A Modei
CHUN-YAO HUANG National Taiwan University cyhuang@management. ntu.edu.tw YONG-ZHENG SHEN Yuan Ze University, Taiwanycs@satu rn .yzu. edu .tw
HONG-XIANG LIN OvisLink Corporation, Taiwan [email protected] SHIN-SHIN CHANG National Taiwan University [email protected]
During the past few years, there has been an exponential growth of biogs, and behind these blogs are numerous bioggers who create and manage them, it is wideiy
expected that bioggers armed with their own biogs wiii make a tremendous impact on both mass communication media and marketers who reiy on such media. However, given the widespread use of biogs, there has been iittle systematic anaiysis of the factors behind biogging activities. To serve as a stepping-stone, this articie presents a modei that addresses the reiationships among biogging motivations and behaviors, and reports the empiricai validation of the modei.
But you cannot afford to close your eyes to them, because they are simply the most explosive outbreak in the information world since the Internet itself.
—BusinessWeek, May 2, 2005
INTRODUCTION
Blogs have been making more and more fanfare in the business press, partly owing to the exponential growth of the blogsphere, and partly owing to the threat as well as opportunities that blogs bring forth to various media-related industries. According to Technorati, an important blog search engine, by April 2007 there were over 70 million blogs that had sur-faced on the service's radar screen. It is estimated that there are 1.6 million new postings per day in the ever-enlarging blogsphere, while each day sees around 120,000 new blogs mushrooming on the in-ternet (Sifry, 2007). The growing momentum of blogs as grass-root, user-generated online media seems unstoppable.
Incumbents in media-related industries are puz-zled not only by the sheer size of the gigantic blogsphere, but also by the very nature of biog-ging activities. A set of "postmodern" conditions (Firat and Venkatesh, 1995) has converged on biog-ging, by which bloggers engage in "multiple con-sumption experiences" relating to multiple types of information behavior and are actually "active producers of symbols and signs of consumption." 4 7 2 JOOBOflL DF flOOEmiSlllG RESEIIRCII December 2 0 0 7
Other than pursuing pieces of information online either rationally or to the extent where self-consciousness disappears,- bloggers also play the roles of communicator, producer, explorer, collec-tor, and player in terms of their consumption of information. By its very nature, blogs are typical C2C platforms (Zhao, Fang, and Whinston, 2006). To the mass media, especially the newspaper industry, the power of blogs leads to the erosion of their audience. Many major media players are trying every avenue to avoid being adversely af-fected by blogs. To advertising and public rela-tions players who are still testing the waters of using commercial websites as a medium for mar-keting communication, blogs represent a new, in-teresting, but uncontrollable platform whose value is largely unproven. Various attempts to utilize blogs for marketing communication are indeed observed, such as commissioning new blogs for specific campaigns (Ives, 2004), buying advertis-ing space on niche blogs with the view of target marketing (Mintz, 2005), inserting advertisements in RSS feeds from some blogs to other blogs that subscribe to them (Nikkei Report, 2005), morphing the print tradition of advertorials to the blogsphere (Ives, 2004), and so forth.
As marketers are seeking a proactive approach to the blogsphere, little has been done to sys-tematically study bloggers' motivations and DOI: 10.2501/S0021849907070493
Little has been done to systematically study bloggers'
motivations and behaviors—^the foundation of
understand-ing the bloggunderstand-ing phenomenon.
behaviors—the foundation of understand-ing the bloggunderstand-ing phenomenon. The present study aims to present a model as a stepping-stone that will help practition-ers and academics to further undpractition-erstand the factors behind blogging activities. In the following, we will first cover the lit-erature related to bloggers' blogging mo-tivations and behaviors. A conceptual model that consists of 10 hypotheses is presented next, followed by the report of an empirical study based on an online survey to validate the conceptual model. We then conclude by discussing the im-plications as well as future research di-rections of the study.
BACKGROUND
What we know about bloggers
Blogs are personal journals on the inter-net arranged in reverse chronological sequence that facilitate interactive com-puter mediated communication through text, images, and audio/video objects. Although the earliest blogs date back to the late 1990s (Dearstyne, 2005), blogs have become a popular tool for inter-active computer mediated communica-tions during the past few years. Thanks to the emergence of various free, easy-to-use blogging services in the market, it is reported that the number of blogs has increased 100 times between mid-2003 and mid-2006 (Sifry, 2006). Behind these many young blogs are their cre-ators, owners, and managers—a cohort of internet users who produce as well as consume content on the internet: the aptly-named bloggers.
Beyond the usual, vague image of par-ticipants in "virtual communities" (Koh and Kim, 2003), who exactly are the bloggers and what do they do? According to a tele-phone survey conducted by Pew Internet & American Life Project, bloggers (those who own their own blogs) are still a mi-nority of internet users—as of July 2006 only 8 percent of internet users keep a blog. More than half of the bloggers are under the age of 30, whereas 84 percent of bloggers keep their blogs as a hobby or pastime. The same survey also reveals that most bloggers are both heavy users of the internet and highly engaged with tech-based social interaction (Lenhart and Fox, 2006) Herring, Scheidt, Wright, and Bonus (2005) found that rather than "external-content-focused, densely in-terconnected journalistic or knowledge-sharing blogs" that have seen quite a few discussions in the early days, most blog-gers actually use blogs for individualistic expression and communication—a finding supported by the Pew survey quoted above. Other than descriptive statistics of blog-gers' profiles directly derived from sur-veys, most efforts in the literature, just like in other cases when people first study a burgeoning phenomenon, are directed at classification. In this light, most re-search studies focus on blogs rather than bloggers. Blood (2002) distinguishes three types of blogs by their functions. Krish-namurthy (2002) proposes to classify blogs into four types by two dimensions of a blog's orientation: personal versus topical and community versus individual. Her-ring, Scheidt, Wright, and Bonus (2005) modify Krishnamurthy's categorization
and ascribes blogs into five categories Dearstyne (2005) suggests five types of blogs by their uses. The ordy sludy we' know of in the literature that focuses on bloggers rather than blogs and attempts to provide a classification scheme is Nardi, J Schiano, Gumbrecht, and Swartz (2004), ^ who propose five motivations for blog- ' "^ gers to blog. ^ „
Beyond all these investigations, little is» known about the links between bloggers' >^ motivations and their blogging activities i Although former studies have shed light i" * on fractural pieces^of the blog phenom-enon, there is a vacuum in the literature that systematically analyzes why blog-gers blog, how they blog, and to what^, " extent bloggers with different blogging' drives are different. The present study plans to fill in the gap by investigating . the links between blogging motivatioris < and behavior. The next sections present a •. brief literature review of the bloggers' mo^ tivations and behavior, followed by the specific research hypotheses to be tested in the empirical study.
Bloggers' motivations ^
To bloggers, a blog may function as a personal diary, a daily pulpit, a coUabora- • tive space, a political soapbox, a collec-tion of links, or a set of memos to the. world (Figueredo, 2005). Bloggers there-fore are likely to blog out of heteroge-neous motivations.
Although several surveys touch the base of bloggers' motivations (e.g.. Blood, 2002; Blumenthal, 2005; Herring, Scheidi, Wright,' and Bonus, 2005; Lenhart and Fox, 2006)^ 4. "^ Nardi, Schiano, Gumbrecht, and Swartz'^'^-(2004) are the first to investigate the issue ^^ ' in depth. Through ethnographic lntefi^^ views, Nardi, Schiano, Gumbrecht, and Swartz (2004) came to the conclusion that ^\"» there are five blogging motivations "to document their life experiences, to pro-vide commentary and opinions, to express
deeply felt emotions, to articulate ideas through writing, and to form and main-tain community forums." They also sug-gest that for a blogger, these motivations may nof be mutually exclusive and may play out simultaneously.
We conducted a prior research by in-terviews with bloggers similar to Nardi, Schiano, Gumbrecht, and Swartz (2004) and fotmd similar results. However, we saw through the interviews that "to ex-press deeply felt emotions" and "to articulate ideas through writing" are in-tertwined; fhe line between the fwo is blurred. Both of them relate fo self-expression and bring process gratifica-tions to bloggers (Swanson, 1992). In fhe related literature discussing blogging or internet use (e.g.. Blood, 2002; Herring, Scheidt, Wright, and Bonus, 2005; Zap-pen, 2005), researchers also treated self-expression as a single, important driver. Therefore, in this study we look at self-expression as a blogging motivation by combing what Nardi, Schiano, Gum-brechf, and Swartz (2004) label as "to express deeply felt emotions" and "to articulate ideas through writing."
Through our interviews, we also found an important factor that has been neglected
so far in the discussion of blogging moti-vation: information seeking. By design, blogs allow bloggers to insert hyperlinks to other sifes/blogs. Blogs also make in-formation updating easy by fhe accommo-dation of information subscription services such as RSS feeds. These features, af least for some bloggers, help them to gather the information they need in a convenient way. To these bloggers, information searching is thus another motivation that drives them fo create and manage their blogs.
Given Nardi, Schiano, Gumbrecht, and Swarfz (2004) as fhe foundation, our own interviews and literature reviews dis-cussed above therefore suggesf that there are five major motivations for a blogger to blog (with which we focus in the fol-lowing study): self-expression, life docu-menting, comdocu-menting, fortun pardcipating, and information searching. Table 1 gives examples of genres and representative blogs associated with these five motiva-tions. It should be noted, however, that most bloggers may blog out of a mix of motivations rather than one single moti-vation. The five major motivations dis-cussed here will serve as the basis for fhe empirical study on fhe motivation— behavior links.
Bloggers' behaviors
Given bloggers' mofivafions, liffle has been discussed in fhe literature abouf blogging behavior, let alone a sensible categoriza-tion scheme. To us, such a scheme can be developed by referring to wider, related literature on people's information behav-ior. As all blogging activities evolve around information created and consumed by bloggers, there are basically fwo behav-ioral orientations of blogging: informa-tion search and social interacinforma-tion, as discussed below.
First, following information economics as pioneered by Sfigler (1961) and Nelson (1970), if can be assumed that bloggers are rational, adaptive, and calculating agents who maximize fheir utilities by searching for information in the blogsphere through links, RSS feeds, and blog search engines, etc. In this utilifarian sense, blog-gers are likely to act like independent agents, who utilize every available design on their own blogs as well as on other blogs that can be reached from their own blogs fo obtain content. The behavioral orientation underlying this "content" focus is information search.
Second, blogging is nof limited fo behav-iors that can be satisfacforily explained by
TABLE 1
Exemplary Blog Genres and Blogs out of Various Blogging Motivations
Motivation Self-expression
Life documenting
Commenting
Community forum participation Information seeking
Exempiary Genre Video blogs
Personal online diaries
Political blog
Project blogs Aggregator blogs
Exemplary Biog
http://www.youtube.com/proflle?user=lonelygirll5,
famous video blog created by Ramesh Flinders and Miles Beckett http://brad.livejournal.com/, by Brad FItzpatrick,
a free software programmer and the creator of LIveJournal http://www.politicalwire.com/, influential political blogs published by Taegan D. Goddard.
Blogs to be found in http://www.whataproject.com/
http://jazter.com/ablog/, an investment-related aggregator blog
There are five major motivations for a blogger to blog:
self-expression, life documenting, commenting, forum
participating, and information searching.
utilitarian rationales. Just like Sherry's (1990) discussion of flea market activities, other than information search, blogging also re-lates to social embeddedness in the sense that the creation and consumption of con-tent are embedded in social interaction. The social aspect of blogging thus relates to vir-tual communities consisting of various blogs. In such a digitally mediated social space, member-generated content is em-phasized and integrated through commu-nication (Hagel and Armstrong, 1997). In the current study, the behavioral orienta-tion forming this "community" focus is so-cial interaction.
The dichotomy of behavioral orienta-tions introduced above is not absolutely
new. Evans, Wedande, Ralston, and Hul
(2001), for example, apply a similar di-chotomy to look at the dynamics of con-sumer interaction in the virtual era. Of course, there may be more elaborate, more complicated schemes to categorize blogging behaviors, but as the first step in the attempt to link up the moti-vations and behaviors of blogging, we adopt the simple dichotomy of behav-ioral orientations with the aim to pave the ground of systematically understand-ing bloggers.
Blog management
How owning a blog is unique to bloggers is that, rather than merely drifting among the sea of websites managed by others, she or he has an online medium with a set of communication tools to manage. From various survey reports (e.g.. Her-ring, Scheidt, Wright, and Bonus, 2005;
Lenhart and Fox, 2006; Sifry, 2006), it is apparent that bloggers differ widely in their efforts in blog management. In these recent surveys, such efforts are reflected by explicit, quantifiable measures such as the scope of interaction through blogroU links and the frequency of content up-date. Having described the distribution of these measures, however, former survey reports fail to explain the obvious hetero-geneity in blogger's blog management ef-forts. Because the explanation of such heterogeneity may provide further in-sights into the blogging phenomenon, in this study we therefore include the possi-ble direct/indirect influences that blog-ging behaviors and motivations may have on blog management efforts.
HYPOTHESES
One major mission of psychologists is to seek to explain the causes of overt hu-man behaviors. Motivation is the major antecedent leading to actual behavior. It is the key driver of behavior that has attracted many research efforts in psy-chology. Early theories in psychology such as that of Maslow (1943) have attempted to build a general theoretical framework of the structure of human motivations, but there seems to be less literature than it deserves to address the links between motivation and behavior. One such frame-work is from Dugree, O'Connor, and Veryzer (1996). Their theory views moti-vations as originating from unfulfilled needs, wants, and desires. These unful-filled needs produce psychological ten-sion and drive that need to be reduced.
The tension and drive in turn lead to visible, goal-directed behavior that satis-fies the unfulfilled needs and achieves individual goals.
This framework provides only a general picture of the links between motivation and behavior in the blogging context. The spe-cific relationships between motivation and the resulting behavior may be different in different contexts. For instance, as illus-trated above, in the blogsphere there is a variety of motivations for bloggers. Their blogging behaviors can be classified into a dichotomy. Exactly how these motivations are related to the blogging behavior is not clear from the general motivation theory discussed above. Thus, the present research investigates the relationships between blog-ging motivations and behavior. According to Dervin's (1983) sense making theory, the "meaning" of the blog to bloggers is shaped through blogging. Out of various motiva-tions, bloggers accumulate knowledge about blogging through longitudinal use with re-gard to the fulfillment of various informa-tion needs. Consequently, they develop a cognitive model, a habitual hunting field, and an activity pattern in their blogging behavior.
Although it has been suggested that in many offline communities, extreme, indi-vidualistic self-expression is not favored by communities highlighting a sense of inclu-sion and interaction (Kozinets, 2002), for communication in the digital space it is found that the self-expression motivation helps explore individual and group iden-tities, facilitating participation and enrich-ing creative collaboration—all "for the purpose of building communities of shared interest" (Zappen, 2005). In this light, blog-gers who have the motivation to express themselves through texts and multimedia content they create may not only blog for creation or expression sake, but are also likely to express themselves to obtain an identity for various social interaction
As all blogging activities evolve around information
created and consumed by bloggers, there are basically
two behavioral orientations of blogging: information
search and social interaction.
activities. In other words, bloggers with the self-expression motivation are not just mo-tivated to express themselves through blog-ging, but would also like to receive feedback from others about themselves. Therefore:
HI: The self-expression motivation leads to interaction-oriented blog-ging behavior.
For bloggers who blog to document their lives, the textual or multimedia records kept on their own blogs are one way to communicate with family and friends, to enrich the ongoing conversa-tions those bloggers are interested in, and to establish an identity in the virtual com-munities they join. For some bloggers, blogging is even "a superior alternative to [sending] mass mail" (Nardi, Schiano, Gumbrecht, and Swartz, 2004). Exchang-ing information is more important than simply gathering information for these bloggers. Therefore:
H2: The life-documenting motivation leads to interaction-oriented blog-ging behavior.
Blogging provides an outlet for blog-gers to express their opinions. More than just private chatting, bloggers may com-ment on issues in the public domain and can get quite serious on a topic (Nardi, Schiano, Gumbrecht, and Swartz, 2004). It has been pointed out that the 2004 U.S. presidential election was a catalyst in the explosive growth of the blogsphere
(Perl-mutter and McDaniel, 2005)—bloggers were urged to comment by the ease of blogging and became excited by the sup-port and/or debate around certain opin-ions. As corrunentaries on blogs invited various responses, interactions are likely to be initiated by the commenting moti-vation. Therefore:
H3: The commenting motivation leads to interaction-oriented blogging behavior.
In the attempt to influence people by com-menting on various topics, people seek more information to solidify their grounds and to elaborate their viewpoints (Lyons and Henderson, 2005). Bloggers who are moti-vated to comment by blogging would like to influence their readers. With the aim of influencing others, bloggers therefore are Hkely to gather content from the blogsphere in support of their arguments. Therefore:
H4: The commenting motivation leads to content-gathering-oriented blog-ging behavior.
For either altruistic or egoistic pur-poses, people who are motivated to par-ticipate in a forum have to interact with other people in the forum to get a sense of involvement. The internet provides some of its users with a sense of belonging to a given online social group upon participat-ing in an ongoparticipat-ing forum (Hiltz and Well-man, 1997; Maignan and Lukas, 1997). The sense of belonging, in turn, is
re-inforced by online interactions made avail-able by the internet. A blog, for people with a forum-participation motivation, is therefore a natural online platform to be involved in various forms of interactions, in which the reinforcement and mobiliz-ing effects (Stanley and Weare, 2004) en-courage interactions. Therefore:
H5: The forum-participation motiva-tion leads to interacmotiva-tion-oriented blogging behavior.
To really participate in an ongoing fo-rum, especially in a forum directed at a specific subject, a blogger need not only socialize with people in the blogsphere, but should also provide relevant informa-tion, opinions, or advice. For a blogger, one of the convenient avenues to prepare the material for contribution to a forum is to gather and/or refer to content from various blogs. Therefore:
H6: The forum-participation motiva-tion leads to content-gathering-oriented blogging behavior.
In the information behavior literature, it is established that the information seek-ing motivation relates to goal-directed, situationally-bound constructing activi-ties (e.g., Dervin, 1983; Savolainen, 1995; Wilson, 1999). Out of the information-seeking motivation, bloggers are more likely to explicitly gather content in the blogsphere. Therefore:
H7: The information seeking motiva-tion leads to content-gathering-oriented blogging behavior.
In the blogsphere, interaction and con-tent gathering are not necessarily inde-pendent of each other. Interactions for self-expression, commenting, and forum participation all can be enriched by the support of more information content. Gathering content can lead to more vibrant
e ^,." social interactions in the virtual
environ-ment. On the other hand, content gather-ing can also be facilitated through advice coming from social interaction in the blogsphere. Therefore:
H8: Interaction-oriented blogging be-havior is positively associated with content-gathering-oriented blogging behavior.
Bloggers blogging with the content gath-ering orientation may resort to many one-way solutions (e.g., search engines, RSS feeds, etc.) to get what they want. In con-trast, blogging with the interaction orien-tation focuses on two-way communication with people the bloggers "know of" to a certain extent. The keener a blogger is to socialize by blogging, the more likely he or she "knows" and interacts with more people in the blogsphere. Therefore:
H9: Interaction-oriented blogging be-havior positively corresponds to the scope of online interaction. Few bloggers go back to their own blogs on a set schedule (Lenhart and Fox, 2006). However, for those who intensely com-municate and interact with other people by blogging, their blogs are more likely to act like their "extended self" (Belk, 1988), and they are more likely to be frequently updated as a consequence. Therefore:
HIO: Interaction-oriented blogging be-havior positively corresponds to the frequency of blog manage-ment.
The set of hypothesized relationships thus constitutes our conceptual model, as Figure 1 illustrates.
RESEARCH METHOD Sample
We tested the conceptual model on blog-gers who kept their own blogs. A random
Self-Expression Life Documenting Commenting Community Forum Participation Information Seeking \ H I - ^ H9/ ] Interaction-Oriented H3^,^ Behavior
Y
/ \^
~~~~-^^^^\
/ \ H 1 0 \ H8 Information-Oriented Behavior , Scope of Online ' Interaction Frequency of Blog ManagementFigure 1 The Model
sample of bloggers whose email accounts are revealed in their blogs serviced by the top-three Taiwanese blog service provid-ers was surveyed to provide data for the empirical study. Of the 1,200 question-naires administered via email, 323 re-sponded. Among these, 311 had complete and internally-consistent information and they constitute the sample for our empir-ical research.
Table 2 compares the demographic and behavioral profile of the sample with other blogger samples that have been reported in the literature (Kumar, Novak, Ragha-van, and Tomkins, 2004; Lenhart and Fox, 2006). Judged by the comparable statistics reported in Table 2, bloggers in our sam-ple have similar profiles to those reported in other surveys.
IVIeasures
Multi-item measures were applied for all the constructs in the model. Tentative mea-sures with a total of 48 items that are relevant to the current study were first developed and the measures were pre-tested on 26 bloggers. At this stage, the
measures were examined for low iterh-to-total correlations. Content validity was also reassessed by two researchers who are familiar with both the blogging phe-nomenon and scale development. The qual-ity of the measurements thus defined was next assessed with confirmatory factor analy-' sis performed on the sample. Only ltems^ * that are loaded in excess of 0.5 were in; '% eluded in the final measures, t h e final'', measures were made up of 25 ilems for.* ,.*-the 9 latent constructs. All paths m.,.*-the*ij5..* measurement model are significant (p < ^-0.05). Except for the two items measurm'g"^\ * the scope of online interaction and an- *i other two measuring frequency of blog"'^',-management, all the other 21 items are-'--measured on 7-point Likert scales ' 'J Table 3 summarizes the items and re- '"* * ports Cronbach's alphas.
ANALYSIS AND RESULTS Analytical approach
Our hypotheses involve the relationships among the five blogging motivations, two blogging behaviors, and two dimensiohs concerning blog management efforts, t o
TABLE 2
Comparisons of Sample Characteristics between the Present Study and Former Surveys
CharacteristicsAge
Present Study Former Surveys
Mean = 23 years old; 70% of the respondents are between 16 and 24 years of age
Three out of four Livejournal bloggers are between 16 and 24 years of age
(Kumar, Novak, Raghavan, and Tomkins, 2004) Blogging history
Number of hours per week spent working on the respondent's own blog
Number of links on the blogroll
Median = 1.1 years Median = 3 hours
Median around 1 year (Lenhart and Fox, 2006) Median = 2 hours (Lenhart and Fox, 2006)
Median = 13 Median = 10 (Lenhart and Fox, 2006)
TABLE 3
Survey Items and Cronbach Alpha Values of the Present Study
Construct Items Cronbach Alpha
Blogging for self-expressing I use my blog to free my mind when I am moody. 0.86 I express myself by writing in my blog.
iVly biog is the piace where I express what i feel.
Blogging for iife documenting I use my biog as my diary to document my iife. 0.81 By writing text and posting video/audio fiies, i keep a record of my life.
Biogging for commenting i'ni wiiiing to comment on what other bioggers say 0.73 i'd iike to respond to other biogs that i read (no matter if i know of the biogger or not).
i'd iike to receive people's comments on what i post on my blog.
Biogging for forum participating Biogging helps me to make more like-minded friends. 0.64 In my biogroll I have friends with whom I can share things.
By biogging i interact with a set of biogs that have contents similar to what I put in my biog.
Blogging for information seeking Biogging heips me extract information behind events that interest me. 0.85 Biogging heips me explore more information about products and/or services.
To me it is convenient to search for information by biogging.
interaction by biogging I'm used to setting up my biog for easy response to visitors' comments. 0.72 I'm used to sharing what I think and feel on my biog.
i'm used to discussing things that interest me by biogging.
information search by blogging i'm used to iooking for information by exploring blogs in my biogroii. 0.79 i'm used to iooking for information by biog search engines such as Technorati and
Googie Biog Search.
i'm used to iooking for information by iooking at ciassified articies in biogs that i visit. i'm used to iooking for information by iooking at biog articies that are frequentiy quoted.
Scope of oniine interaction There are around blogs in my biog's biogroii. 0.85 iVIy biog is inciuded in around biogs' biogroii.
Frequency of biog management On average i update my biog every days. 0.78 I updated rriy biog days ago.
4 tivation leads to the engagement of online
Correlations among Constructs interaction by blogging, is confirmed
(co-efficient = 0.19, t-value = 2.32). Hypothesis
A B C D E F G H I H4 is also confirmed, which predicts that
the commenting motivation also leads to A. Interaction behavior 1 °
interaction behaviors by blogging (coeffi-B;..C9nterrt gathering behavior 0.33 1 ^^^^ ^ 0^9^ ^ . ^ ^ 1 ^ ^ ^ 2.05). The only
di-C. Scope of online interaction 0.21 0.07 1 vergence from our expectation in this p. Frequency of management 0.34 0.11 0.07 1 empirical analysis relates to the forum par-ticipation construct. We hypothesize that
E. Self-expression motivation 0.19 0.72 0.04 0.07 1 , , . . . .
the forum partictpation motivation leads to .':-...':.'?;^..^9P!J.™.".l'r'I..T.9^.'y?!^.'.°.". 9.-.?2 ..9.-21 .0-13 0.21 O.I6 l both interaction (Hypothesis H5) and con-G. Commenting motivation 0.65 0.25 0.13 0.22 0.20 0.54 1 tent gathering (Hypothesis H6) behaviors.
^ ,^ „ ^^ „ ^^ ^ ^^ „ ,^ ^-„ ^ ^^ ^ ' The data, however, reveal that Hypothesis
H. Forum participating motivation 0.46 0.55 0.09 0.16 0.45 0.38 0.41 1 . '^ H5 is not empirically supported (coeffi-I. Inforniation seeking nnotivation 0.33 0.6^ ^,-gj^j = -0.051, t-value = -0.69), whereas Hypothesis H6 has marginal support (co-efficient = 0.20, t-value = 1.73, and signif-icant at the 0.1 level).
verify the proposed model and related 0.34, t-value = 4.33). Hypothesis H2, which All of the remaining hypotheses find hypotheses simultaneously, a structural predicts that the higher the life-documenting empirical support. The motivation for in-equation modeling (SEM) approach is ap- motivation that a blogger has, the more formation seeking is found to positively plied for the analysis. The hypothesized likely he or she is to interact with people correspond to content gathering behav-model is presented by Figure 1 and is by blogging, is supported as well (coeffi- iors (Hypothesis H7, coefficient = 0.52, estimated by maximum likelihood esti- cient = 0.41, t-value = 4.45). Hypothesis H3, t-value = 5.43). Looking at the two behav-mation with LISREL 8, whereby the co- which proposes that the commenting mo- ioral dimensions, interaction by blogging variance matrix is an input. In model
estimation, correlations between within-construct items are allowed. The fit
indi-ces (x^ with 244 degrees of freedom = T A B L E O
526.22; RMSEA = 0.061; GFi = 0.88; CFi = Summary of R e s u l t s of t h e P r e s e n t Study
0.92; IFI = 0.92) for the model indicate
that the model captures the underlying •^.y.P?*^fsi.f ^.*.P.ff*.f"..?.'?" ^^}*].9.?^^.^}^.'^..{*'^^^*'.^}. !*^.'".f.''.
relationships in the dataset to an accept- HI -i- 0.34 (4.33) Supported able degree. Table 4 reports correlations ,,„ , ^ „ . , , / , ^ ^ - ^ r- ^ . j ° ^ H2 -I- 0.41 (4.45) Supported among the nine latent constructs asesti-mated by the structural model. T. t 9.:^?. (2.32) Supported
H4 + 0.19 (2.05) Supported
•*®^"'*® H5 + -9-9.51 (-9;69) ^"Supported Table 5 reports the model estimates and
H6 + 0.20 (1.73) Marginally supported t-values of the model. From most of the •
hypotheses, the corresponding model co- \}J, + .9;52 (5.43) Supported efficients have the expected signs and are H 8 + 0.082* (2.20) Supported
'"S"^^^'^^"*- H9 + 0.21 (3.15) Supported For H I , the hypothesis that the
self-expression motivation leads to interaction ^B + 9:34 (5.06) ; Supported by blogging is supported (coefficient = *Error covariance.
is found to be positively associated with content gathering by blogging (Hypoth-esis H8, validated by error covariance, coefficient = 0.082, f-value = 2.20). Fur-thermore, the hypotheses that more inter-action by blogging leads to a larger scope of online interaction (Hypothesis H9, co-efficient = 0.21, f-value = 3.15) and a higher frequency of blog management (Hy-pothesis HIO, coefficient = 0.34, t-value = 5.06) are both empirically confirmed.
To further ensure that our conceptual model (Figure 1) does not miss any prob-able causal relationships among the constructs under the current analytical frame, we also fit a model with addi-tional causal links. This alternative model is thus constructed so that each of the five motivations leads to the two blog-ging behavioral dimensions (i.e., with the addition of self-expression -^ content gathering, life documenting -^ content gathering, and information seeking -> in-teraction), whereas both of the blogging behavioral dimensions lead to the scope of online interaction and the frequency of blog management (i.e., with the addi-tion of content gathering —> scope of on-line interaction and content gathering -^ frequency of blog management).
We empirically find that none of the five additional relationships have significant coefficients (at the 0.05 level) in this alternative model. At the same time, the signs and levels of significance of coefficients do not change in this alter-native model. Both AIC (692.79 for the alternative model versus 688.22 for the proposed model) and CAIC (1086.20 ver-sus 1072.15) comparisons indicate that the proposed model, while losing no relevant information about the data, is more parsimonious.
DISCUSSION
In this study we propose a conceptual model of bloggers' blogging activities.
The current study proposes and empirically validates
an analytical framework of blogging as a new mode of
computer mediated communication.
Having clarified heterogeneous motiva-tions and behaviors of blogging, a set of 10 hypotheses that relate the blogging motivations to blogging behaviors and usage patterns is proposed. An empirical study based on data from an online sur-vey validates most of the hypotheses in the conceptual model. It is concluded that interaction by blogging is driven by the motivations of self-expression, life docu-menting, and commenting. On the other hand, content gathering by blogging is found to be driven by the motivations of commenting, forum participation, and information seeking. Furthermore, the in-tensity of the bloggers' interaction-oriented blogging behavior is found to positively influence their scope of on-line interaction and frequency of blog management.
Implications
Maintained by bloggers' passions, blogs are expected to profoundly change the world of mass media (Baker and Green, 2005). To marketers, the emergence of blog-ging implies that there is no longer a scarcity of media, but an even more frac-tural media space. The media incumbents no longer control the shape and flow of the messages they provide to the market, and the "audience" who receives mono-logues from mass media can be itself a new web of media. Up to this stage, mar-keters recognize that blogging is an issue to be faced and a new platform to utilize, but the environment is mostly in a "wait-and-see" mode (Mintz, 2005) mainly be-cause people have not yet figured out
what is working behind the mosaic of the blogsphere.
Most bloggers are admittedly ordinary people blogging for a very small audience (Nardi, Schiano, Gumbrecht, and Swartz, 2004), and the digital hinterland (Baker and Green, 2005) they build up shows a typical "long-tail" (Anderson, 2004) pattern in that the majority of blogs attract little attention. However, it has been quantitatively dem-onstrated in the recent literature (Huang and Lin, 2006) that even following the mass media, eye-ball-counting thinking, accom-modating the "long tail" in an online me-dia plan that can enhance a campaign's cost-effectiveness. Furthermore, as more and more people become bloggers, understand-ing the very nature of bloggunderstand-ing activities is crucial for marketers to utilize their ever-growing blogsphere.
Our analysis provides insights into mo-tivations and behaviors of blogging and may serve as the foundation for market-ers to look for proactive utilizations of blogs. The five blogging motivations dis-cussed in this study show directions for the exploration of new brand communi-cation opportunities. A set of such oppor-tunities is summarized in Table 6. From the results, it is obvious that different blogging motivations should be served by different communication strategies and tactics. Upon facing blog-related commu-nication tasks, managers therefore are sug-gested to pay attention to heterogeneous motivations in addition to conventional demographic variables in market segmen-tation, target selection, positioning, and detailed message design and execution.
TABLE 6
Managing Brand Communication via Blogs by Addressing Various Blogging Motivations
Blogging Motivations Self-expression
Life documenting
Communication Opportunities for Brands
• Provide platforms (e.g., events, competitions) to encourage brand-related self-expressions. • Explore opportunities to link such self-expressions with the brand's communication messages
for conventional media.
• Create brand-related experiences for bloggers to document.
• Make bloggers' brand experiences a part ofthe brand's experiences (e.g., have a meta-blog run by the brand that empathetically documents brand-related personal anecdotes recorded by bloggers). Commenting • Locate the influential commentator blogs; subscribe to their RSS feeds so as to sense the pulses
of the blogsphere.
• Make quick and proactive responses to unfriendly comments. • Invite bloggers to join the brand's public relations activities. Forum participation Encourage staff to participate in brand-related online communities.
Sponsor forums to discuss the brand.
Provide stimulating information for discussion to keep the dialogue ongoing. Information seeking Pay attention to SEO (search engine optimization) on blog-specific search engines.
Provide rich information and easy-to-find paths for eyeballs to converge.
Synthesize internal and external pro-brand blogging activities on the brand's main website.
Furthermore, the dichotomy of blogging behavioral dimensions and the drivers be-hind each dimension expounded by this study are also relevant for advertising and public relations practitioners. Cou-pled with the knowledge of the motiva-tions leading to the two different blogging behaviors, our model suggests that for any online campaigns that attempt to attract bloggers' attention, one has to specify which behavioral dimension that the campaign is targeting. Execution and media planning for information-gathering-oriented blog-gers should provide rich information and easy-to-find paths for bloggers' eyeballs to converge. Other than satiating the bloggers' urge for information seeking, the content provided in this direction should be handy for bloggers to refer to (for their commenting and forum participation motivations).
Execution and media planning tar-geted at interaction-oriented blogging ac-tivities may best attempt to get individual, nurturing empathy, and encourage self-expression by audience bloggers, which will keep the dialogue going. Practition-ers should also pay attention to the con-clusion that a blogger's scope of online interaction and frequency of blogging are determined by the degree of his or her interaction behaviors, and not by that of his or her content-gathering behaviors. If the purpose of a campaign in the blogsphere is to get a simple message diffused as quick and as far as possible, then the logical target should be blog-gers who are blogging mainly to interact with other bloggers. Instead, if the cam-paign has more information to commu-nicate and has a longer-term agenda, then a more balanced targeting plan or even
more emphasis on content-gathering-oriented bloggers will help maximize a campaign's effectiveness.
In terms of research implications, the current study proposes and empirically validates an analytical framework of blogging as a new mode of computer me-diated communication (CMC). Our model addresses bloggers' motivations, behav-iors, and management effort heterogene-ity, and it also clarifies the relationships among them. Given the vacuum in the cur-rent literature that maps out why and how bloggers blog, the analytical framework serves as a stepping-stone for further un-derstanding of the blogging phenomenon.
Researcli iimitations and future researcii
Our conceptual model consisting of a set of hypotheses is largely validated in the empirical analysis, except that in our
data the motivation of forum participation does not significantly correspond to the behavior of interaction by blogging. We are not able to provide a satisfying expla-nation for this relatively counterintuitive result at this stage. However, looking at the Cronbach alpha coefficient of the fo-rum participation construct (Table 3), it is apparent that this construct has relatively low reliability versus other constructs un-der study. The alpha admittedly is lower than the threshold of 0.7 that Nunnally (1978, p. 245) recommends. We acknowl-edge that this low-reliability construct is the weak point in our empirical analysis and needs to be addressed in future studies. The current study looks at five motiva-tions of blogging, two behavioral dimen-sions of blogging, and two indications of efforts in blog management. Although the constructs under study are mostly dis-cussed in the literature and are important in understanding the blogging phenom-enon, they are not exhaustive. For exam-ple, in our prior study that interviewed bloggers, an informant explicitly pro-posed that "seeking help" and "giving help" are motivations that are important for him and some of his fellow bloggers to blog. Another informant revealed that to escape from the "real world" is why she blogs. Furthermore, the current study focuses solely on individual bloggers who blog for nonfinancial purposes without considering motivations and behaviors that relate to business or money making. These cases indicate that although our model accommodates the most common factors in the blogging phenomenon, the picture it provides is certainly not comprehensive. Beyond these limitations, given the model presented here as the foundation, there are various directions for future research. First, the current study looks at bloggers rather than blogs. Future studies that link up the bloggers (their motivations and behaviors as addressed in this study) and their blogs
(e.g., content, direction of outbound hyp-erlinks in the blogroUs, etc.) may provide valuable insights into blogging activities. Second, as there are some attempts to cat-egorize bloggers (e.g.. Herring, Scheidt, Wright, and Bonus, 2005; Lenhart and Fox, 2006; Nardi, Schiano, Gumbrecht, and Swartz, 2004), they mostly focus on some single dimension of bloggers. The analyti-cal framework in this study may help future research to arrive at a more elabo-rate and comprehensive categorization scheme to classify bloggers. Third, an in-teresting and potentially important ques-tion that has been neglected so far is that, because blogging is so easy, costs so little, and seemingly satisfies various informa-tion and noninformainforma-tion needs, why do the majority of internet users up to this stage not keep a blog? There are people who own blogs, there is another group of internet users who do not own a blog, but read blogs, and there are internet users who are dis-tant from the blogsphere. Future studies that provide explanations for such facts will ben-efit marketers in gaining a more realistic picture about the attractiveness of the blogsphere. Fourth, although the model psented in the current study clarifies the re-lationships between various blogging motivations and behavioral orientations, the modeling framework by its nature is not a segmentation tool. A managerially rele-vant extension of our model, given the mo-tivations and behaviors identified, is to empirically apply the means-end chain model (e.g., Reynolds, 2006) so as to ex-tract attributes, consequences and values of blogging in more details. Relevant deci-sion segments can be produced in this way. To conclude, the study herein is an at-tempt to objectively analyze the drivers and behaviors of blogging activities. Be-ing preliminary in nature, we do not claim the model to be comprehensive. How-ever, given the empirical validation re-ported above, we believe that the model
presented in this article provides a reason-able platform for further analysis of the blogging phenomenon.
CHUN-YAO HUANG is an associate professor of market-ing in the Department of Business Administration at Nationai Taiwan University. He received his Ph.D. in marketing from the London Business School. His pri-mary research interests lie in the analysis of internet users' online information behavior and the develop-ment of quantitative marketing models. His previous research has appeared in Marketing Science, the Jour-nai of Advertising, the Journal of t/ie American Society for information Science and Technoiogy, the Inter-nationai Journal of Eiectronic Commerce, among others.
YUNG-CHENG SHEN is an assistant professor in the Department of Business Administration of Yuan-Ze University in Taiwan. He received academic training in psychology for both undergraduate and graduate edu-cation. His major research interests focus on con-sumer decision making, branding, advertising, and online consumer behavior. He teaches courses in consumer behavior, marketing research, and market-ing management.
HoNG-XiANG LIN is a manager at OvisLink Corporation. He holds an MBA from National Tsing Hua University, Taiwan. He is interested in studying the blogging phenomena.
SHIN-SHIN CHANG is a Ph.D. candidate in marketing in the Department and Graduate Institute of Business Administration at National Taiwan University. Her re-search interests include consumers' judgment and behavioral decision making. She has published in the Journal of Management and the Journal of Business Administration.
ACKNOWLEDGMENT
Research funding from the National Science Council, Taiwan (No. 96-2628-H-002-069), is ac-knowledged by the first author.
REFERENCES
ANDERSON, C. "Long Tail." Wired, October 2004: [URL: http://www.wired.com/wired/archive/ 12.10/tail.html].
BAKER, S., and H. GREEN. "Blogs Will Change Your Business." BusinessWeek, May 2, 2005.
BELK, R. W. "Possessions and the Extended Self." lournal of Consumer Research 15, 2 (1988): 139-68.
BLOOD, R. The Weblog Handbook: Practical Advice
on Creating and Maitaining Your Blog.
Cam-bridge, MA: Perseus Publishing, 2002.
BLUMENTHAL, M. M. "Toward an Open-Source Methodology: What Can We Learn from the Blogsphere." Public Opinion Quarterly 69, 5 (2005): 655-68.
DEARSTYNE, B. W. "Blogs: The New Informa-tion RevoluInforma-tion?" InformaInforma-tion Management
jour-nal 39, 3 (2005): 38-44.
DERVIN, B. "An Overview of Sense-Making Research: Concepts, Methods, and Results to Date." Presented at the Annual Meeting of International Communication Association, May 1983: [URL: http://communication.sbs. ohio-state.edu/sense-making/art/artdervin83. html].
DuGREB, J. R, G. C. O'CONNOR, and R. W.
VERYZER. "Observations: Translating Values into Product Wants." Journal of Advertising Research 36, 6 (1996): 90-100;
EVANS, M., G. WEDANDE, L. RALSTON, and S.
HUL. "Consumer Interaction in the Virtual Era: Some Qualitative Insights." Qualitative Market
Research 4, 3 (2001): 150-59.
FiGUEREDO, p. "Blogs and Blog Advertising: Smart Online Marketing, or a Waste of Money?"
Target Marketing 28, 3 (2005): 37.
FiRAT, A. F., and A. VENKATESH. "Liberatory Postmodernism and the Reenchantment of Con-sumption." Journal of Consumer Research 22, 3 (1995): 239-67.
HAGEL, J., and A. ARMSTRONG. Net Gain:
Ex-panding Markets through Virtual Communities.
Cambridge, MA: Harvard Business School Press, 1997.
HERRING, S. C , L. A. SGHEIDT, E. WRIGHT, and
S. BONUS. "Weblogs as a Bridging Genre."
In-formation Technology & People 18,2 (2005): 142-71.
HILTZ, S. R., and B. WELLMAN. "Asynchronous
Learning Networks as a Virtual Classroom."
Communications of the ACM 40, 9 (1997): 44-49.
HUANG, C., and C. LIN. "Modeling the Audi-ence's Baruier Ad Exposure for Internet Adver-tising Planning." Journal of AdverAdver-tising 35, 2 (2006): 129^3.
IvES, N. "Nike Tries a New Medium for Ad-vertising: The Blog." New York Times, June 7, 2004.
KoH, JooN, and YOUNG-GUL KIM. "Sense of Virtual Community: A Conceptual Framework and Empirical Validation." International Journal
of Electronic Commerce 8, 2 (2003): 75-93.
KOZINETS, R. V. "Can Consumers Escape the Market? Emancipatory Illuminations from Bum-ing Man." Journal of Consumer Research 29, 1 (2002): 20-38.
KRISHNAMURTHY, S. "The Multidimensionality
of Blog Conversations: The Virtual Enactment of September 11." Internet Research 3.0, October 2002.
KUMAR, R., J. NOVAK, P. RAGHAVAN, and A.
ToMKiNS. "Structure and Evolution of Blogspace." Communications of the ACM 47, 12 (2004): 35-39.
LENHART, A., and S. Fox. "Bloggers: A Portrait of the Internet's New Storytellers." Pew Internet
& American Life Report, July 19, 2006: [URL:
http://www.pewinternet.org/pdfs/PIP%20 Bloggers%20Report%20July%2019%202006.pdf].
LYONS, B., and K. HENDERSON. "Opinion
Lead-ership in a Computer-Mediated Environment."
Journal of Consumer Behaviour 4,5 (2005): 319-29.
MAIGNAN, I., and B. A. LUKAS. "The Nature
and Social Uses of the Internet: A Qualitative Investigation." Journal of Consumer Affairs 31, 2 (1997): 346-71.
MASLOW, A. H. "A Theory of Human Motiva-tion." Psychological Review 50, 4 (1943): 370-96.
MINTZ, J. "Many Advertisers Find Blogging Frontiers Is Still Too Wild," Wall Street Journal, March 25, 2005.
NARDI, B. A., D. J. SCHIANO, M . GUMBRBCHT,
and L. SWARTZ. "Why We Blog."
Communica-tions of the ACM 47, 12 (2004): 41-46.
NELSON, P. "Information and Consumer Behav-ior." Journal of Political Economy 78,2 (1970): 311-29.
NIKKEI REPORT. "Large Companies Set Sights
on Blog Ad Services Amid Growing Use," June 20, 2005.
NuNNALLY, J. C. Psychometric Theory. New York: McGraw-Hill, 1978.
PERLMUTTER, D . D., and M. MCDANIEL, "The
Ascent of Blogging." Nieman Report 59,3 (2005): 60-64.
REYNOLDS, T. J. "Methodological and Strategy Development Implications of Decision Segmen-tation." Journal of Advertising Research 46,4 (2006): 445-61.
SAVOLAINEN, R. "Introduction to the Special Issue: Everyday Life Information Seeking: Ap-proaching Information Seeking in the Context
of Way of Life." Library and Information Science
Research 17, 3 (1995): 259-94.
SHERRY, J. E "A Sociocultural Analysis of a
Midwestern American Flea Market." Journal of
Consumer Research 17, 1 (1990): 13-30.
SIFRY, D. "State of the Blogsphere, August 2006": [URL: http://www.sifry.com/alerts/].
. "The State of the Live Web, April 2007": [URL: http://www.sifry.com/alerts/].
STANLEY, J. W., and C. WEARE. "The Effects of
Internet Use on Political Participation: Evi-dence from an Agency Online Discussion Fo-rum." Administration & Society 36, 5 (2004): 503-27.
STIGLER, G. J. "The Economics of Information." Journal of Political Economy 69, 3 (1961): 213-25.
SwANSON, D. L. "Understanding Audiences: Continuing Contributions of Gratifications Re-search." Poetics 21, 4 (1992): 305-28.
WILSON, T. D. "Models in Information Behav-iour Research." Journal of Documentation 55, 3 (1999): 249-70.
ZAPPEN, J. P. "Digital Rhetoric: Toward an In-tegrated Theory." Technical Communication
Quar-terly 14, 3 (2005): 319-25.
ZHAO, XIA, FANG FANG, and ANDREW B.
WHIN-STON. "Designing On-Line Mediation Services
for C2C Markets." International Journal of Elec-tronic Commerce 10, 3 (2006): 71-93.
