember the .890 .636
world and my life.
Feel good about myself. .851 .815
Satisfy my desire to showcase my writings,

. . .772 .716
musrc, pictures, or other works.
Tighten connectrons wrth ofﬂine and onlrne .762 .713
frrends.
Get attention. .665 .506

 

Table 5. 6. Negative Outcome Expectations for Blogging Item Loadings

 

 

American Sample Chinese Sample
Items Factor Loading Factor Loading
Bring negative employment
consequences like dismissals and .878 .842
reprimands.
Jeopardize my relationships. .876 .748
Fee] shame because my weakness or my 809 .735

private life might be exposed.

Harm my health. .740 .650
Have physrcal danger if people know .681 .690
where I live and work.

 

Table 5. 7. Factor Loadings for Online Posting Anxiety Items

 

American Sample Chinese Sample

 

Items Factor Loading Factor Loading
I feel uncomfortable when I think my blog
is going out to the world and on a .860 .840

permanent record.
I am afraid the emotions shown on my

.847 .723
blog are too temporary.
A constant awareness in the back of my
mind of my audience causes me to feel .812 .744
tense.

 

73

5.4 Structural Equation Modeling Test

5.4.1 Operational Measures

Writing apprehension Daly and Miller’s writing apprehension test (WAT) is the
most widely used scale to measure writing-related anxiety (Daly & Miller, 1975a).
However, scholars have questioned the construct validity of the WAT. After reviewing
the literature, Poff concluded that most instruments used to measure writing apprehension
actually measure hypothesized causes of these constructs, not the constructs themselves
(Poff, 2004). Studies argued that the WAT may measure variables other than writing
apprehension. For example, it may be inﬂuenced by self-conﬁdence in the ability to write
(Richmond & Dickson-Markman, 1985), attitudes toward writing (Reed & Keeley, 1986),
transitory/situation-speciﬁc anxiety (Fowler & Knoll, 1980), self-efﬁcacy and outcome
expectations (McKain, 1991), positive emotions (McKain, 1991), and writing enjoyment
(Davida et al., 1995). In the current study, the WAT items were examined, and the items
lacking construct validity were eliminated. Studies found that the WAT measure other
variables other than writing apprehension (Poff, 2004). For example, self-conﬁdence in
the ability to write (Richmond & Dickson—Markman, 1985), attitudes toward writing
(Reed & Keeley, 1986), transitory/situation-speciﬁc anxiety (Fowler & Knoll, 1980),
self-efﬁcacy and outcome expectations (McKain, 1991), positive emotion (McKain,
1991), and writing enjoyment (Davida et al., 1995). In this study, ﬁve valid items were

used to measure writing apprehension.2

 

2 We would like to ask about your experience in your real life when you write. Please indicate the degree to
which each statement applies to you by marking whether you: I have no fear of my writing’s being
evaluated; I’m nervous about writing; I look forward to writing down my ideas (N); Taking a composition
course is a very frightening experience.

74

Social anxiety Four out of 24 items from Liebowitz’s social anxiety scale
(Liebowitz, 1987) were used to measure social anxiety. Items are scored on a seven-point
Likert scale ranging from “strongly disagree” (1) to “strongly agree” (7).3

Blogging self-efﬁcacy Six items were used to assess blogging self-efﬁcacy. Items
were scored on a seven-point Likert scale ranging from “no conﬁdence at all” (1) to
“complete conﬁdence” (7). Total scores were obtained by summing over all the items and
taking the mean; higher scores indicated greater perceived blogging self-efﬁcacy.

Positive outcome exmctations for blogging. Positive outcome expectations for
blogging were measured by a seven-point Likert scale consisting of ﬁve items, ranging
from “strongly disagree” (1) to “strongly agree” (7).

Negative outcome exp_e£tations for blogging. Negative outcome expectations for
blogging were also measured by a seven-point Likert scale consisting of ﬁve items,
ranging from “strongly disagree” (1) to “strongly agree” (7).

Online posting anxiety Three items were used to measure online posting anxiety.
Items were scored on a seven-point Likert scale ranging from “strongly disagree” (1) to
“strongly agree” (7). Total scores were obtained by summing over all the items and
taking the mean; higher scores indicated greater online posting anxiety.

Blog-maintenance intentions. Blog-maintenance intentions was assessed using
three items adapted from Armitage and Conner (1999). Each item was measured on a

seven-point Likert scale (1 = strongly disagree to 7 = strongly agree).4

 

3 I feel anxious when... Writing while being observed; Talking with people you don’t know very well;
Meeting strangers; Expressing a disagreement or disapproval to people you don’t know very well.

4 I will try to update my blog in the future; I want to update my blog in the future; I intend to update my
blog in the future.

75

5. 4. 2 Analysis

The Statistical Package for the Social Sciences (SPSS) version 14.0 was used to
analyze the data. Pearson product-moment correlations were computed to test the
hypotheses at the .05 level of signiﬁcance. Experiment-wise error rate (BER) might be
present. It is worth noting that one or more of the signiﬁcant test results may be
signiﬁcant just by chance. However, EER is not necessarily a problem. O’Keefe (2003)
proposed that the practice of adjusting the alpha level to control for experiment-wise
error should be abandoned because it has no principled basis, commits one to absurd
beliefs and practices, and reduces statistical power. Tutzauer (2003) argued that alpha
adjustment should be applied only in the narrowly circumscribed instance when the
researcher wants to make a strong claim that there is no Type I error in a speciﬁc
collection of tests. Yuan-Bentler’s model was applied for dealing with missing data in the
EQS procedure. For examining the causal relationships among variables under the
proposed model, the EQS program (Bentler, 1995) was employed. Each study variable
was a latent variable with one indicator formed by the relevant items. The error terms for

each of the indicators was computed according to the formula e = variance * (1 — alpha).

76

CHAPTER 6

RESULTS

The Pearson product-moment correlation coefficients between variables are
presented in Table 6.1 and Table 6.2. The ﬁgures in the diagonal are the reliability alpha
coefﬁcients. The means and standard deviations for each variable are shown in the
columns to the right of the table. It should be noted that in the US group, the correlations
between positive outcome expectations for blogging and social anxiety and blogging self-
efﬁcacy were very high (r=-.597, r=.641). Multicollinearity might present. VIF
(Variance Inﬂation Factor) and Tolerance values were calculated in regression analysis
using SPSS 14.0 to check for multicollinearitys. The effect of multicollinearity was
excluded in the current study.

To examine causal relationships between blog-maintenance intentions and other
variables, a structural model was constructed. Following Raykov, Tomer, and
Nesselroade’s ( 1991) recommendation on reporting goodness of ﬁt indices, normed ﬁt
index (NFI), nonnorrned ﬁt index (NNFI), and comparative ﬁt index (CFI), as well as the
misﬁt measure known as root-mean-square error of approximation (RMSEA) suggested
by Hu & Bentler (1999), indications for acceptable ﬁt were provided by ﬁt indices that

exceed 0.90 and misﬁt RMSEA index under or equal to 0.07.

 

5 The criteria that any VIF of 10 or more and/or tolerance values of .10 or less (Cohen et al. 2003, p. 422)
was applied to examine multicollinearity.

77

Model 6.1, the model with blog-maintenance intention as the dependant variable
constituted a good fit to the data. 12 (df = 25, N = 323) = 55.18, p=.000, NFI = .909,
NNFI =. 910, CFI =. 946, RMSEA =. 061. Yuan-Bentler correction model indicated that:
12 (df = 25, N = 323) = 48.404, p=.003, NFI = .914, NNFI =. 924, CF12. 955, RMSEA

=. 054. Parameter estimates are shown in Figure 6.1.

As depicted in Figure 6.1, 13 out of 16 hypotheses were supported in both the
American and Chinese samples. Speciﬁcally, writing anxiety has signiﬁcant direct impact
on blogging self-efﬁcacy (the American sample: B: -.527; the Chinese sample: B: -.522),
positive outcome expectations for blogging (the American sample: B=-.426; the Chinese
sample: B=-.416), and negative outcome expectations for blogging (the American sample:
B=-.241; the Chinese sample: B=-.244). Social anxiety has signiﬁcant direct impact on
positive outcome expectations for blogging (the American sample: B: -.215; the Chinese
sample: B: -.178), negative outcome expectations for blogging (the American sample:
B=.253; the Chinese sample: B=.217). Blogging self—efﬁcacy had signiﬁcant direct impact
on positive outcome expectations for blogging (the American sample: B=.566; the
Chinese sample: (3:.558), negative outcome expectations for blogging (the American
sample: B=—.253; the Chinese sample: B=-.259), and online posting anxiety (the American
sample: B: -.546; the Chinese sample: B: -.587). Positive outcome expectations for
blogging had signiﬁcant direct impact on online publishing anxiety (the American
sample: B=.323; the Chinese sample: 8:352), and blog-maintenance intentions (the
American sample: B=.817; the Chinese sample: B=.804). Negative outcome expectations
for blogging had signiﬁcant direct impact on online posting anxiety (the American

sample: B: .562; the Chinese sample: [3: .591). Online posting anxiety had signiﬁcant

78

direct impact on blog-maintenance intentions (the American sample: [3: -.269; the

Chinese sample: B: -.242).

Additionally, writing anxiety had signiﬁcant indirect impact on blog-maintenance
intentions through positive outcome expectations for blogging (the American sample: B:
-.426, B: .817; the Chinese sample: [3: -.416, B: .804), blogging self—efﬁcacy and online
posting anxiety (the American sample: B: -.527, B: -.546, [3: -.269; the Chinese sample:
B: -.522, B: -.587, B: -.242), and negative outcome expectations for blogging and online
posting anxiety (the American sample: B: .241, B: .562, B: -.269; the Chinese sample:
[3: .244, B: .591, B: -.242). Social anxiety had signiﬁcant indirect impact on blog-
maintenance intentions through positive outcome expectations for blogging (the
American sample: B: -.215, B: .817; the Chinese sample: B: -.178, B: .804), and
negative outcome expectations for blogging and online posting anxiety (the American
sample: B: .253, B: .562, B: -.269; the Chinese sample: [3: .217, B: .591, B: -.242).
Blogging self-efﬁcacy (the American sample: B: -.546; the Chinese sample: B: -.587),
positive outcome expectations for blogging (the American sample: B: .323; the Chinese
sample: B: .352), and negative outcome expectations for blogging (the American sample:
B: .562; the Chinese sample: B: .591) had signiﬁcant indirect impact on blog-
maintenance intentions through online posting anxiety (the American sample: [3: -.269;

the Chinese sample: B: -.242).

However, three paths were not signiﬁcant for both samples. Speciﬁcally, the path
from social anxiety to blogging self-efﬁcacy (H3a) was not signiﬁcant. The path from

blogging self-efﬁcacy to blogging-maintenance intentions (H4d) and the path from

79

negative outcome expectations for blogging to blogging-maintenance intentions (H6b)

were not signiﬁcant.

The results of the group comparison showed that there were signiﬁcant
differences between the betas of the paths for H3b (social anxiety to positive outcome
expectations for blogging, p=.030), H4c (blogging self-efﬁcacy to online publishing
anxiety, p=.070), H4d (blogging self-efﬁcacy to blog-maintenance intentions, p=.000),
and H5a (positive outcome expectations for blogging to online publishing anxiety,

p=.073).

To examine the possibility that demographic differences between the Chinese and
American samples could explain differences in blog-maintenance intentions, correlational
analyses were performed, shown in Table 6.3. Gender was treated as a dummy variable
and further correlated with other variables. The results suggested that gender correlated
with social anxiety in the American sample.

Table 6.3. Gender’s Correlations with Other Variables in both Samples

 

American sample Chinese sample

 

Social anxiety -.204(*) .098
Writing anxiety -.075 .051
Blogging self-efﬁcacy .023 .021
Posrtrve outcome expectatrons for .039 _.053
bloggrng

Negative outcome expectatrons for .073 .000
bloggrng

Online postrng anxrety _.1 10 .049
Blog-maintenance intentions —.066 -.069

 

To further examine the effects of gender, a structural equation model was

developed which incorporated gender as a control variable. The results suggested that

80

gender did not show a signiﬁcant inﬂuence. After controlling for gender, the path
coefﬁcients, ﬁx indexes only had very tiny changes. Thus, gender differences between
the two samples were unlikely to explain the differing results.

In addition to the model showed above, rival theories suggest alternate models.
For example, theories of social anxiety and writing anxiety have no role for outcome
expectations and self-efﬁcacy. This is depicted in Model 6.2 (Figure 6.2). In this model,
writing anxiety and social anxiety predict online posting anxiety directly. The model
showed that 2'2 (df = 14, N = 323) = 67.207, p=.000, NFI = .890, NNFI =. 720, CFI
=.907, RMSEA 2. 109. Yuan-Bentler correction model indicated that: 12 (df = 14, N =
323) = 54.588, p=.000, NFI = .909, NNFI =. 781, CFI =.927, RMSEA =. 095. This was
not a good ﬁtting model.

Social Cognitive Theory suggests an alternate causal ordering among anxiety
and self-efﬁcacy in which anxiety precedes self-efﬁcacy and outcome expectations and
has no direct relationship to intention. This was tested through Model 6.3, shown in
Figure 6.3. In this model, writing anxiety and social anxiety inﬂuenced online posting
anxiety directly. Online posting anxiety then predicted blogging self-efﬁcacy and
outcome expectations. The model did not ﬁt. ,‘(2 (df = 20, N = 323) = 145.727, p=.000,
NFI = .762, NNFI =. 537, CFI =.779, RMSEA =. 140. Yuan-Bentler correction model
indicated that: 12 (df = 20, N = 323) = 131.113, p=.000, NFI = .780, NNFI =.580, CFI
=.800, RMSEA =. 132.

Another possibility is that anxiety precedes outcome expectations. That is
because efﬁcacy beliefs will inﬂuence anxiety. Self-efﬁcacy and anxiety will then predict

outcome expectations. This was tested through Model 6.4 in Figure 6.4. The model

81

showed 12 (df = 18, N = 323) = 94.736, p=.000, NFI = .845, NNFI =. 686, CFI =.865,
RMSEA =. 115. Yuan-Bentler correction model indicated that: 12 (df = 18, N = 323) =

76.825, p=.000, NFI = .871, NNFI =.753, CFI =.894, RMSEA =.101. This was a worse
ﬁt comparing to the model with blogging self-efﬁcacy and blogging outcome
expectations beliefs preceding online posting anxiety.

Finally, in the interest of parsimony, we must consider whether online posting
anxiety is a necessary construct. In ﬁgure 6.5 we present a model (Model 6.5) in which

outcome expectations and self-efﬁcacy are directly connected to blog-maintenance
intentions in which other anxiety measures directly cause intentions as well. 12 (df = 2,
N = 323) =4 .476, p=.128, NFI = .999, NNFI = 1.050, CFI =1.000, RMSEA =. 000.
Yuan-Bentler correction model indicated that: 12 (df = 2, N = 323) = 4.383, p=.139, NFI
= .999, NNFI =1.056, CFI =1.000, RMSEA =.000.

Kline suggested two ways to compare ﬁt indexes of models: values and

signiﬁcances of 2'2 index and values of lzldf. Low and nonsigniﬁcant values of the 12

are desired. And the value of Zzldf should be smaller than 3 (Kline, 1998). Based on
these criteria, Model 6.5 was a worse ﬁtting model comparing with the model containing
online posting anxiety (Model 6.1) because it had fewer degree of freedom and higher
2’2 /df values.

Akaike Information Criterion (AIC) is another goodness-of-ﬁt measure which
can be used to compare models with differing numbers of latent variables. AIC adjusts

model chi-square to penalize for models lacking of parsimony and overpararneterization.

The lower AIC reﬂects the better-ﬁtting model (Bumham & Anderson, 1998).

82

Thus, ﬁve models were tested. The original proposed model with blog-maintenance
intentions as the dependent variable was still the best ﬁt model. Table 6.4 present the ﬁt

indexes of these models.

Table 6.4. Fit Indexes of Structural Models.

 

 

12 (if 1ng p AIC

Model 6.1 48.404 25 1.936 .003 -l.596
Model 6.2 54.588 14 3.899 .000 26.588
Model 6.3 131.113 20 6.556 .000 91.113
Model 6.4 76.825 18 4.268 .000 40.825
Model 6.5 4.383 2 2.191 .139 -3.617

 

83

 

mom.—

com;

mmm._

“um—.—

8m;
w:._
pm;

Om

mama

mooN

Elam

owoé

mmwe

oowd

wood
:32

005.

5225::
8:58:88

-woa

. . . . ftmwv. .- 25:52:
FLOO— - chzmm - chNmm fivmom . Avwvmoﬁ oogcoucﬁerw2m
Se. 56%. 8a.. 9955.. ...Vmem. from? 58.5.
wagon cam—:0

wEwwo—n

mmw. Nmor ooor fvmom. Atnom. 5m 5235098

0:535 033ch
$23.03

5m 523055
0:585 03:5;

ﬁmw. A**:3.A**Vhomu. TILWNNH

. 9.83 m. . 38%

@0w 1 fcvaNm .. rH—Ow wcmwwAzm

oww. $.54. 58.5 use?

5mm. 58:8 58m
.3258 wEwon 5m 93on 5a .3550 .9258 .9258
wagon 2538895 25380095 .28 went? 38m

0:585
gamma Z

0:585 wEmon
035$

955

 

055mm Snow—08¢. 05 5“ 333.2; 53m 5m macaw—050885 EB .wcﬁauwm bzﬁmzom £5533 “5:28am .532 . No Swish

84

 

 

0mm:

in:

90m:
ave:
CNN:
moo:

:m:
Om

wEw

hmﬁm

mad
m .56
mmmé
comm

Nona
0002

50:00::
00005508 5552

3

w. fewvmcm. Whor

one. 93.39.

35.

-020 650.0

frenzom.

mwor

cmo.

vow.

050050
03:52

Tcwvoov. TcLNmVr

£02.- ftoom.

02. owe.
Titan. fervmmvr

as. 0.054.-
mam.

30:50 50205098 50205090 50050200 30:50
0:535
0>s0w0z

wEwwoa watt?

. 50:00::

000 - 000000508505
$3..ng 50:50
2:502 002:0

. 500000095

NE 005050 0>20w0z

. 502000090

2o 050050 03:52

. 2002.20

mo: - .200 wEwon

02. 060.5. 050.5
4%. 08:0 068
30:50
38m

 

0EE0m 805:0 05 5.: 0030E0> >55 5.: 5020—0500025 0:0 50085.2 3:30:32 £52038 E02505 .5002 Nd 030k

85

                   

mQ .nwgghumzoz

..0. 50230006
050050

0300002

.0... a
-----------------

       

0000250080.... 5:
0000:2505
.m0_m

guwsx ..QQwNSm

30.52
95000 05.00

wwm.u~5- ..8M.H~z0-

         
        

>08_=m.=om 05000.0

 

08.n080 0.3550
@5305

.00 50280003

050050 033000

  

1 ”500% 500 5 502.55%

All: ”000.5 .Z.U 05 5 50050me

TI. 0:20 .50 05 5 0530000

03000> 080530 20 0%. H20

00 555005 00:0:0555-m05 ﬁe 3002 <00. ENG. .804 ”m3

86

33x5

9.5.;

 

 

 
   

     
   

 

        

 

 

 

 

   
   
    

  

  

my: . mwlvn
am. \ can? “Ewan.
mmm. m3 uXX .8 2.035033
olrxx 2.83 ,
2c. 3592
IE. 6,2
XXX xmoor
mmvwnaz ..Rm.u~5z 02.. 8?st ”Shauna T
scum—”Ham“? NQNV Eggs 4 a waV MQQ."NMDQ .WQQ."NZUQ
-uoﬁ 9.53.. 2.2.5 >ouo_tm.=om mﬁmmoﬁ
\ mp.
SP Rm.
“mm. own. Senna: ”Swans:
R3 ﬁlm. 9.82..
«MAI. a .8 30350096
x «E350 2;sz .
>553 AI ”8:on Eon E “gamma—Em
Ecom

I ”@320 .20 05 E Emoﬁawmm
I x505 .m: 05 a 385:me
.5252 wagon 25:0 8c. HZU

855%: 2325 5252 mg; 2:. 3252 38m 3 382 3 mod “m:

87

 

       
       

32x5 9.2.:

 

Ewngz ”Sank:

.8 2.2.8033
2:350
95¢qu

           

          

N§.HNWD“ ..NOM."NZU~

occur—25¢:—
.msm

gawk: ..mmmuNEz

38:2
9.68 25.5

      

a: «use ..nS,.n~zux
>035»
£8 9.32m

3252
300.0.

      

   
    

Khnwsx “Sums:
9.595

.2 acozﬁooaxm
2.32:0 2,an .

  

AI ”350% Eon E Haaoumawﬁ
All... ”@380 .Z.U 05 E “58$:me
AIIII ”@280 .m.D 05 E anmeﬁ

8%. H20

20:8on 830on >632 wagon 25:0 m6 382 g coo. ”m3

      

88

 

30:23 95:3

      
         
         

.6. Vacanﬁihumx
050230
03592

   
           

gowns: KENS:
3352
953a 05:5

." MD .. ." ZU
envwsz ”Shaka 2» N 2 ex. N x

moan-.0235:
boa

 

a

  

avgumusz ﬂaws:
..2 2.0230096
3.08:0

02:3“.

1&9.on Son 5. EmoﬁEwE
Al ”9.80 .20 05 E Emu—bawﬁ
I “9.80 .m.D 2: E EmoEcw_m
8c. HZU
olodu ”m:

mcosﬁoumxm 080850 8385 525?. wagon 25:0 v.0 .322 g

20.52
300m

 

89

            
          
      
 

 

aqumiuitu 5..
.2 302800an
2.38:0

       

.I b . .l
5. -m&_..$=._..£
coca—50:35:.— a u a
62m . - - - -

VMMHNEM ..Numﬂmzuk ................................ 32x5
300$
635?...“ 9.53. n

 

   

Rows: Kuhn»...
9.595

.2 2.03500an

2:350 «>52...

AIII ”anew Eon E .585:me
AI ”95.0 .Z.U on. E .585:me
3252 3.8. 25.5 505.3 .082 no .082 .2. Elam... anew w: 2.. 5 .awmmﬁww

ooo. ”m3

9O

CHAPTER 7

DISCUSSION

Guided by social cognitive theory and related literature, this dissertation sought to
develop a better theoretical explanation for predicting American and Chinese bloggers’
blog-maintenance intentions. The goals of this research were: (1) establish evidence for
the existence of online posting anxiety; (2) develop valid and reliable scales to measure
blogging self-efﬁcacy, positive outcome expectations for blogging, negative outcome
expectations for blogging, and online posting anxiety; (3) empirically test the four new
scales, (4) investigate the relationships between more general anxiety variables, speciﬁc
blogging cognition, and speciﬁc blogging anxiety and their effects on blog-maintenance
intentions; and (5) examine the inﬂuence of culture on blog-maintenance intentions. The
unique contribution of this research was to employ the social cognitive theoretical
perspectives to examine anxiety in the CMC realm.

To achieve these goals, three studies were conducted. Study 1 applied in—depth
interview methods to understand American and Chinese bloggers’ salient beliefs of self-
efﬁcacy and outcome expectations and to explore the existence and characteristics of
online posting anxiety. Study 2 employed a pretest and posttest survey design to establish
the Cronbach’s coefﬁcient alpha, test-retest reliability, content validity, criterion-related
validity, and construct validity of the new scales. The Main Study used an online survey
to assess the 16 hypotheses and the structural model.

Overall, results from this dissertation research provided evidence for the existence

of an online posting anxiety construct. The results of Study 2 showed that online posting

91

 

anxiety demonstrated moderate correlations with relevant anxiety measures with
correlations ranging from .281 to .586. The Main Study provided further evidence of
concurrent and predictive validity. Online posting anxiety again showed moderate
correlations with writing anxiety, social anxiety, and blog-maintenance intentions.

To summarize, the blogging self-efﬁcacy, positive outcome expectations for
blogging, negative outcome expectations for blogging, and online posting anxiety scales
demonstrated good reliabilities and validities. The proposed structural model was
supported. Cultural inﬂuences on blogging were discovered. In this chapter, the
discussions of the model will be presented, followed by research implications, future
research areas, and limitations.

7.1. The Role of General Anxiety Variables

Trait perspectives have been criticized for neglecting the impact of a situation on
behavior. This criticism does not discount the trait perspective. Perhaps equally important
is the issue of how trait variables could contribute to the social cognitive model. The
structural model in this dissertation employed a new strategy to develop a trait-situation
congruity theoretical model. Two blogging-relevant anxiety types (social anxiety and
writing anxiety) were employed. These situation-relevant anxiety concepts were treated
as distal inﬂuences on blog-maintenance intentions with the blogging-speciﬁc cognitive
and anxiety variables as mediators. The results suggested that writing anxiety showed a
signiﬁcant impact on blogging self-efﬁcacy as well as positive and negative outcome
expectations for blogging. Social anxiety had a signiﬁcant impact on positive and
negative outcome expectations for blogging. This ﬁnding added to social cognitive theory.

Social cognitive theory has been criticized for underemphasizing trait perspectives.

92

Instead of abandoning trait perspectives, this model integrated trait and social cognitive
perspectives. Speciﬁcally, the model indicated that people’s general tendencies to be
uneasy in relevant situations (social and writing) are personal factors that reﬂect an
interaction between P (personal factors) and E (environment factors). So instead of
positing vague and general principles of the triad causations among P (personal
factors)——E (environment factors)—B (behavioral factors), this research speciﬁed a type
of PE that is relevant to intention.

Writing anxiety was a powerful independent variable in the model, with its
signiﬁcant impact on all the three blogging cognitive variables. It should be noted that
writing is not required for blogging. Many blogs only consist of pictures, links, or cited
articles. Bloggers can bypass writing and use other multimedia content to blog. Blogs,
however, are an online text communication genre. Other multimedia content can help to
achieve the desired goals of bloggers. To really beneﬁt from blogging, people have to
employ writing. That’s why writing anxiety was important in our research.

Social anxiety signiﬁcantly predicted positive and negative outcome expectations
for blogging. Its impact on blogging self-efﬁcacy, however, was not signiﬁcant. The
results of Study 2 may provide explanations as to why social anxiety did not predict
blogging self-efﬁcacy. The ﬁndings suggested that the majority of bloggers think that
blogging is a simple task. Socially anxious people become tense when an environment
provides threatening stimuli. When socially anxious people think about blogging, they
might not experience visceral arousal because they may not anticipate threats to their
efﬁcacy beliefs. This might explain why social anxiety does not predict blogging self-

efﬁcacy.

93

To summarize, the general anxiety variables demonstrated mixed ﬁndings. Their
predictabilities varied. Writing anxiety was an important predictor. Five items of WAT
with good validity were selected to measure writing anxiety. The exclusion of
confounding items improved writing anxiety’s predictability. Social anxiety predicted
both positive and negative outcome expectations for blogging. A tentative conclusion—
general anxiety variables inﬂuence speciﬁc blogging cognitions—may be drawn.

A more important task for future studies is to ﬁnd more variables that are
connected with relevant situations or classes of situations for which they have predictive
implications, like writing anxiety in this model. For example, fear of negative evaluation
should be relevant for blogging. Fear of negative evaluation is deﬁned as “apprehension
about others’ evaluations, distress over their negative evaluations, avoidance of
evaluative situations and the expectation that others would evaluate oneself negatively”
(Watson & Friend, 1969). The results of Study 2 and the Main Study suggested that
audience is an important source of blogging outcome expectations and online posting
anxiety. The presence of audiences implies evaluations. Fear of negative evaluations may
be a source of concern for bloggers. However, before employing the measure of fear of
negative evaluations, scholars should check its validity because its deﬁnition is not
consistent with negative outcome expectations and behavioral avoidance.

7.2 The Role of Blogging Self-Efficacy and Outcome Expectations for Blogging

Self-efﬁcacy and outcome expectations are important variables that inﬂuence
people’s intention. In this study, a blogging self-efﬁcacy scale, positive outcome
expectations for blogging scale, and negative outcome expectations for blogging scale

were developed and tested.

94

For both the American sample and the Chinese sample, blogging self-efﬁcacy had
a signiﬁcant impact on positive outcome expectations for blogging, negative outcome
expectations for blogging, and online posting anxiety. The results provided another
support for social cognitive theory. The current model supported the view that those
bloggers who are conﬁdent about their blogging skills will conceive higher positive
outcome expectations for blogging, lower negative outcome expectations for blogging,
and lower online posting anxiety. Following this argument, a signiﬁcant positive path
from blogging self-efﬁcacy to blog-maintenance intentions was anticipated. Contrary to
the hypothesis, however, the path from blogging self-efﬁcacy to blog-maintenance
intentions was a nonsigniﬁcant one.

Bloggers generally perceive blogging as a simple task that does not require
speciﬁc skills. Blog websites provide blogging software and professional-looking design
templates that make the process of blogging easy. Nevertheless, there are difﬁcult skills
required for high-end blogging, for example, modifying the existing templates, designing
a technically savvy blog, and using a blog to present a unique personality. But difﬁcult
skills usually do not serve as a threshold for basic blogging. A minimal understanding of
word processing software and Internet access can be enough for people to blog. To be a
good blogger, however, blogging self-efﬁcacy is important. This argument was partly
supported by Marlow’s social network analysis of blogs. He randomly sampled blogs
using an aggregator, ﬁnding a strong relationship between investment in the blog and
payoff in terms of audience size and feedback. His data showed that blog communities
reward bloggers who put time into their work (Marlow, 2006b). It is reasonable to

hypothesize that the more time bloggers invest, the more skills they can master and the

95

higher the self—efﬁcacy levels they can achieve. Put another way, blogging self-efﬁcacy
might not be related to blog-maintenance intentions after people set up a blog. It should
be related to how well people blog. This might be the reason that blogging self-efﬁcacy
predicted positive outcome expectations for blogging and negative outcome expectations
for blogging and online posting anxiety, but not blog—maintenance intentions. Future
studies may apply other dependent variables that relate to how well people blog, such as
blogging achievement and the size of the audience, to examine the effects of blogging
self—efﬁcacy.

Additionally, blogging self-efﬁcacy generality can be a variable for further
analysis. Bandura deﬁned self-efﬁcacy generality in terms of how well judgments of
efﬁcacy for one task transfer into self-efﬁcacious behavior of other tasks similar in nature
(Bandura et al., 1980). Blogging can be seen as multiple tasks: blog writing, blog reading,
and blog commenting. In the current research, the newly developed blogging self-
efﬁcacy measure focuses on blog writing self-efﬁcacy. To test the generality of this
measure, its correlations with blog writing, blog reading, and blog commenting were
examined. Blogging self-efﬁcacy was signiﬁcantly correlated with all three blogging
behaviors in both samples. The correlations ranged from .208 to .287 (p < .01). The
generality of the blogging self-efﬁcacy scale was supported. The scale can be used in
future studies about blog reading or commenting.

The blogging outcome expectations construct is divided into positive outcome
expectations for blogging and negative outcome expectations for blogging. The two types
of outcome expectations showed mixed results to their determinants—online posting

anxiety and blog-maintenance intentions. The path from positive outcome expectations

96

for blogging to blog-maintenance intentions was signiﬁcant. Contrary to our hypothesis,
however, it had a positive inﬂuence on online positing anxiety.

It should be noted that the sign of Pearson product-moment correlation between
positive outcome expectations for blogging and online posting anxiety was negative (r = -
.020 in the U.S. sample and r = -.085 in the Chinese sample). Statistically, the ﬂipping of
the signs in the structural equation model might be caused by multicollinearity and
suppression effect (Cohen et al., 2003; Tzelgov & Henik, 1991). VIF (variance inﬂation
factor) and tolerance values were calculated in regression analysis using SPSS 14.0 to
check for multicollinearity (See Table 7.1). Cohen et al. (2003, p. 422) posited that a
commonly used rule of thumb is that (1) any VIF of 10 or more and/or (2) tolerance
values of .10 or less provide evidence of serious multicollinearity involving the
corresponding independent variable. The effect of multicollinearity was excluded in the
current study.

Table 7.1. VIF and Tolerance Value of Factors

 

 

 

Factors Tolerance VIF

American Chinese American Chinese
Sample Sample Sample Sample

Writing anxiety .677 .704 1.476 1.421

Social anxiety .512 .902 1.953 1.108

Blogging self-efﬁcacy .51 1 .632 1.955 1.582

POS’HV" .‘mtwme . .453 .685 2.205 1.460

expectatlons for blogging

Negauv‘? outcome . .824 .645 1.214 1.551

expectanons for blogging

Online posting anxiety .580 .542 1.725 1.845

 

Assuming Y is a dependent variable and X 1 and X 2 are two independent

variables, Cohen et al. (2003) stated that suppression is present when either r“ or r,2 is

97

less than the product of the other and r12 , or when r12 is negative (assuming positive r“

and r”):
ry1< ryzx’iz (1)
ry2< ry1xr12 (2)

If Equation 1 or Equation 2 holds, the partialed coefﬁcients of X , and X 2 will be

larger in value than the zero-order coefﬁcients and one of the partialed coefﬁcients may
become negative (Cohen et al., 2003, p. 77).

Suppression was found using Cohen et al.’s (2003) equations in the American
sample. There was suppression among blogging self-efﬁcacy, positive outcome
expectations for blogging, online posting anxiety, and blog-maintenance intentions.
Moreover, negative suppression was identiﬁed in the current model (Maassen & Bakker,
2001; Tzelgov & Henik, 1991). The suppression effect was the reason for the positive
path from positive outcome expectations for blogging to online posting anxiety.

Negative outcome expectations for blogging had a signiﬁcant impact on online
posting anxiety. The path from it to blog-maintenance intentions was not signiﬁcant.
Thus, both positive and negative outcome expectations for blogging generated complex
inﬂuences on online posting anxiety and blog-maintenance intentions. In general,
scholars have suggested that, when using technology, communicative outcomes are
emergent from people’s interactions with the technology and others with whom they are
communicating (Waskul & Douglass, 1997). In the case of blogging, outcomes are
jointly decided by bloggers and audiences.

The existence of audiences is part of what makes a blog a unique medium.

Audiences are one of the major forces that shape blogging outcomes; self-presentation

98

exists because there are audiences to present the self to; audiences are the friends
bloggers want to make and the relationship bloggers manage to maintain; audiences are
the source of the social support and attention bloggers want. Additionally, an unknown
and unwanted audience can lead to social pressures, safety concerns, embarrassment,
privacy invasions, and other negative outcomes. Thus, the outcome expectation measure
used here may have been incompletely speciﬁed.

Previous literature suggested diverse blogging gratiﬁcations. LaRose et al. applied
social cognitive theory to frame the uses and gratiﬁcations perspective. They believed
that media gratiﬁcation can be seen in terms of the media user’s behavioral outcome
expectations (LaRose et al., 2001). Following LaRose’s argument, previous literature
identiﬁed diverse blogging outcome expectations. One ﬁnding of Study 1 was that
bloggers’ outcome expectations were very diverse. Trammell suggested that blogs are
truly individual and are what the bloggers make them (Trammell, 2005). Li suggested
that blogs serve a variety of ends for people. Bloggers can consciously satisfy their
unique motivation by taking advantage of the characteristics of blogs as an open and free
social venue for intrapersonal, interpersonal, and mass communication. The use of blogs
is systematically associated with an individual’s motivation for blogging (Li, 2005).
Blogs do not ﬁt into a single method of communication as other communication
technologies do. Depending on bloggers and audiences, a blog could act as intrapersonal
communication, one-to—many communication, one-to—one communication, or group
communication. Different communication modes generate different communication
outcome possibilities. Blogs could accommodate various outcomes for all communication

modes. This is a challenge for the blog outcome expectation concept and measure

99

development. In this study, in the process of blogging outcome expectation measure
development, 139 items were generated from the results of Study 1 and a brief content
analysis of newspapers’ portrayal of blogs. However, the ﬁnal scales of positive outcome
expectations for blogging and negative outcome expectations for blogging had only ﬁve
items. It should be noted that most culturally speciﬁc items were removed to make the
scales equivalent across cultures. This strategy improves generalization of the scales, but
it decreases the power of the scales. This is a trade-off that blogger researchers may have
to face when they conduct blog research.

One possible cause of the heterogeneity of the outcome expectation beliefs in this
research is that most of the interviewees in Study 1 were diarists. Only one interviewee in
the Chinese sample was a journalist. And two other interviewees in the American sample
kept K-blogs. Compared to other blog types, personal—joumal-type bloggers may have
more diverse outcome expectations. It should be noted that it is exactly this population
that has attracted more and more blog motivation researchers.

The concept of modally salient beliefs, which was proposed by the theory of
planned behavior, may provide a solution. The theory of planned behavior suggested that
the most frequently mentioned beliefs in the pilot sample constitute the set of beliefs
salient in the population, named modally salient beliefs (Ajzen, 1991). Scholars further
extended the construct of modally salient beliefs to incorporate heterogeneity in the
construct. Bel argued that certain behavior is mainly an individual experience and that
belief structures are expected to be highly idiosyncratic. Modally salient beliefs may have
limitations to be applied to the behavior because few beliefs are shared among the sample.

The author suggested that heterogeneity should be modeled along with modally salient

100

beliefs. In other words, Bel applied heterogeneity as another dimension of modally salient
beliefs. Bel used the discrepancy between the most important outcome beliefs and
modally salient beliefs to measure the degree of heterogeneity (Bel, 2004). Bel’s
approach may be used in future blogging research. Blogging modally salient beliefs and
the degree of heterogeneity may provide explanatory power for blog-maintenance
behavior.

The heterogeneous nature of blogging outcome expectations concepts can be an
advantage for blogging research as well. First of all, blogging outcome expectations
should be a powerful factor in predicting blogging behavior, since bloggers purposefully
employ different communication modes to blog. This argument was supported by the
structural model. Positive outcome expectations for blogging was the most powerful
predictor of blog-maintenance intentions. Secondly, previous literature in interpersonal
communication, group communication, and mass communication can be examined to
help develop and deﬁne the blogging outcome expectations concept.

The concept of negative outcome expectations for blogging was not a signiﬁcant
predictor of blog-maintenance intentions. Generally, when behavior has been associated
with negative expectations, people may be forced to choose between withdrawing from
the behavior and engaging in the behavior and expecting negative outcomes. It is possible
that the bloggers with negative outcomes self-selected from the author’s samples. They
were not as reachable through the websites from which participants were recruited. This
is an important limitation of this study.

It is also reasonable to suggest that bloggers accept the risks associated with

blogging and blog anyway. Perhaps the negative outcomes are simply an unavoidable

101

consequence of blogging. Bortree suggested that bloggers carved out a blogging
community for themselves, a community in which they believe they can express their
opinions and thoughts freely without fear of interference. On the other hand, by sharing
as they do, they make themselves vulnerable (Bortree, 2005). Blogging is a special
practice in that every advantage can lead to a disadvantage. For example, an audience can
bring friendship and social support. It can also bring social pressure and
misunderstanding. Recording one’s life and releasing one’s feelings can generate
happiness. It can also encourage privacy invasions and information misuse. Bloggers are
mindful of this feature of blogging. Ada, an interviewee for Study 1, has two blogs at
different blogging websites. She mentioned the advantages and disadvantages of both
blogs:

For my xxx blog the one where I promote my writing: advantages - satisfying a

desire to showcase writing and having a large number of people reading it;

disadvantages - can’t share with friends, completely anonymous... For my

personal private journal: advantages - the ability to write whatever without

judgment; disadvantages - inability to share with friends.

There are also compromises that bloggers have to make. Another interviewee,
Mike, is an artist, who displays his artwork on his blog. He worried that “someone uses
my image without permission,” however, he is “willing to take a risk to have my work
seen.” Jason puts his blog link on his portfolio page. He admitted that, “I have to be
careful to not post things that might be detrimental to my career. However, I don’t want it
to be sterile and boring.”

Every coin has two sides. The tight connections between positive outcome

102

 

expectations for blogging and negative ones lead bloggers to accept the possibilities of
negative outcomes and still blog. Their desire for the positive outcomes outweighs their
perceptions of the negative ones. This might explain the insigniﬁcant path from negative
outcome expectations for blogging to blog-maintenance intentions. It also serves as
evidence for the existence of online posting anxiety.

7.3 Online Posting Anxiety

This dissertation introduced, deﬁned, and empirically tested a construct that
captures a situational anxiety associated with blogging. Online posting anxiety is deﬁned
as a joint effect of two separate factors, visceral arousal and cognitive self-labeling of the
internal state, as anxiety in blogging.

7.3.1 Social Cognitive Deﬁnition of Online Posting Anxiety

There are many deﬁnitions of anxiety. For example, Beck deﬁned anxiety as a
reaction to perceived threats (Beck, 1976; Beck et al., 1985). Izard believed that anxiety
is a hybrid or blend of a number of emotions—fear, distress, shame, anger, and interest
(Izard, 1972, 1977). Reviewing the literature, Spielberger and Sarason concluded that
anxiety has been conceptualized as a stimulus for behavior, as a learned drive, as a
personality variable, and as a complex response (Spielberger & Sarason, 1991).

The deﬁnition of online posting anxiety is adapted from Bandura’s deﬁnition of
anxiety. The results of the model testing suggested that the social cognitive deﬁnition of
anxiety can better capture the characteristics of anxiety than other deﬁnitions. Anxiety is
not just a reaction to perceived threats. Situational congruent self-efﬁcacy and positive
and negative outcome expectations generate anxiety. Anxiety is also not a hybrid of

emotions associated with fear, distress, shame, and anger. Visceral arousal is the common

103

 

phenomenon that anxiety shares with fear, distress, shame, and anger. The cognitive
labeling of the state is what differentiates anxiety from other emotions. Therefore, social
cognitive theory can enrich anxiety theories by adding the cognitive labeling process to
the emotional state view of anxiety and can also add important variables that integrate
environment and personal factors to explain anxiety’s etiology.

7.3.2 Online Posting Anxiety in Lasswell’s Transmission Model of Communication

In line with social cognitive theory, online posting anxiety is rooted in a unique
form of online communication that is characterized by personal media with public
attention. Lasswell’s transmission model of communication can explain features of
blogging. In blogging, a communicator only publishes the content that he or she wants to
publish. Communicators have a high level of control over the content. However, the level
of selectivity is low. Communicators cannot selectively reveal information to one
audience and not to others.

As a communication channel, blogging does not require extra costs. A computer,
Internet access, and a basic knowledge of installing software might be the only
requirements. Additionally, the Internet is a network that is characterized by hyperlinks,
interactivity, and anonymity. Turkle argued that the Internet gives us a new boundary that
we stand on:

We stand on the boundary between worlds we understand through transparent

algorithm and worlds we understand by manipulating opaque simulations. Our

current experience of life “betwixt and between” recalls what the anthropologist

Victor Turner termed a “liminal moment,” a moment of passage. It is a moment of

anxiety, but it is also a moment of invention and creativity. When Turner spoke of

104

liminality, he understood it as a transitional experience, but for us, living the

tension between physical and virtual and between analysis and simulation seems a

permanent state of affairs, our permanent existence on the edge of things. (Turkle,

2004)

The new boundary also inﬂuences the message the channel transmits. Waskul and
Douglass suggested that electronic media allow us to View public activities privately and
view private activities publicly (Waskul & Douglass, 1997). The distinction between
private and public information is blurred online (Papacharissi, 2007). Besides
ambiguousness of private and public information, blog messages are also marked by high
volumes, diversity, and unequal qualities. There are no formal gate-keeping processes in
blogging. The varied qualities of the content, however, do not mean that there are no
rules governing blogging. The results of the Main Study suggested that the variable
blogging positive outcome expectation was the most important determinant of blog-
maintenance intentions. Low-quality blogging is unlikely to be awarded with desired
outcomes. This will decrease the bloggers’ blog-maintenance intentions; they even may
quit altogether. In the world of blogs, self-selectiveness, rather than formal gate-keeping,
regulates the process of content creation. This may provide another direction for future
studies of blog-maintenance intentions. Low-quality content may be a cause of high
negative outcome expectations and low positive outcome expectations, which will in turn
determine blog-maintenance intentions.

Receiver is not a ﬁxed role in blogs. Depending on the situation, a given person
can be a communicator or a receiver. Both the communicator and the receiver are blog

content contributors. From a communicator’s perspective, receivers can be divided into

105

 

wanted or unwanted, known or unknown (Lenhart, 2005). Online receivers are
fragmented. Communicators cannot fully comprehend their audience. Lenhart suggested
that the blog audience is made up of people from different parts of an author’s life, or
from different experiential, geographic, or moral backgrounds (Lenhart, 2005). The “easy
entry, easy exit” model also applies to receivers. The receiver is on a shifting base.

It should be noted that in Study 1, interviewees were asked the following
questions about audiences directly: “Who do you imagine your readers to be? Do you
think about your readers when you blog?” The results indicated that bloggers were
constantly aware of their audience. They did not perceive their audiences as speciﬁc
individuals. Instead, they put their audiences into different social categories: family
members, friends, and colleagues were the answers they gave when asked about “whom
do you imagine your readers to be?” Moreover, the results of the interviews also
suggested that bloggers further divided these social categories into social subcategories.
For example, Alex mentioned that he blogs to communicate with his family in California,
his relatives in Boston, his high school and college friends all over the United States, his
football friends in Pittsburgh, his colleague at the institute where he works, and his
colleagues in India. Another interviewee, Ming (not her real name), told me that:

I broke up with my boyfriend last month. I cannot write this on my blog. There

are so many people on my MSN list, including my colleagues in the company I

work and the colleagues in other companies. I really can’t talk about my private

life on my blog My company colleagues may feel sorry for me and I will feel

like I lost face. It could impair my relationship with my colleagues in other

106

companies. If they knew my weaknesses, they might have taken advantage of

them.

It seems that bloggers use these social subcategories when they think about their
audiences and related outcome expectations. Although audiences have been noted as a
shifting base, gaining or losing an individual reader might not have a very signiﬁcant
inﬂuence on most bloggers because the social subcategories of audiences are still present.
Previous studies developed typologies to identify audiences, such as wanted versus
unwanted. Future studies may examine the subsocial categories. These categories are
directly related to how bloggers perceive blogging outcomes.

The speciﬁc features of blogging communicator, message, channel, and receiver
constitute a unique online environment that may bring online posting anxiety. Buss listed
a few situations that may bring about anxiety (Buss, 1980). The author suggested that
novel situations, feeling conspicuous, unfamiliarity with the norms, and excessive
attention can create anxiety. In novel situations and when unfamiliar with the norm
situation, people do not know to how to react. The uncertainty and unpredictability of the
situation generates anxiety. In situations when people feel conspicuous or receive
excessive attention, they experience anxiety because of social evaluations (Buss, 1980).
Blogging is a new Internet application in which the norms have not fully emerged.
Bloggers are the centers of attention because it is their lives, thoughts, and perspectives
that attract the audience. For blogging, novelty, unfamiliarity, dissimilarity, uncertainty,
and excessive attention induce online posting anxiety. It is worth pointing out again that
these environmental factors do not solely directly inﬂuence online posting anxiety. They

interact with social cognitive variables. Novelty, unfamiliarity, dissimilarity, uncertainty,

107

and excessive attention inﬂuence self-efﬁcacy beliefs, which predict bloggers’ outcome
perceptions. The self-efﬁcacy and outcome expectation beliefs will then determine online
posting anxiety as well.

7.3.3 Online Posting Anxiety as a Process

Any theory of anxiety must address both content and process (Barrett et al., 2007).
The theory of online posting anxiety presented here considers which factors in blogging
situations cause anxiety. In integrating social cognitive theory and anxiety theories, the
theory of online posting anxiety argues that blogging congruent anxiety (social anxiety
and writing anxiety) leads to blogging cognitive beliefs. The cognitive beliefs determine
online posting anxiety. The concept of blog-maintenance intentions was decided by these
cognitive beliefs and online posting anxiety. The results of the model supported the
theory. The factors directly contributing to online posting anxiety include blogging self-
efﬁcacy, positive outcome expectations for blogging, and negative outcome expectations
for blogging. Writing anxiety and social anxiety also indirectly inﬂuenced online posting
anxiety through blogging self-efficacy and outcome expectations for blogging.

The direct paths from self-efﬁcacy and outcome expectations to online posting
anxiety provide further evidence of social cognitive theory. Cognition is a signiﬁcant
determinant of anxiety. The indirect paths from writing anxiety and social anxiety
suggest a solution to the conﬂicting arguments of social cognitive theory. On one hand,
Bandura ﬁrmly posited that cognition determines anxiety (Bandura, 1986). On the other
hand, he acknowledged that one source of self-efﬁcacy is affective arousal (Bandura,
1997). The online posting anxiety model proposed in this dissertation combined these

two conﬂicting statements. This model argued that cognition inﬂuencing anxiety is the

108

basic principle. However, the causal relationship between cognition and anxiety is also
decided by the scopes of the cognitive and anxiety variables. In the same level, cognition
inﬂuences emotion. On different levels, higher-level anxiety inﬂuences lower-level
cognition. The results of the structural model provided support for this new perspective.

A moderate amount of variance of online posting anxiety can be explained by
blogging self-efﬁcacy, positive outcome expectations of blogging, negative outcome
expectations for blogging, social anxiety, and writing anxiety (R2 = .442 in the American
sample and R2 = .459 in the Chinese sample). The variances might place doubts on the
validity of new construct development. Since existing concepts can cover a large amount
of online posting anxiety, do we really need to develop a speciﬁc concept of online
posting anxiety? Additionally, the interview results suggested that writing anxiety,
computer and Internet anxiety, and social anxiety are related to online posting anxiety.
This brings the same question: is online posting anxiety a unique concept? Or it is just a
form of self-efﬁcacy and outcome expectations? Or it is a dimension of social anxiety,
writing anxiety, or computer and Internet anxiety? The three items of the measure of
online posting anxiety were examined to guarantee validity. Self—efﬁcacy and outcome
expectations were not confounded with online posting anxiety. Model 6.5 provides
evidence for the parsimony of online posting anxiety. Online posting anxiety is a unique
concept that can predict blog-maintenance intentions better than other types of anxieties.
7.3.4 Online Posting Anxiety as a Unique Concept

Social anxiety, writing anxiety, and computer and Internet anxiety are well-
studied areas. Moreover, there has been a proliferation of self-report measures of social-

communicative anxiety for new technology communication during the past few years.

109

Some people may suggest that adopting previous measures to examine blogging might be
a more efﬁcient approach than developing a new measure. Using existing measures may
beget valid results. However, using general measures (e. g., social anxiety, writing anxiety,
and computer anxiety) may omit important domain- or population-speciﬁc variables
(Adamopoulos & Kashima, 1999). Existing theories and measures cannot fully capture
the nature of blogging, as demonstrated in model 6.5.

Blogging is a unique practice that is different from other computer-mediated
communication practices. Psychologists argue that it is essential to develop speciﬁc
concepts because they can be provided with unambiguous deﬁnitions and thus lead to
precision in making observations, taking measurements, and testing hypotheses. It also
helps to avoid confusion in the interpretation of research ﬁndings (Crozier, 2001).
Moreover, the focus on application-speciﬁc constructs provides a mechanism for
ensuring that research attends to those factors that are intimately related to the application
in question, which in turn can afford greater predictive abilities (Benbasat & Zmud,
2003).

Additionally, existing related theories and measures have limitations.
Spielberger’s trait—state theory was helpful in advancing the study of anxiety but is
limited because it fails to specify any of the antecedents that are likely to cause A-state
(Martens et al., 1990). McCroskey developed a typology of communication apprehension
and argued that communication apprehension is a trait that is mostly decided by
biological factors (McCroskey, 1984). The author’s solution to measure this trait—adding
four situational communication apprehension scores—betrayed his trait argument. Daly

and Hailey (1984) realized the role of writing tasks, essay types, and classroom

110

environments in inﬂuencing writing apprehension. Unfortunately, they failed to generate
a pure and valid measure of writing apprehension. Clarke’s computer-mediated
communication apprehension scale had the same problem (Clarke, 1991). Brown and
colleagues developed a CMC anxiety construct and tried to explain the causes of CMC
anxiety (Brown et al., 2004). In their model, computer anxiety and communication
apprehension were employed as causes of CMC anxiety. Correlation, or indirect
causation, instead of a direct causal relationship might be a more appropriate description
of the relationship between computer anxiety, communication apprehension, and CMC
anxiety. Subsequently, this dissertation deﬁned and developed a new concept to describe
blogging anxiety. Online posting anxiety is not the sum of these ﬁve anxiety types. It is
related to these anxiety types. However, online posting anxiety is a unique concept
because blogging is a unique practice.
7.4 Blog-Maintenance Intentions as the Dependent Variable

Blog-maintenance intention is a justiﬁed dependent variable for the current study.
Many theories, such as the theory of planned behavior, posit that intention is a valid
dependent variable (Ajzen, 1991). Many previous studies regarding technology use also
applied behavioral intention as the dependent variable (Agarwal & Karahanna, 2000;
Dabholkar & Bagozzi, 2002; Hsi-Peng et al., 2005; Huang & Liaw, 2005; Luarn & Lin,
2005; Mathieson, 1991; Park, 2003b; Yining & Hao, 2002).

People choose to/not to perform a task for a variety of reasons. The results of the
Main Study indicated that blog-maintenance intentions vary according to the bloggers’
general anxiety types (social and writing anxiety), cognition (perceived efﬁcacy beliefs,

positive and negative outcome beliefs), and blogging-speciﬁc anxiety (online posting

111

anxiety). These variables can explain 54.7% of the variance of blog-maintenance
intentions in the American sample and 50.6% of the variance in the Chinese sample.

Intention is also linked to subsequent action (Ajzen, 1991). Studies examining the
intention-behavior relationship have reported a wide range of correlations. In their meta-
analysis of the theory of reasoned action, Sheppard et al. reported intention-behavior
correlations ranging from .10 to .94 (Sheppard et al., 1988). In this dissertation, blog-
maintenance intentions were examined. For most personal diary-type bloggers, blogging
is a voluntary action. Many bloggers play with it as they would a “toy.” Bloggers blog
not because they have to, but because they want to. It is reasonable to hypothesize that
blog-maintenance intentions will lead to actual blog-maintenance behavior.

In fact, blog-maintenance behavior was measured as well. It was measured by
questioning participants as to how many hours and minutes they spend blogging per
weekday, per weekend, and yesterday, and how many hours and minutes they spend
writing, reading, and commenting on blogs per week. The results indicated that a model
that deﬁned biog-maintenance intentions as a dependent variable was a better model than
a model with blog-maintenance behavior as a dependent variable. Moreover, online
posting anxiety failed to predict blog-maintenance behavior. The reasons for this
insigniﬁcant path could be the sample bias. The bloggers who participated in this study
may not experience high online posting anxiety. It is reasonable to hypothesize that
bloggers who feel anxious about blogging will have low blogging frequency or stop
blogging.

In the Main Study, for the American sample, 35.2% of participants updated their

blogs during the last month, which indicated that 35.2% of participants in the sample

112

were still frequently updating their blogs. In the Chinese sample, 72.8% of participants
updated their blogs during the last month. Effort was made to recruit participants who
may have high online posting anxiety scores. Twenty invitation emails were sent to
addresses retrieved from abandoned blogs. Fifty recruitment messages were posted as
comments on abandoned blogs. These trials failed. It seems that abandoned bloggers are
a hidden population that is hard to reach. A longitudinal study design may help to reach
this population. Lenhart and Fox’s Pew report about blogging provided this evidence. To
recruit bloggers, they asked a nationally representative sample of American adults if they
maintained a blog. They then called back these self—identiﬁed bloggers between July
2005 and February 2006. Twenty-nine percent of those called back said they were no
longer keeping a blog or were not willing to take another survey (Lenhart & Fox, 2006).
Therefore, to reach the high online posting anxiety population, we may Start with
contacting current bloggers. Another possible way is to post messages on currently
updated blogs and invite those readers who were bloggers but stopped blogging to
participate. The method used in this study failed to recruit the sample most valid for
online posting anxiety studies, but it can beneﬁt future studies.

Measurement problems may be the reason for the poor ﬁt of the model, with blog-
maintenance behavior as the dependent variable. Self-reported media behavior measures
have long been criticized as unreliable (Collopy, 1996). Participants might have
difﬁculties calculating the time they spend writing, reading, and commenting on blogs
every week. Also, blog writing, reading, and commenting are totally different blog
activities. Aggregating these three activities together might decrease the scale’s reliability

and validity. Future studies may develop more valid and more reliable measures of blog-

113

maintenance behavior. Moreover, factors that are important to blog-maintenance
behavior, for example, perceived ease of use, social norms, and coping self-efﬁcacy, may
fail to be incorporated into the model. High coping self-efﬁcacy can help bloggers to
develop strategies to avoid negative outcomes and increase the performance of blog-
maintenance behavior. Social norms can also determine blog-maintenance behavior
because they are an important element of self-regulation. Incorporating these variables in
future studies may increase the predictability of blog-maintenance behavior models.

7.5 Cultural Inﬂuences

The results of the Main Study suggested that the proposed structural model can be
applied to both Chinese bloggers and American bloggers. Hl—blogging self-efﬁcacy,
positive outcome expectations for blogging, negative outcome expectations for blogging,
and online posting anxiety will predict blog-maintenance intentions for both American
and Chinese bloggers—was supported. Bandura’s social cognitive view of culture was
again supported. Self-efﬁcacy, outcome expectations, and anxiety are universal factors
that will inﬂuence individuals in different cultures.

The research question—what are the signiﬁcant differences between the Chinese
model and the American one—can be examined based on the results of the model. The
results of the group comparison showed that there were signiﬁcant differences between
the betas of the paths for H3b (social anxiety to positive outcome expectations for
blogging, p = .030), H4c (blogging self-efﬁcacy to online publishing anxiety, p = .070),
H4d (blogging self-efﬁcacy to blog-maintenance intentions, p = .000), and H5a (positive
outcome expectations for blogging to online publishing anxiety, p = .073). Speciﬁcally,

social anxiety had a signiﬁcantly larger inﬂuence on positive outcome expectations for

114

blogging in the American sample than in the Chinese sample. As to blogging self-
efﬁcacy, comparing to its inﬂuence in the Chinese sample, blogging self—efﬁcacy had
signiﬁcantly less inﬂuence on online posting anxiety and more inﬂuence on blog-
maintenance intentions in the American sample. Positive outcome expectations for
blogging had a larger inﬂuence on blog—maintenance intentions in the American sample
than in the Chinese sample.

It should be noted that the mean of the social anxiety scores was lower in the
Chinese sample (M = 3.562) than in the American sample (M = 3.995). Chinese bloggers
were generally less socially anxious than American bloggers. Chinese who are high in
social anxiety might not even consider setting up a blog because of their social anxiety
and perceptions of negative social outcomes. Put another way, the entry level for
blogging is higher among Chinese than among American bloggers. Social anxiety had
less inﬂuence on Chinese bloggers, though this might simply be because those who blog
do not have many social anxiety concerns.

Blogging self-efﬁcacy had a greater inﬂuence on online posting anxiety in the
Chinese sample than in the American sample. Chinese bloggers with low self-efﬁcacy
were more likely to experience online posting anxiety than American bloggers. The
results of Study 1 indicated that Chinese bloggers believe blogging is “high technology.”
One interviewee, Vivian (not her real name), has a master’s degree and needs to use a
computer to work every day. She believed that a blog was “high technology,” and she
didn’t start blogging until requested by her colleagues. This might explain why blogging
self-efﬁcacy plays a more important role in the experience of online posting anxiety. In

comparison with American bloggers, Chinese bloggers perceived blogging to be more

115

difﬁcult. This low self-efﬁcacy makes them feel more anxious when blogging. This
argument may be supported by the evidence that the mean values of blogging self-
efﬁcacy were larger in the American sample (M = 4.823) than in the Chinese sample (M
= 4.222).

Blogging self-efﬁcacy and positive outcome expectations for blogging played a
more important role in blog-maintenance intentions in the American sample than in the
Chinese sample. Generally, the data from the American sample ﬁt the model better than
the data from the Chinese sample. This might because culturally speciﬁc items were
removed to conduct structural equation model testing. More Chinese culturally relevant
items, if incorporated into the model, might improve the strength of the paths in the
model. Future research might continue to explore the interactions among self-efﬁcacy,
outcome expectations, anxiety, and intentions for blogging across cultures.

Besides the statistically signiﬁcant differences for these four paths, the results of
Study 1 suggested that cultural differences can be reﬂected in negative outcome
expectations for blogging. Chinese culture is a collectivist culture. In collectivist cultures,
the major normative task is to maintain harmony with others by coming to terms with
their needs and expectations.

Chinese bloggers are more concerned with negative outcomes that would destroy
the harmony they cherish. This was supported by the mean values of individual items of
negative outcome expectations in the two samples. The mean value of the two negative
outcome expectations most relevant to social relationship and collectivist culture
(jeopardize my relationships, and feel shame because my weakness or my private life

might be exposed) was higher (M = 2.89) in the Chinese sample than in the American

116

sample (M = 2.40). In the Chinese sample, the measure consisting of these two items’
correlation with online posting anxiety was higher (r = .424, p < .01) than that of the
American sample (r = .389, p < .01).

It should be noted that American bloggers also have negative outcome concerns.
For example, an American interviewee, Jason (not his real name), needs to take his four
types of audiences into account: his family, colleagues, friends, and himself. He worries
that his colleagues and supervisors might judge him based on his blog. However, he
revealed that if his colleagues do not like him because of his blog, he would not want to
work with them. Jessica also expressed a similar view. These results might indicate that
although American bloggers worry about negative outcomes, they equally value the
selves they present on their blogs. Negative outcomes for blogging were not a very
salient factor for American bloggers.
7. 6 Implications

The study has theoretical and practical implications. First of all, it adds to social
cognitive theory in the following two ways: ( 1) it provides a model that resolves the
conﬂicting views of cognition and anxiety; (2) it integrates cultural and trait perspectives
in social cognitive theory. Speciﬁcally, one discrepancy of social cognitive theory
arguments is the relationship between cognition and anxiety. On one hand, social
cognitive theory states clearly that cognition inﬂuences anxiety. On the other, the theory
states that affective states are a source of self-efﬁcacy. Social cognitive theory fails to
provide an explanation for these conﬂicting views. This research proposed a structural
model that integrates the opposing perspectives. The proposed mechanism is that general

anxiety will inﬂuence speciﬁc cognition. The speciﬁc cognition then will determine

117

speciﬁc anxiety. The causal relationship between cognition and anxiety is determined by
the level and generality of the cognition and anxiety concepts.

Social cognitive theory has been criticized for underemphasizing the inﬂuence of
culture on behavior. This research integrated the individualism/collectivism cultural
dimension in a social cognitive theoretical model. The ﬁndings suggested that culture is
compatible with social cognitive theory’s triadic causal model. The
individualism/collectivism cultural dimension is an E that produces and is produced by Ps
and Bs. Incorporating the individualism/collectivism cultural dimension enriches social
cognitive theory.

Second, previous research on blogging has usually focused on online journalism,
self-presentation, identity, genre analysis, uses and gratiﬁcations, and social network
analysis. No previous study, to the author’s knowledge, has examined blogs within social
cognitive models. Historically, the ﬁeld of blogging research looked ﬁrst at what a blog is
(the format), what is being blogged (the content), and who blogs (the bloggers). Over
recent years, blogging research added why blogging and how, the motivations and the
processes. This research introduced blogging self-efﬁcacy, negative outcome
expectations for blogging, and positive outcome expectations for blogging. These
variables can help to examine the effects of blogging, including how blogging inﬂuences
the subjective well-being of bloggers. This current study might contribute to our
understanding of blogs from the social cognitive perspective and may beneﬁt new blog
theory development.

Third, a new concept, online posting anxiety, is developed based on the literature

of communication apprehension, writing apprehension, social anxiety, computer-

118

mediated communication apprehension, and computer anxiety. Many previous theories
and measures of these concepts have limitations. For example, there has been a huge
amount of research on computer anxiety in the last two decades. However, after critically
reviewing this body of research, one is left with a sense of excitement over an area that is
newly developed, but also a sense of disappointment at the (lack of) intellectual rigor and
contribution to wider understanding that these studies combine to make. Much of the
research has been rather atheoretical in its orientation, conceptualization, and design.
Research in general simply addressed rather descriptive-level questions, and correlation
analysis was the most commonly used data analysis method. Additionally, with the rapid
development of computer technology and applications, the deﬁnitions and measurements
of computer anxiety, which were developed in the late 19805 and early 19903, might be
inappropriate and outdated. Although the new construct of online posting anxiety focuses
especially on blogging, it relates to diverse aspects of Internet use apprehension, such as
computer anxiety, Internet anxiety, and privacy concerns. Online posting anxiety can
contribute to our understanding of users’ perceptions of and attitudes toward broad
applications of the Internet and computing technology. Speciﬁcally, the current research
made a unique and original contribution in that it applied social cognitive concepts to
explain anxiety in a CMC domain. Anxiety is described as visceral arousal that is caused
by low self-efﬁcacy and negative outcome expectations. A structural model that
demonstrates the process through which a speciﬁc type of anxiety interacts with cognitive
factors and other anxiety factors on the Internet in different cultures was developed.
Online posting anxiety can contribute to the analysis of an important trend of the

Intemet—Web 2.0. Blogs are an important application of Web 2.0, along with video

119

sharing and the development of collective or open-source repositories of information
such as Wikipedia. One of the important differences between the Internet and traditional
mass media is that content creators on the Internet may include a wider variety of people,
including those who are not media professionals. One of the common reasons behind the
popularity of Web 2.0 is that anyone can create content and broadcast it to millions of
potential audiences. A Pew report found that 44% of adult Internet users have used the
Internet to publish their thoughts, respond to others, post pictures, share ﬁles, and
otherwise contribute to the explosion of content available online. Twenty-one percent of
Internet users say they have posted photographs to websites. Thirteen percent of Internet
users maintain their own websites. Around 7% have webcams running on their computers,
allowing other Internet users to see live pictures of them and their surroundings (Lenhart
et al., 2004). Lenhart et al.’s survey was conducted three years ago. It is reasonable to
propose that a much higher percentage of Internet users now create online content. Online
posting anxiety can beneﬁt Web 2.0 research because online posting anxiety is an
important obstacle for blogging or, more generally, creating content online. The
understanding of the nature and the treatment of online posting anxiety can help to
promote the use of Web 2.0.

There are practical implications. For usability researchers and interface designers,
this dissertation can help to develop more user-friendly blogging services and authoring
tools, thus offering bloggers a better user experience. Second, the knowledge about how
different factors, such as experience, personality traits, and cognitive variables, inﬂuence
the development of online posting anxiety can help researchers plan and design programs

to change or reduce online posting anxiety. Meanwhile, online posting anxiety may

120

undermine self-efﬁcacy and outcome expectations and decrease people’s intention to set
up a blog or update blogs. Different measures to reduce online posting anxiety can
beneﬁt people’s cognitive self-evaluations and blog use. Third, blogs are now being
portrayed as voices of people and political players and as catalysts for breaking stories
(Perlmutter & McDaniel, 2005). The Internet is strictly regulated in China. Some people
anticipate blogs to play a role that leads to more freedom and democracy in China. Before
enthusiastic claims—including that blogs are rupturing China’s social, cultural, and
political fabric; blogs are promoting freedom of expression; and so on—are made about
the revolutionary impact of blogs on Chinese society, scholars need to examine the
complex adoption patterns of this online service by the Chinese. The diverse implications
of blogs can only be realized after people start to use them. Examining the factors that
inﬂuence Chinese’s blog-maintenance intention might have implications for positive
changes in China’s Internet environment.
7.7 Future Study Areas

There are many issues future studies might investigate. Future studies might add
other variables to the model and examine their power for explaining blogging patterns
and behavior. For example, social cognitive theory’s self-regulatory variables may be
incorporated. Bandura posited that self-regulation, or people’s control over their
behavior, is an important determinant of behavior (Bandura, 1986). Self-regulation is
concerned with norms, social standards, personal standards, and performance attributions
(Bandura, 1986). Self-reﬂection helps individuals make sense of their experiences,
explore their own cognitions and self-beliefs, engage in self-evaluation, and alter their

thinking and behavior accordingly (Bandura, 1986). For blogging, interactions with and

121

feedback from audiences may be a source of self-regulatory beliefs. How bloggers’ self-
regulatory capabilities inﬂuence blogging behavior can be a topic for future study.

Anxiety sensitivity may also be considered. Anxiety sensitivity is conceptualized
as the extent to which individuals fear their own anxiety and anxiety—related sensations
(Taylor, 1999) Speciﬁcally, anxiety sensitivity may amplify individuals’ responses to
their own anxiety sensations and thereby produce a positive feedback cycle of anxiety
(Asmundson et al., 2002). This concept may be applied to examine culture differences of
blogging. Compared to bloggers in individualistic cultures, bloggers in collectivistic
cultures may have higher scores of anxiety sensitivity because they worry that the anxiety
they experience may jeopardize their social relationships.

Anxiety sensitivity might also interact with blogging self-efﬁcacy and blogging
outcome expectations to predict blog-maintenance behavior. Like general anxiety types,
anxiety sensitivity may precede blogging self-efﬁcacy and blogging outcome
expectations. Future studies are needed to test this argument.

Blogging can be seen as a series of behaviors (e.g., set up a blog, update a blog,
read others’ blogs, and comment on others’ blogs). Similar arguments can be found in
stage models [health action process approach (Schwarzer, 1999, 2001, 1992) and the
precaution-adoption process (Weinstein, 1988; Weinstein & Sandman, 1992)]. Stage
models posit that, from a precontemplation stage through a motivation stage to the
initiation and maintenance of behavior, different cognitive beliefs may be important at
different points (Norman & Conner, 1996; Sandman & Weinstein, 1993). Following the

logic of stage models, future studies may focus on those who have not started blogging

122

yet. A detailed analysis of the variables that are important in determining blogging set-up
intention will provide a better account of the nature of blogging.

Future study might also examine different types of blogs, for example, focusing
mainly on personal journal type blogs. As Bruns and Jacobs stated, “Our discussion of
blogs, bloggers, and blogging must become more sophisticated; it makes as little sense to
discuss the uses of blogs as it does to discuss, say, the uses of television unless we
specify clearly what genres and contexts of use we aim to address” (Bruns & Jacobs,
2006). Future study may analyze how different genres and contexts inﬂuence blog
cognitions, anxiety, intentions, and behavior. Employment status, gender, generation
differences, and nonvoluntary blogging may be especially valuable topics for future
investigations.

7. 8 Limitations

There are several limitations of the current study. First of all, the causality and
directions of causality of the inﬂuences from writing anxiety, social anxiety, blogging
self-efﬁcacy, negative outcome expectations for blogging, positive outcome expectations
for blogging, and online posting anxiety to blog-maintenance intentions are still open to
question, although the theory and the model’s overall goodness-of—ﬁt both support the
inclusion.

Researchers often assume that stable personality traits must inﬂuence what they
believe to be the more transient feeling of anxiety; however, there is evidence that the
reverse causal direction may be true (Cunningham, 1988). It is also possible that the
relationship between online posting anxiety and blog self-efﬁcacy and blog outcome

expectations might be reciprocal (i.e., anxiety is both the antecedent and the outcome of

123

efﬁcacy and outcome beliefs) (Crick & Dodge, 1994). Anxiety may increase or inhibit
attention to emotion-relevant information, but, in turn, anxiety may be ampliﬁed or
attenuated by information to which attention is directed (Vasey et al., 1996). Social
cognitive theory might suggest that as individuals experience higher anxiety, they report
lower levels of efﬁcacy and outcome expectations; however, as their efﬁcacy and
outcome belief levels rise, individuals report a corresponding decrease in anxiety
(Thatcher & Perrewe, 2002).

The second limitation is related to the third variable effect. Other factors might
have signiﬁcant contributions to blog-maintenance intentions. Social norms, social
network size, habit, and other demographic and psychological factors might play
essential roles in explaining blog—maintenance intentions and other variables in the
current model. Future studies might consider incorporating other factors in their models.

Another important limitation of our study is that it relies on self-reported,
adjective-based emotion measures, which may bring bias (Davies et al., 1998) Self-report
measures assume that respondents are aware of their emotional experiences and are
accurate in their observations of their own regulatory behavior (Heimberg et al., 2004).
However, differences in anxiety scores may arise because of one or more of the following
reasons: (a) real differences in the intensity of the anxiety evoked by a stressful situation;
(b) differences in the extent to which people recognize their own emotions; and (c)
differences in the degree of conscious or unconscious distortion of responses due to
social-cultural expectations and projected self-images (Spielberger & Sarason, 1991).

More objective measures may be employed in future studies.

124

The same limitation is applicable to cognition and blog-maintenance behavior
measures. As discussed earlier in this chapter, participants might experience difﬁculties
when they try to assess the hours and minutes they blog. Other blog-use measures, such
as content analysis or running a panel study that includes monitoring blog content, may
be helpful to access blogging usage. With regarding to blogging cognition measures, self-
report might be a valid method. Barrett et al. suggested that self-report is useful and
indeed essential for revealing the ontological structure of consciousness (Barrett et al.,
2007). Additionally, Li argued that blogging is different from other forms of media
consumption. It is a highly active action. It demands higher levels of cognition and
motivation. Self—reports of bloggers offer relatively reliable materials to analyze their
cognition (Li, 2005).

The fourth limitation is the small sample size and nonrandom sampling method.
An increased sample size may generate more generalizable data for the blogger
population. The Pew national blogging survey found that bloggers are evenly divided
between men and women. More than half (54%) of bloggers are under the age of 30.
Sixty percent of bloggers are white, 11% are African American, 19% are English-
speaking Hispanic, and 10% identify as some other race (Lenhart & Fox, 2006).

In this research, the sample was 71.3% female and 28.7% male, and 43.8% of the
participants were under the age of 30. The majority of respondents were Caucasian
(64.8%). Other ethnic groups represented include African American (6.2%), Asian
(16.6%), and others (4.8%). Differences were found between this sample and that of the

Pew report.

125

CNNIC’s China national blogging survey found that the sample had 51.1%
female and 48.9% male respondents. The majority, 78.2%, of bloggers were under the
age of 30. The Chinese sample in this research was 51.2% female and 48.8% male. A
similar 79.6% of bloggers were under the age of 30. The sample was very comparable to
the CNNIC’s national sample.

This study used online postings of the survey URL and announcements to recruit
participants. This method posed problems with tracking precise and reliable response
rates. It is also very likely that truly anxious bloggers and those who experienced
negative outcomes never even saw the recruiting websites. Babbie posited that
nonprobability sampling methods are acceptable when random sampling is not possible
(Babbie, 2001). Although results from Study 2 and the Main Study may suffer from self-
selection bias (the sample may only represent those bloggers motivated to ﬁll out the
surveys), the ﬁndings may contribute to the understanding of blogging phenomena.
Another limitation related to sampling method is that we only recruited bloggers between
ages of 18 and 65 years. Herring suggested that adolescent bloggers are an important age
group for blogging studies (Herring et al., 2004b). The ﬁndings of the current study
cannot be applied to bloggers who are above or below the age range studied (18 to 65
years).

The ﬁnal limitation concerns the cross-cultural research bias. Inherently different
populations may blog in the U.S. and China and inherently different populations may see
the recruitment websites. Additionally, comparing responses to scales of cross-cultural
studies requires that researchers assume that what respondents mean when they say that

they agree is sufﬁciently similar cross-culturally to make comparisons meaningful

126

(Oyserman et al., 2002). The equivalence of the constructs across cultures might also be a
threat to the validity of the study. Running measurement models is recommended for
multiple group comparison structural equation modeling testing. However, measurement
models could not be examined because of the small sample sizes. The sample sizes of the
Main Study cannot afford to have to two indicators for each latent variable. Although the
equivalence of the constructs was tested with factor analysis to establish the reliability of

the measures, the lack of measurement model testing is a limitation of this study.

127

 

References

Adamopoulos, J ., & Kashima, Y. (Eds.). (1999). Social psychology and cultural context.
Thousand Oaks: Sage Publications.

Agarwal, R., & Karahanna, E. (2000). Time ﬂies when you're having fun: Cognitive
absorption and beliefs about information technology usage. MIS Quarterly, 24(4),
665-694.

Agarwal, R., Sambamurthy, V., & Stair, R. M. (2000). Research report: The evolving
relationship between general and speciﬁc computer self-efﬁcacy - an empirical
assessment. Information System Research, 11(4), 418-430.

Ajzen, I. (1991). The theory of planned behavior. organizational Behavior and Human
Decision Processes, 50, 179-211.

Al-Khaldi, M. A., & Al-Jabri, I. M. (1998). The relationship of attitudes to computer
utilization: New evidence from a developing nation. Computers in Human
Behavior, 14(1), 23-42.

Albirini, A. (2005). Cultural perceptions: The missing element in the implementation of
ict in developing countries. International Journal of Education and Development
using Information and Communication Technology, 2(1), 49-65.

Allen, M., & Bourhis, J. (1992). Meta-analysis of the relationship between
communication apprehension and cognitive performance. Communication
Education, 41(1), 68-76.

Allen, M., & Bourhis, J. (1996). The relationship of communication apprehension to
communication behavior: A meta-analysis. Communication Quarterly, 44(2),
214-226.

Amichai-Hamburger, Y., & McKenna, K. Y. (2006). The contact hypothesis reconsidered:
Interacting via the intemet. Journal of Computer-Mediated Communication, 11(3).

Armitage, C. J ., & Conner, M. (1999). The theory of planned behaviour: Assessment of
predictive validity and ‘perceived control’. British Journal of Social Psychology,
38, 35-54.

Arnold, M. B. (1960). Emotion and personality. Vol. I: Psychological aspects. New York:
Columbia University Press.

Asmundson, G. J. G., Taylor, S., & Cox, B. J. (Eds.). (2002). Health anxiety:
Hypochondriasis and related disorders. London: John Wiley & Sons.

128

Averill, J. R. (1980). A constructivist view of emotion. In R. Plutchik & H. Kellerman
(Eds.), Theories of emotion (pp. 305-340). New York: Academic Press.

Ayres, J. (1997). A component theory of communication apprehension. Ruston, WA:
Communication Ventures.

Ayres, J ., & Hopf, T. (1985). Visualization: A means of reducing speech anxiety.
Communication Education, 34, 318-323.

Ayres, J ., & Schliesman, T. S. (2002). Paradoxical intention: An alternative for the
reduction of communication apprehension? COMMUNICATION RESEARCH
REPORTS, 19(1), 38-45.

Babbie, E. (2001). Survey research methods. Belmont: Wadsworth.

Bagozzi, R. P., & Lee, K.-H. (2002). Multiple routes for social inﬂuence: The role of
compliance, internalization, and social identity. Social Psychology Quarterly,
65(3), 226-247.

Bandura, A. (1969). Principles of behavior modiﬁcation. New York: Holt, Rinehart &
Winston.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Englewood Cliffs: Prentice Hall.

Bandura, A. (1997). Self-efﬁcacy: The exercise of control. NY: Freemen.

Bandura, A. (2001). Social cognitive theory of mass communication. Media Psychology,
3(3), 265-299.

Bandura, A. (2002). Growing primacy of human agency in adaptation and change in the
electronic era. European Psychologist, 7(1), 2-16.

Bandura, A., Adams, N. E., Hardy, A. B., & Howells, G. N. (1980). Tests of the
generality of self-efﬁcacy theory. Cognitive Therapy and Research, 4, 39-66.

Bandura, A., Reese, L., & Adams, N. E. (1982). Microanalysis’ of action and fear arousal
as a function of differential levels of perceived self-efﬁcacy. Journal of
Personality and Social Psychology Quarterly(5-21).

Bani Ali, A. S. (2005). An assessment of the impact of the ﬁt among computer self-
eﬂicacy, task characteristics and systems characteristics on performance and
information systems utilization. Unpublished Ph.D., The George Washington
University, United States -- District of Columbia.

129

Barbeite, F. G., & Weiss, E. M. (2004). Computer self-efﬁcacy and anxiety scales for an
intemet sample: Testing measurement equivalence of existing measures and
development of new scales. Computers in Human Behavior, 20, 1-15.

Barlow, D. H. (2002). Anxiety and its disorders: The nature and treatment of anxiety and
panic (2nd ed.). New York: The Guilford Press.

Barrett, L. F., Mesquita, B., Ochsner, K. N., & Gross, J. J. (2007). The experience of
emotion. Annual Review of Psychology, 58, 373-403.

Bassam, H. (2006). Effectiveness of computer training: The role of multilevel computer
self-efﬁcacy. Journal of Organizational and End User Computing, 18(1), 50.

Beatty, M. J. (1987). Communication apprehension as a determinant of avoidance,
withdrawal and performance anxiety. Communication Quarterly, 35(2), 202-217.

Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York:
International Universities Press.

Beck, A. T. (1993). Cognitive approaches to stress. In R. Woolfolk & P. Lehrer (Eds.),
Principles and practice of stress management (2nd ed., pp. 333-372). New York:
Guilford.

Beck, A. T., Emery, G., & Greenberg, R. L. (1985). Anxiety disorders and phobias: A
cognitive perspective. New York: Basic Books.

Beckers, J. J ., & Schmidt, H. G. (2001). The structure of computer anxiety: A six-factor
model. Computers in Human Behavior, 17, 35-49.

Bel, D. (2004). Heterogeneity in belief structures regarding cultural consumption: A case
study on the reading of ﬁction. Paper presented at the SABE/IAREP 2004
Conference, Philadelphia: LeBow.

Benbasat, I., & Zmud, R. W. (2003). The identity crisis within the is discipline: Deﬁning
and communicating the discipline's core properties. MIS Quarterly, 27(2), 183-
194.

Blood, R. (2000). Weblogs: A history and perspective. Retrieved November 12, 2004,
from http://wwwrebeccab]ood.net/essays/weblog historyhtml

Blood, R. (2002). The weblog handbook: Practical advice on creating and maintaining
your blog. Cambridge MA: Perseus Publishing.

Boice, R. (1993). Writer's blocks and tacit knowledge. The Journal of Higher Education,
64(1), 19-54.

130

Bortree, D. S. (2005). Presentation of self on the web: An ethnographic study of teenage
girls' weblogs. Education, Communication, & Information, 5(1), 25-39.

Brady, M. (2006, l2-14 September). Blogs: Motivations behind the phenomenon. Paper

presented at the Information Communication and Society Conference, University
of York, York, UK.

Brown, S. A., Fuller, R. M., & Vician, C. (2004). Who’s afraid of the virtual world?
Anxiety and computer-mediated communication. Journal of the Association for
Information S ystems, 5(2), 79-107.

Bruns, A., & Jacobs, J. (2006). Introduction. In A. Bruns & J. Jacobs (Eds.), Uses of
blogs (pp. 1-8). New York: Peter Lang.

Burgoon, J. K., & Hale, J. L. (1983). A research note of the dimensions of
communication reticence. Communication Quarterly, 31, 238-248.

Bumham, K. P., & Anderson, D. R. (1998). Model selection and inference: A practical
information-theoretic approach. New York: Springer-Verlag.

Burroughs, N. F., Marie, V., & McCroskey, J. C. (2003). Relationships of self-perceived
communication competence and communication apprehension with willingness to
communicate: A comparison with ﬁrst and second languages in micronesia.
COMMUNICATION RESEARCH REPORTS, 20(3), 230-239.

Buss, A. H. (1980). Self-consciousness and social anxiety. San Francisco, CA: W.H.
Freeman & Company.

Cambre, M. A., & Cook, D. (1985). Computer anxiety: Deﬁnition, measurement and
correlates. Journal of Educational Computing Research, 1(1), 37-54.

Campbell, A., Cumming, S. R., & Hughes, 1. (2006). Internet use by the socially fearful:
Addiction or therapy? C YBERPS YCHOLOGY & BEHAVIOR, 9(1), 69-81.

Campbell, 8., & Neer, M. R. (2001). The relationship of communication apprehension
and interaction involvement to perceptions of computer-mediated communication.
COMMUNICATION RESEARCH REPORTS (Fall), 391-398.

Chelley, V., & Larry, R. D. (2002). Investigating computer anxiety and communication
apprehension as performance antecedents in a computing-intensive learning
environment. The Journal of Computer Information Systems, 43(2), 51.

Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a new general self-efﬁcacy
scale. Organizational Research Methods, 4(1), 62-83.

131

Chen, G., Gully, S. M., Whiteman, J ., & Kilcullen, R. N. (2000). Examination of
relationships among trait-like individual differences, state-like individual

differences, and learning performance. Journal of Applied Psychology, 85(6),
835-847.

Chou, C. (2003). Incidences and correlates of intemet anxiety among high school
teachers in taiwan. Computers In Human Behavior, 19(6), 731-749.

Chou, H.-W. (2001). Effects of training method and computer anxiety on learning
performance and self-efﬁcacy. Computers in Human Behavior, 17, 51-69.

Chua, S. L., Chen, D., & Wang, A. F. L. (1999). Computer anxiety and its correlates: A
meta-analysis. Computers in Human Behavior, 15, 609-623.

Clarke, C. T. (1991). Rationale and development of a scale to measure computer-
mediated communication apprehension. Kent State University, Ohio.

CNNIC. (2006). Chinese blogs research report.

Cohen, J. (1977). Statistical power analysis for the social sciences. New York: Academic
Press.

Cohen, J ., Cohen, R, West, S. G., & Aiken, L. S. (2003). Applied multiple
regression/correlation analysis for the behavioral sciences (Third Edition ed.).
Mahwah, New Jersey: IEA.

Cole, D. A., Martin, N. C., & Steiger, J. H. (2005). Empirical and conceptual problems
with longtitudinal trait-state models: Introducing a trait-state-occasion model.
Psychological Methods, 10(1), 3-20.

Collis, B. (1999). Designing for differences: Cultural issues in the design of www-based
course-support sites. British Journal Of Educational Technology, 30(3), 201-215.

Collopy, F. (1996). Biases in retrospective self-reports of time use: An empirical study of
computer users. Management Science, 42(5), 758 - 767.

Compeau, D., & Higgins, C. A. (1995). Computer self-efﬁcacy: Development of a
measure and initial test. MIS Quarterly, 19(2), 189-211.

Compeau, D., Higgins, C. A., & Huff, S. L. (1999). Social cognitive theory and
individual reactions to computing technology: A longitudinal study. MIS
Quarterly, 23(2), 145-158.

Cooper, A., & Sportolari, L. (1997). Romance in cyberspace: Understanding online
attraction. Journal of Sex Education and Therapy, 22, 7-14.

132

Corbett-Whittier, C. (2004). Writing apprehension in adult college undergraduates: Six
case studies. University of Kansas, Kansas.

Crable, E. A., Brodzinski, J. D., Scherer, R. F., & Jones, P. D. (1994). The impact of
cognitive appraisal, locus of control, and level of exposure on the computer

anxiety of novice computer users. Journal of Educational Computing Research,
10(4), 329-340.

Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-

processing mechanisms in children's social adjustment. Psychological Bulletin,
115, 74-101.

Crozier, W. R. (2001). Understanding shyness: Psychological perspectives. London:
Palgrave Macmillan.

Crumbo, G. B. (1999). Writing apprehension and the effects of "i think i can, i think i
can ”. Spalding University, Kentucky.

Culpan, O. (1995). Attitudes of end-users towards information technology in
manufacturing and service industries. Information & Management, 28, 167-176.

Cunningham, M. R. (1988). What do you do when you're happy or blue? Mood,
expectancies, and behavior interest. Motivation and Emotion, 12, 309-331.

Dabholkar, P. A., & Bagozzi, R. P. (2002). An attitudinal model of technology-based
self-service: Moderating effects of consumer traits and situational factors. Journal
of the Academy of Marketing Science, 30(3), 184-201.

Daly, J. A. (1978). Writing apprehension and writing competency. Journal of
Educational Research, 72, 10-14.

Daly, J. A., & Hailey, J. L. (1984). Putting the situation into writing research: Situational
parameters of writingapprehension as disposition and state. In R. E. Beach & L.
Bidwell (Eds.), New directions incomposition research. New York: Guiford.

Daly, J. A., & Miller, M. D. (1975a). The empirical development of an instruction to
measure writing apprehension. Research in the Teaching of English, 9, 242-249.

Daly, J. A., & Miller, M. D. (1975b). Further attitudes in writing apprehension: Sat scores,
success expectations, willingness to take advanced courses and sex differences.
Research in the Teaching of English, 9, 250-256.

Davida, C., Newman, J ., & Palmguist, M. (1995). "i'm just no good at writing:

Epistemological style and attitudes toward writing." Written Communication,
12(3), 298-329.

133

 

Davies, M., Stankov, L., & Roberts, R. D. (1998). Emotional intelligence: In search of an
elusive construct. Journal of Personality and Social Psychology, 75, 989-1015.

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer
technology: A comparison of two theoretical models. Management Science, 35(8),
982—1003.

Davis, L. (2003). An investigation of the use of technology as a teaching tool in the
collegiate business communication course: Does the use of technology decrease
writing apprehension and increase writing ability? Mississippi State University.

Doostdar, A. (2004). "the vulgar spirit of blogging": On language, culture, and power in
persian weblogestan. American Anthropologist, 106(4), 651-662.

Dumdell, A., & Haag, Z. (2002). Computer self-efﬁcacy, computer anxiety, attitudes
towards the intemet and reported experience with the intemet, by gender, in an
east european sample. Computers in Human Behavior, 18, 521-535.

Dwyer, K. K., & Fus, D. A. (2002). Perceptions of communication competence, self-
efﬁcacy, and trait communication apprehension: Is there an impact on basic
course success? COMMUNICATION RESEARCH REPORTS, 19(1), 29-37.

Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth: Harcourt
Brace J ovanovich.

Eastin, M. S. (2005). Internet use: Relating social perceptions and cognitive models to
behavior. CyberPsychology & Behavior, 8(1), 62—75.

Eastin, M. S., & LaRose, R. (2000). Internet self-efﬁcacy and the psychology of the
digital divide. Journal of Computer Mediated Communication, 6(1).

Eastin, M. S., & LaRose, R. (2005). Alt.Support: Modeling social support online.
Computers in Human Behavior, 21(6), 977.

Eﬁmova, E. (2004). Blogs: The stickiness factor. Retrieved March 9, 2007, from
http://wwwhyperorgcom/blogger/mtarchive/OO1545.html

Fagan, M. H., Neill, S., & Wooldridge, B. R. (2003). An empirical investigation into the
relationship between computer self-efﬁcacy, anxiety, experience, support and

usage. Journal Of Computer Information Systems, 44(2), 95-104.

Fishman, B. J. (1997). Student traits and the use of computer-mediated communication
tools: What matters and why?

134

Fiske, A., Kitayama, S., Markus, H., & Nisbett, R. (1998). The cultural matrix of social
psychology. In D. Gilbert, S. Fiske & G. Lindzey (Eds.), The handbook of social
psychology (4th ed., Vol. 2, pp. 915-981).

Flaherty, L. M., Pearce, K., & Rubin, R. B. (1998). Internet and face-to-face
communication: Not functional alternatives. Communication Quarterly, 46(3),
250-268.

Fowler, B., & Knoll, B. M. (1980). Relationship of apprehension about writing to
perofrmance as measured by grades in a college course on composition.
Psychological Reports, 46, 583-586.

Fulk, J ., Steinﬁeld, C. W., Schimtz, J ., & Power, J. G. (1987). A social information
processing model of media use in organizations. Communication Research, 14,
529-552.

Fuller, R. M., Vician, C., & Brown, S. A. (2006). E-leaming and individual
characteristics: The role of computer anxiety and communication apprehension.
The Journal of Computer Information Systems, 46(4), 103-115.

Gillmor, D. (2003). Moving toward participatory journalism. Nieman Reports, 57(3), 79-
80.

Greene, J. O., & Sparks, G. G. (1982). Toward a reconceptualization ofcommunication
apprehension: A cognitive approach. Paper presented at the International
Communication Association, Boston, Mass.

Greene, J. O., & Sparks, G. G. (1983). The role of outcome expectations in the
experience of a state of communication apprehension. Communication Quarterly,
31(3), 212-219.

Gumbrecht, M. (2004). Blogs as "protected space", Workshop on the Weblogging
Ecosystem: Aggregation, Analysis, and Dynamics: WWW2004. New York: ACM
Press.

Harris, J ., & Grandgenett, N. F. (1996). Correlates among teachers' anxieties,
demographics, and relecomputing activity. Journal of Research on Computing in
Education, 28, 300-317.

Hartman, K., Neuwirth, C. M., Kiesler, S., Sproull, L., Cochran, C., Palmquist, M., et al.
(1991). Patterns of social interaction and learning to write: Some effects of

network technologies. Written Communication, 8, 79-113.

Hasan, B. (2003). The inﬂuence of speciﬁc computer experiences on computer self-
efﬁcacy beliefs. Computers in Human Behavior, 19(4), 443.

135

Hebert, M., & Benbasat, I. (1994). Adopting information technology in hospitals: The
relationship between attitudes/expectations and behavior. Hospital & Health
Services Administration, 39(3), 369-383.

Heimberg, R. G., Turk, C. L., & Mennin, D. S. (2004). Generalized anxiety disorder:
Advances in research and practice. New York: Guilford Press.

Hemby, K. V. (1998). Predicting computer anxiety in the business communication
classroom: Facts, ﬁgures, and teaching strategies. Journal of Business and
Technical Communication, 12(1), 89.

Hening, S., Kouper, I., Paolillo, J. C., Scheidt, L., Tyworth, M., Welsch, P., et a1. (2005).
Conversations in the blogosphere: An analysis "from the bottom up ". Paper
presented at the the Thirty-Eighth Hawai'i International Conference on System
Sciences, Hawaii.

Herring, S., Scheidt, L., Bonus, S., & Wright, E. (2004a). Bridging the gap: A genre
analysis of weblogs., Proceedings of HICSS ’ 04. Big Island, Hawaii.

Herring, S., Scheidt, L., Kouper, I., & Wright, E. (2006). A longitudinal content analysis
of weblogs: 2003-2004. In M. Tremayne (Ed.), Blogging, citizenship and the
future of media. London: Routledge.

Herring, S. C., Kouper, 1., Scheidt, L., & Wright, E. (2004b). Women and children last:
The discursive construction of weblogs. Into the blogosphere: Rhetoric,
community, and culture of weblogs. Retrieved March 1, 2007, from
http://blogdibumn.edu/blogosphere/women and children.html

High, A. C. (2006). Does communicating via a mediated environment reduce the
debilitating effects of social anxiety on interpersonal impression management?
University of Delaware.

Hiltz, S. R., & Turoff, M. (1993). The network nation: Human communication via
computer (revised ed.). Cambridge, MA: The MIT Press.

Hofstede, G. (1991). Cultures & organizations: Software of the mind: Intercultural
cooperation and its importance for survival. New York: McGraw-Hill.

Hofstede, G., & Hofstede, G. J. (2005). Cultures and organizations: Software of the mind.
New York: McGraw-Hill USA.

Holt, C. S., Heimberg, R. G., Hope, D. A., & Liebowitz, M. R. (1992). Situational
domains of social phobia. Journal of Anxiety Disorders, 6, 63-77

Hopf, T., & Colby, N. (1992). The relationship between interpersonal communication
apprehension and self-efficacy. Communication Research reports, 9(2), 131-135.

136

Horwitz, B. (2002). Communication apprehension: Origins and management. New York:
Singular, Thomson Learning.

Hsi-Peng, L., Chin-Lung, H., & Hsiu-Ying, H. (2005). An empirical study of the effect of
perceived risk upon intention to use online applications. Information Management
&. Computer Security, 13(2/3), 106.

Hsu, M., & Chiu, C. (2003). Internet self-efﬁcacy and electronic service acceptance.
Decision Support Systems, 38(3), 369-381.

Huang, H.-M., & Liaw, S.-S. (2005). Exploring users' attitudes and intentions toward the
web as a survey tool. Computers in Human Behavior, 21(5), 729.

Igbaria, M. (1993). User acceptance of microcomputer technology: An empirical test.
Omega, 21(1), 73.

Igbaria, M., & Iivari, J. (1995). The effects of self-efﬁcacy on computer usage. Omega,
23(6), 587.

Izard, C. E. (1972). Patterns of emotions: A new analysis of anxiety and depression. New
York: Academic Press.

Izard, C. E. (1977). Human emotions. New York: Plenum.

Izard, C. E. (1993). Four systems for emotion activation: Cognitive and noncognitive
processes. Psychological Review, 100(1), 68-90.

J akobs, E., Fischer, A. H., & Manstead, A. S. R. (1997). Emotional experience as a
function of social context: The role of the other. Journal of Nonverbal Behavior,
21(2), 103-130.

Johnson, R. D. (2005). An empirical investigation of sources of application-speciﬁc
computer-self-efﬁcacy and mediators of the. efﬁcacy - performance relationship.
International Journal Of Human-Computer Studies, 62(6), 737-758.

Johnson, T. J ., & Kaye, B. K. (2004). Wag the blog: How reliance on traditional media
and the intemet inﬂuence credibility perceptions of weblogs among blog users.

Journalism and Mass Communication Quarterly, 81(3), 622-642.

Kanfer, R. (Ed.). (1990). Motivation theory and industrial and organizational psychology
(2nd ed. Vol. 1). Palo Alto, CA: Consulting Psychologists Press.

Karahanna, E., Ahuja, M., Srite, M., & Galvin, J. (2002). Individual differences and
relative advantage: The case of gss. Decision Support Systems, 32(4), 327-341.

137

Karen, D. L., Detmar, W. S., & Sherif, K. (2003). Diffusing the intemet in the arab world:
The role of social norms and technological culturation. IEEE Transactions on
Engineering Management, 50(1), 45.

Kavanaugh, A., Zin, T. T., Carroll, J. M., & Schmitz, J. (2006). When opinion leaders
blog: New forms of citizen interaction. Paper presented at the International
conference on Digital government research, San Diego, CA.

Kawaura, Y., Kawakanri, Y., & Yamashita, K. (1998). Keeping a diary in cyberspace.
Japanese Psychological Research, 40(4), 234-245.

Kaye, B. K. (2004, August 2004). Web site story: An exploratory study of why weblog
users say they use weblogs. Paper presented at the AB] MC Annual Conference,
Toronto, Canada.

Kaye, B. K. (2005). It’s a blog, blog, blog, blog world. ATLANTIC JOURNAL OF
COMMUNICATION, 13(2), 73-95.

Kaye, B. K. (2007 ). Blog use motivations: An exploratory study. In M. Tremayne (Ed.),
Blogging, citizenship, and the future of media (pp. 127-148). New York:
Routledge.

Keren, M. (2004). Blogging and the politics of melancholy. Canadian Journal of
Communication, 29(1), 5-23.

Kivela, R. (1996). Working on networked computers: Effects on esl writer attitude and
comprehension. Asian Journal of English Language Teaching, 6, 85-93.

Kline, R. B. (1998). Principles and practice of structural equation modeling. New York:
Guilford.

Krishnamurthy, S. (2002). The multidimensionality of blog conversations. Internet
Research 3.0.

LaRose, R., & Eastin, M. S. (2004). A social cognitive explanation of intemet uses and
gratiﬁcations: Toward a new theory of media attendance. Journal of Broadcasting
and Electronic Media, 48(3), 458-477.

LaRose, R., Eastin, M. S., & Gregg, J. (2001). Reformulating the intemet paradox: Social

cognitive explanations of intemet use and depression. Journal of Online Behavior,
1 (2).

LaRose, R., Lai, Y. J ., Lange, R., Love, B., & Wu, Y. (2005). Sharing or piracy? An

exploration of downloading behavior. Journal of Computer Mediated
Communication, 11(1), 1-21.

138

LaRose, R., Lin, C. A., & Eastin, M. S. (2003). Unregulated intemet usage: Addiction,
habit, or deﬁcient self-regulation? Media Psychology, 5(3), 225-253.

Lasica, J. D. (2002). Blogging as a form of journalism. Retrieved November 13, 2006,
from http://wwwojr.org/ojr/1asica/1019166956.php

Lasica, J. D. (2003). Blogs and journalism need each other. Nieman Reports(57), 3.

Lawson-Borders, G., & Kirk, R. (2005). Blogs in campaign communication. American
Behavioral Scientist, 49(4), 548-559.

Lazarous, R. (1991). Emotions andadaptation. New York: Oxford University Press.

Lazarous, R. (1995). Vexing research problems inherent in cognitive-mediational theories
of emotion and some solutions. Psychology Inquiry, 6, 183-196.

Lazarus, R. S. (1991). Emotions andadaptation. New York: Oxford University Press.

Lazarus, R. S. (1995). Vexing research problems inherent in cognitive-mediational
theories of emotion and some solutions. Psychology Inquiry, 6, 183-196.

Lazarus, R. S., Averill, J. R., & Opton, E. M. (1970). Toward a cognitive theory of
emotion. In M. B. Arnold (Ed), Feelings and emotions: The loyola symposium
(pp. 207-232). New York: Academic Press.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer
Publishing Company.

Lea, M., & Spears, R. (1995). Love at ﬁrst byte? Building personal relationships over
computer networks. In J. T. Wood & S. Duck (Eds.), Under-studied relationships:
Off the beaten track (pp. 197-245). Thousand Oaks, CA: Sage.

Leary, M. R., & Kowalski, R. M. (1995). Social anxiety. New York: Guilford Press.
Lenhart, A. (2005). Unstable texts: An ethnographic look at how bloggers and their
audience negotiate self-presentation, authenticity and norm formation.

Georgetown University, Washington DC.

Lenhart, A., Fallows, D., & Horrigan, J. (2004). Content creation online: Pew Internet &
American Life Project.

Lenhart, A., & Fox, S. (2006). Bloggers: A portrait of the intemet’s new storytellers: Pew
Internet & American Life Project.

Lento, T., Welser, H. T., & Gu, L. (2006). The ties that blog: Examining the relationship
between social ties and continued participation in the wallop weblogging system.

139

Paper presented at the 3rd Annual Workshop on the Weblogging Ecosystem,
Edinburgh, UK.

Li, D. (2005). Why do you blog: A uses and gratiﬁcation inquiry into bloggers's
motivations. Marquette University, Milwaukee, Wisconsin.

Li, N., Kirkup, G., & Hodgson, B. (2001). Cross-cultural comparison of women students'
attitudes toward the intemet and usage: China and the united kingdom.
CyberPsychology & Behavior, 4(3), 415-426.

Liebowitz, M. R. (1987). Social phobia. Modern Problems of Pharmacopsychiatry, 22,
141-173.

Littlejohn, J. W. (1994). The effects of situational contexts on writing apprehension in
managers. Mississippi State University, Mississippi.

Loch, K. D., Straub, D. W., & Kamel, S. (2003). Diffusing the intemet in the arab world:
The role of social norms and technological culturation. IEEE TRANSACTIONS
ON ENGINEERING MANAGEMENT, 50(1), 45-63.

Loyd, B. H., & Loyd, D. E. (1985). The reliability and validity of an instrument for the
assessment of computer attitudes. Educational and Psychological Measurement,
45, 903-908.

Luarn, P., & Lin, H.-H. (2005). Toward an understanding of the behavioral intention to
use mobile banking. Computers in Human Behavior, 21 (6), 873.

Lucchetti, A. E., Phipps, G. L., & Behnke, R. R. (2003). Trait anticipatory public
speaking anxiety as a function of self-efﬁcacy expectations and self-handicapping
strategies. COMMUNICATION RESEARCH REPORTS, 20(4), 348-356.

Maassen, G. H., & Bakker, A. B. (2001). Suppressor variables in path models:
Deﬁnitions and interpretations. Sociological Methods & Research, 30(2), 241-270.

Mabrito, M. (1989). Writing apprehension and computer-mediated peer response groups:
A case study of four high- and four low-apprehensive writers communicating
face-to-face versus electronic mail. Illinois State University, Illinois.

Mabrito, M. (2000). Computer conversations and writing apprehension. Business
Communication Quarterly, 63(1), 39-49.

MacDougall, R. (2005). Identity, electronic ethos and blogs - a technologic analysis of

symbolic exchange on the new news medium. American Behavioral Scientist,
49(4), 575-599.

140

Madell, D., & Muncer, S. (2006). Internet communication: An activity that appeals to shy
and socially phobic people? C YBERPS Y CHOLOGY & BEHA VIOR, 9(5), 618-622.

Makrakis, V. (1992). Cross-cultural comparison of gender differences in attitude towards
computres in japan andsweden. Scandinavian Journal of Educational Research,
36(4), 275-287.

Mandler, G. (1984). Mind and body: Psychology of emotion and stress. New York:
Norton.

Manstead, A. S. R. (1991). Emotion in social life. Cognition and Emotion, 5, 353-362.

Manstead, A. S. R., & Van-Eekelen, S. A. M. (1998). Distinguishing between perceived
behavioral control and self-efﬁcacy in the domain of academic intentions and
behaviors. Journal of Applied Social Psychology, 28, 1375-1392.

Marakas, G. M., Yi, M. Y., & Johnson, R. D. (1998). The multilevel and multifaceted
character of computer self-efﬁcacy: Toward clariﬁcation of the construct and an
integrative framework for research. Information Systems Research, 9(2), 126-163.

Marcoulides, G. A., Stocker, Y.-O., & Marcoulides, L., D. (2004). Examining the
psychological impact of computer technology: An updated cross-cultural study.
Educational and Psychological Measurement, 64(2), 311.

Marks, D. F., Murray, M., Willig, C., & Evans, B. (2005). Health psychology: Theory,
research and practice (2nd ed.). Thousand Oaks: Sage Publication.

Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition,
emotion, and motivation. Psychological Review, 98(2), 224-253.

Markus, M. L. (1990). Toward a "critical mass" theory of interactive media. In J. Fulk &
C. W. Steinﬁeld (Eds.), Organizations and communication technology (pp. 491-
511). Newbury Park, CA: Sage.

Marlow, C. A. (2006a). Investment and attention in the weblog community. Paper

presented at the AAAI Symposium on Computational Approaches to Analyzing
Weblogs, Stanford.

Marlow, C. A. (2006b, June). Linking without thinking: Weblogs, readership and online
social capital formation. Paper presented at the International Communication

Association Conference, Dresden, Germany.

Martens, R., Vealey, R. S., & Burton, D. (1990). Competitive anxiety in sport (New Ed
edition ed.). Champaign, IL: Human Kinetics Publishers.

Mason, J. (2002). Qualitative researching. London: Sage Publications.

141

Matheson, D. (2004). Weblogs and the epistemology of the news: Some trends in online
journalism. New Media & Society, 6(4), 443-468.

Mathews, A., & MacLeod, C. (1994). Cognitive approaches to emotion and emotional
disorders. Annual Review of Psychology, 45, 25-50.

Mathieson, K. (1991). Predicting user intentions: Comparing the technology acceptance
model with the theory of planned behavior. Information Systems Research, 2(3),
173-191.

Maurer, M. (1994). Computer anxiety correlates and what they tell us: A literature review.
Computers in Human Behavior, 10, 369-376.

Mazur, M. A., Burns, R. J ., & Emmers-Sommer, T. M. (2000). Perceptions of relational
interdependence in online relationships: The effects of communication
apprehension and introversion. COMMUNICATION RESEARCH REPORTS,
17(4), 397-406.

McCoy, S. (2002). The effect of national culture dimensions on the acceptance of
information and technology: A trait based approach. Unpublished Ph.D.,
University of Pittsburgh, United States -- Pennsylvania.

McCoy, S., Everard, A., & Jones, B., M. (2005). An examination of the technology
acceptance model in uruguay and the us: A focus on culture. Journal of Global
Information Technology Management, 8(2), 27.

McCroskey, J. C. (1970). Measures of communication bound anxiety. Speech
Monographs, 37, 269-277.

McCroskey, J. C. (1982). Oral communication apprehension: A reconceptualization. In
M. Burgoon (Ed.), Communication yearbook 6 (pp. 136-170). New Brunswick,
NJ: Transaction Books.

McCroskey, J. C. (1984). The communication apprehension perspective. In J. C.
McCroskey & J. A. Daly (Eds.), Avoiding communication: Shyness, reticence,
and communication apprehension (pp. 13-38). London: Sage Publications.

MCcullough, S. C., Russell, S. G., Behnke, R. R., Sawyer, C. R., 8c Witt, P. L. (2006).
Anticipatory public speaking state anxiety as a function of body sensations and
state of mind. Communication Quarterly, 54(1), 101-109.

McDowell, E. (1998). An investigation of the relationships among technology

experiences, communication apprehension, writing apprehension, and computer
anxiety. Joumal of Technical Writing and Communication, 28(4), 345.

142

McFarland, D. J ., & Hamilton, D. (2006). Adding contextual speciﬁcity to the technology
acceptance model. Computers in Human Behavior, 22(3), 427.

McInemey, V., Marsh, H., & McInemey, D. (1999). The designing of the computer
anxiety and leaming measure (calm): Validation of scores on a multidimensional
measure of anxiety and cognitions relating to adult learning of computing skills
using structural equation modeling. Educational and Psychological Measurement,
59(3), 451-470.

Mcinemey, V., Mcinemey, D. M., & Sinclair, K. E. (1994). Student teachers, computer
anxiety and computer experience. Journal of Educational Computing Research,
11(1), 27-50.

McKain, T. L. (1991). Cognitive, affective, and behavioral factors in writing anxiety.
Catholic University of America.

Mikkelsen, A., Ogaard, T., Lindoe, P. H., & Einar Olsen, 0. (2002). Job characteristics
and computer anxiety in the production industry. Computers in Human Behavior,
18(3), 223.

Miller, C. R., & Shepherd, D. (2004). Blogging as social action: A genre analysis of the
weblog. In L. Gurak, S. Antonijevic, L. Johnson, C. Ratliff & J. Reyman (Eds.),
Into the blogosphere: Rhetoric, community, and culture of weblogs: University of
Minnesota.

Miller, M. D. (1987). The relationship of communication reticence and negative
expectations. Communication Education, 36(July), 228-235.

Milne, J. M. (2004). Weblogs and the technology lifecycle: Context, geek-chic and
personal community. University of South Florida.

Nardi, B. A., Schiano, D. J ., & Gumbrecht, M. (2004a). Blogging as social activity, or,
would you let 900 million people read your diary? CSC W. Chicago, Illinois.

Nardi, B. A., Schiano, D. J ., Gumbrecht, M., & Swartz, L. (2004b). Why we blog?
Communications of the ACM, 4 7(12), 41-46.

Norman, P., & Conner, M. (1996). The role of social cognition models in predicting
health behaviors: Future directions. In M. Conner & P. Norman (Eds.), Predicting
health behavior: Research and practice with social cognition models (pp. 197-
225). Buckingham, Philadelphia: Open University Press.

Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and
collectivism: Evaluation of theoretical assumptions and meta-analyses.
Psychological Bulletin, 128(1), 3-72.

143

Pajares, F., & Johnson, M. J. (1994). Conﬁdence and competence in writing: The role of
self-efﬁcacy, outcome expectancy, and apprehension. Research in the Teaching of
English, 29, 313-331.

Papacharissi, Z. (2004, May). The blogger revolution? Audiences as media producers.
Paper presented at the International Communication Association, New Orleans,
LA.

Papacharissi, Z. (2007). Audience as media producers: Content analysis of 260 blogs. In
M. Tremayne (Ed.), Blogging, citizenship, and the future of media (pp. 21-38).
New York: Routledge.

Parboteeah, D. V., Parboteeah, K. P., John, B. C., & Choton, B. (2005). Perceived
usefulness of information technology: A cross-national modell. Journal of Global
Information Technology Management, 8(4), 29.

Park, D. (2003a, October). Bloggers and warbloggers as public intellectuals: Charging
the authoritative space of the weblog. Paper presented at the Internet Research 4.0,
Toronto, Canada.

Park, J .-J . (2003b). Understanding consumer intention to shop online: A model
comparison. Unpublished Ph.D., University of Missouri - Columbia, United
States -- Missouri.

Parks, M. R., & Floyd, K. (1996). Making friends in cyberspace. Journal of Computer
Mediated Communication, 1.

Patterson, B. R., & Gojdycz, T. K. (2000). The relationship between computer-mediated
communication and communication related anxieties. COMMUNICATION
RESEARCH REPORTS, 17(3), 278-287.

Perlmutter, D. D., & McDaniel, M. (2005). The ascent of blogging. Nieman Reports,
59(3), 60-65.

Perseus. (2003). The blogging iceberg.

Peter, J ., & Valkenburg. (2006). Research note: Individual differences in perceptions of
intemet communication. European Journal of Communication, 21 (2), 213-226.

Phillips, G. (1980). On apples and onions: A reply to page. Communication Education,
29, 105-108.

Phinney, M. (1991). Word processing and writing apprehension in ﬁrst and second
language writers. COMPUTERS and COMPOSITION, 11(1), 65-82.

144

Poff, S. I. (2004). Regimentation: A predictor of writer's block and writing apprehension.
University of Southern California.

Poor, N. (2005). Mechanisms of an online public sphere: The website slashdot. Journal
of Computer Mediated Communication, 10(2).

Presno, C. (1998). Taking the byte out of intemet anxiety: Instructional techniques that
reduce computer/intemet anxiety in the classroom. Journal of Educational
Computing Research, 18, 147-161.

Prickel, D. O. (1994). The development and validation of a writing self-eﬁ‘icacy scale for
adult basic writers and its use in correlational analysis. Oregon State University,
Oregon.

Rak, J. (2005). The digital queer: Weblogs and intemet identity. Biography, 28(1), 167-
182. .

Reed, M. K., & Keeley, S. M. (1986). Writing apprenhension: Its relationship to writing
quality. Educational and Psychological Research, 6, 83-98.

Reinsch, N. L. (1985). Technology aversion (with implications for education and
training). Ofﬁce Systems Research Journal, 4, 9-20.

Reisenzein, R. (2001). Appraisal processes conceptualized from a schema-theoretic
perspective: Contributions to a process analysis of emotions. In K. R. Scherer, A.
Schorr & T. J ohnstone (Eds.), Appraisal processes in emotion: Theory, methods,
research (pp. 3-19). Oxford, UK: Oxford University Press.

Richmond, V. P., & Dickson-Markman, F. (1985). Validity of the writing apprehension
test: Two studies. Journalism Quarterly, 65, 384-391.

Richmond, V. P., & McCroskey, J. C. (1998). Communication apprehension, avoidance,
and effectiveness (5th ed.): Allyn & Bacon.

Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York: The Free Press.

Roseman, I., Antoniou, A. A., & Jose, P. E. (1996). Appraisal determinants of emotions:
Constructing a more accurate and comprehensive theory. Cognition and Emotion,
10, 241-277.

Rosen, J. (2004, December, 19). Your blog or mine? The New York Times.

Rosen, L. D., Sears, D. C., & Weil, M. M. (1993). Treating technophobia:Alongitudinal

evaluation of the computer phobia reduction program. Computers in Human
Behavior, 9, 27-50.

145

Rosen, L. D., & Weil, M. M. (1995). Computer anxiety: A cross-cultural comparison of
university students in ten countries. Computers in Human Behavior, 11, 45-64.

Rosenberg, M. (Ed.). (1986). Self-concept from middle childhood through adolescence.
(Vol. 3). Hillsdale, NJ: Erlbaum.

Sam, H. K., Othman, A. E. A., & Nordin, Z. S. (2005). Computer self-efﬁcacy, computer
anxiety, and attitudes toward the intemet: A study among undergraduates in
unimas. Educational Technology & Society, 8(4), 205-219.

Sandman, P. M., & Weinstein, N. D. (1993). Predictors of home radon testing and
implications for testing promotion programs. Health Education Quarterly, 20, 1-
17.

Schafer, S. (2006, Feb 27). Blogger nation; a proliferation of voices is slowly dismantling
the status quo in china. Newsweek International Edition.

Scheidt, L. (2006). Adolescent diary weblogs and the unseen audience. In D.
Buckingham & W. Rebekah (Eds.), Digital generations: Children, young people
and new media. London: Lawrence Erlbaum.

Scherer, K. R., Schorr, A., & Johnstone, T. (2001). Appraisal processes in emotion:
Theory, methods, research. Oxford: Oxford University Press.

Schneier, F., & Welkowitz, L. (1996). The hidden face of shyness. New York: Avon
Books.

Schuster, M. L. (2006). “where are the women? ” rhetoric and gender in weblog
discourse. UNIVERSITY OF MINNESOTA.

Schwarzer, R. (1999). Self-regulatory processes in the adoption and maintenance of
health behaviors: The role of optimism, goals, and threats. Journal of Health
Psychology, 4(2), 115-127.

Schwarzer, R. (2001). Social-cognitive factors in changing health-related behavior.
Current Directions in Psychological Science, 10, 47-51.

Schwarzer, R. (Ed). (1992). Self-efﬁcacy: Thought control of action. Washington, DC:
Hemisphere.

Scott, C. R., & Rockwell, S. C. (1997). The effect of communication, writing, and

technology apprehension on likelihood to use new communication technologies.
Communication Education, 46(1), 44.

146

Scott, C. R., & Timmerman, C. E. (2005). Relating computer, communication, and
computer-mediated communication apprehensions to new communication
technology use in the workplace. Communication Research, 32(6), 683-725.

Seidman, I. (1998). Interviewing as qualitative research: A guide for researchers in
education and the social sciences: Teachers College Press.

Sheppard, B. H., Hartwick, J ., & Warshaw, P. R. (1988). The theory of reasoned action:
A meta-analysis of past research with recommendations for modiﬁcations and
future research. The Journal of Consumer Research, 15(3), 325-343.

Singer, J. B. (2005). The political j-blogger: "normalizing" a new media form to ﬁt old
norms and practices. Journalism, 6(2), 173-198.

Smith, 3., & Caputi, P. (2001). Cognitive interference in computer anxiety. Behaviour &
Information Technology, 20(4), 265-273.

Song, I., LaRose, R., Eastin, M. S., & Lin, C. A. (2004). Internet gratiﬁcations and
intemet addiction: On the uses and abuses of new media. CyberPsychology &
Behavior, 7(4), 384 -394.

Spielberger, C. D. (1972). Anxiety: Current trends in theory and research. New York:
Academic Press.

Spielberger, C. D. (1985). Anxiety, cognition and affect: A state-trait perspective. In A. H.
Tuma & J. Maser (Eds.), Anxiety and the affective disorders. Hillsdale, NJ:
Erlbaum.

Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1983). Manual for the state-trait
anxiety inventory. Palo Alto, CA: Consulting Psychologists Press.

Spielberger, C. D., & Sarason, I. G. (Eds.). (1991). Stress and anxiety (Vol. 13).
Bristol,PA: Hemisphere Publishing.

Strauss, A. L., & Corbin, J. M. (1998). Basics of qualitative research techniques and
procedures for developing grounded theory (2nd ed.). Thousand Oaks: CA: Sage
Publications.

Su, N. M., Wang, Y., Mark, G., & Aiyelokun, T. (2005). A bosom buddy ofor brings a
distant land near: Are bloggers a global community? Paper presented at the
Proceedings of the Second International Conference on Communities and
Technologies.

Sugiyarna, S. (2003, May 23—26). Exploration of instructional communication
environment: Mediated communication and communication apprehension. Paper

147

presented at the International Communication Association Annual Conference,
New Orleans.

Sullivan, A. (2002). Andrew sullivan: An honest blogger will never make a quick buck.
Retrieved June 12, 2007, from
http://www.freerepublic.com/focus/news/76823S/posts

Sundar, S. S., Edwards, H. H., Hu, Y., & Stavrositu, C. (2007). Blogging for better health:
Putting the "public" back in public health. In M. Tremayne (Ed), Blogging,
citizenship, and the future of media (pp. 83-102). New York: Routledge.

Susan, R. (2006). The mission of the j-blog: Recapturing journalistic authority online.
Journalism, 7(1), 65-83.

Takhteyev, Y., & Hall, J. (2005). Blogging together: Digital expression in a real—life
community, Social Software in the Academy Workshop. Los Angeles.

Taylor, S. (1999). Anxiety sensitivity. Mahwah, NJ: Erlbaum.

Thatcher, J. B., & Perrewe, P., L. (2002). An empirical examination of individual traits as

antecendents to computer anxiety and computer self-efﬁcacy. MIS Quarterly,
26(4), 381.

Thomas, R. M. (1987). Computer technology: An example of decision-making in
technology transfer. In R. M. Thomas & V. N. Kobayashi (Eds.), Educational
technology—its creation, development and cross-cultural transfer (pp. 25-34).
Oxford: Pergamon Press.

Torkzadeh, G., & van Dyke, T. P. (2001). Development and validation of an intemet self-
efﬁcacy scale. BEHAVIOUR & INFORMATION TECHNOLOGY, 20(4), 275-280.

Traﬁmow, D., Triandis, H. C., & Goto, S. (1991). Some tests of the distinction between

private self and collective self. Journal of Personality and Social Psychology, 60,
649-655.

Trammell, K. D. (2005, May 2005). Looking at the pieces to understand the whole: An
analysis of blog posts, comments, and trackbacks. Paper presented at the
International Communication Association, New York City.

Trammell, K. D., & Keshelashvili, A. (2005). Examining the new inﬂuencers: A self
presentation study of a-list blogs. Journalism &. Mass Communication Quarterly,
82(4), 968-982.

Trevino, E. M. (2005). Blogger motivations: Power, pull, and positive feedback.
Retrieved March 1, 2007, from
http://blog.erickamenchen.net/MenchenBlogMotivations.pdf

148

Triandis, H. C. (1989). The self and social behavior in differing cultural contexts.
Psychological Review, 96(3), 506-520.

Triandis, H. C. (1995). Individualism and collectivism. San Francisco, CA: Westview
Press.

Turkle, S. (2004). The second self: Computers and the human spirit (Twentieth
Anniversary Edition ed). Cambridge, MA: The MIT Press.

Tzelgov, J ., & Henik, A. (1991). Suppression situations in psychological research:
Deﬁnitions, implications, and applications. Psychological Bulletin, 109(3), 524-
536.

Vasey, M. W., El-Hag, N ., & Daleiden, E. L. (1996). Anxiety and the processing of
emotionally threatening stimuli: Distinctive patterns of selective attention among
high- and low-test-anxious children. Child Development, 67(3), 1173-1185.

Vician, C., & Davis, L., R. (2002). Investigating computer anxiety and communication
apprehension as performance antecedents in a computing—intensive learning
environment. The Journal of Computer Information Systems, 43(2), 51.

Viegas, F. B. (2005). Bloggers' expectations of privacy and accountability: An initial
survey. Journal of Computer-Mediated Communication, 10(3).

Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal,
and hyperpersonal interaction. Communication Research, 23(1), 3-43.

Walther, J. B. (1997). Group and interpersonal effects in international computer-mediated
collaboration. Human Communication Research, 23, 342-369.

Wang, Y. S., Wang, Y. M., Lin, H.-H., & Tang, T. I. (2003). Determinants of user
acceptance of intemet banking: An empirical study. International Journal of
Service Industry Management, 14(5), 501-519.

Waskul, D, & Douglass, M. (1997). Cyberself: The emergence of self in on- -line chat.
The Information Society, 13(4), 375- 397.

Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of
Consulting and Clinical Psychology, 33, 448-457.

Weil, M. M., & Rosen, L. D. (1995). The psychological impact of technology from a
global perspective: A study of technological sophistication and technophobia in

university students from 23 countries. Computers in Human Behavior, 11(1), 95-
133.

149

Weinstein, N. D. (1988). The precaution adoption process. Health Psychology. 7 (4), 355-
386.

Weinstein, N. D., & Sandman, P. M. (1992). A model of the precaution adoption process:
Evidence from home radon testing. Health Psychology, 11, 170-180.

Wheeless, L. R., Eddleman-Spears, L., Magness, L. D., & Preiss, R. W. (2005).
Informational reception apprehension and information from technology aversion:

Development and test of a new construct. Communication Quarterly, 53(2), 143-
158.

Wierzbicka, A. (1999). Emotions across languages and cultures: Diversity and
universals. Cambridge: Cambridge University Press.

Wijnia, E. (2004). Understanding weblogs: A communicative perspective. Twente
University, The Netherlands.

Wright, K. (2000). Computer-mediated social support, older adults, and coping. Journal
of Communication, 50, 100.

Yining, C., & Hao, L. (2002). Toward an understanding of the behavioral intention to use
a groupware application. Journal of End User Computing, 14(4), 1.

Ziomek, J. (2005). Journalism, transparency, and the public trust. Washington, DC: The
Aspen Institute.

150

‘ulililjjjrrljtjwith *