Psychological and Social Influences on Blog Writing: An Online Survey of Blog Authors in Japan


  • Asako Miura,

    1. Kobe Gakuin University
    Search for more papers by this author
    • Asako Miura is an associate professor in the Department of Psychology at Kobe Gakuin University, Japan. Her research interests are online communication, human-computer interaction, and group creativity.

      Address: 518 Arise Ikawadan-cho Nishi-ku Kobe, 651–2180, Japan

  • Kiyomi Yamashita

    1. Senshu University
    Search for more papers by this author
    • Kiyomi Yamashita is a professor in the School of Network and Information at Senshu University, Japan. Her research interests are self-expression in CMC, the network society, and the family.

      Address: 2–1–1 Higashi-mita Tama-ku Kawasaki, Kanagawa, 214–8580, Japan


We conducted a questionnaire survey of personal blog authors (N = 1,434) and examined two hypothesized models using structural equation modeling to clarify the psychological and social process associated with why authors continue to write their blogs. Two final models with good fit were obtained. It was confirmed that being satisfied with benefits to self, relationships with others, and skill in handling information had significant positive effects on the intention to continue blog writing. The psychological traits of private self-consciousness, reassurance-seeking, and information need were hypothesized to be effective in establishing consciousness of the benefits; these also had significant positive effects. In contrast, only positive feedback had a significant influence on satisfaction related to information handling skill, whereas both negative and positive feedback had significant influences on satisfaction related to information handling skill. This suggests that communication with readers who gave positive feedback strongly encouraged blog authors to continue writing. Similarities and differences between the two models and recommendations for further theoretical development are discussed.


From its early stages, the Internet, and especially the World Wide Web, was a place where people could communicate with one another. The primary means for individuals to communicate information on the web was through the construction of websites or “homepages” (Döring, 2002). Around the middle of the 1990s, construction and publication of personal websites gained momentum on a global scale. Also in Japan, a large number of personal websites were constructed at that time (Hashimoto, Tsuji, Fukuda, Mori, & Yanagisawa, 1996). Personal websites offer their owners an unprecedented opportunity to present almost anything they want to in an inexpensive, highly flexible mass medium to a potentially global audience (Döring, 2002). Over time, many personal websites were connected to each other by hyperlinks and formed electronic communities, demonstrating “the strength of weak ties” (Granovetter, 1973).

Personal websites can be divided roughly into two types. The first is a “self-description” using the web as sort of a stage for website owners. Such personal websites consist almost entirely of identity claims (Vazire & Gosling, 2004). Contents for self-description can include a detailed profile, for example, job, hobbies, or special skills. Most self-description contents are prepared and updated only by website owners. The second type involves “interaction with others.” In such websites, which can incorporate bulletin board systems, guest books, chat rooms, etc., the website owner is a participant as well. In other words, this type can exist only when there are participants other than the website owner. The communication on such websites can take various forms, such as dialogues between two people or large-scale communication in which the participants can be viewed as a community.

Personal Blogs on the Web

Although personal websites can be thought of as either one of the two types in the earliest stages, online diaries emerged around 1995 as a fusion of the two types (e.g., Herring, Kouper, Scheidt, & Wright, 2004). An online diary is a personal diary or journal published on a personal website or a diary hosting website. Authors keep running accounts of their personal lives in their online diaries. An online diary can be defined structurally as a type of website where entries are made and displayed in reverse chronological order. Websites called “diary” (nikki, in Japanese) or similar names were seen from the earliest days of the web (Kawaura, Kawakami, & Yamashita, 1998; McNeill, 2005). The owners of the websites, the online diarists, were at the same time active participants, communicating with others. As online diarists began to learn about each other, webrings were formed to connect the various online diaries. Lists of online diaries by topic allowed people to find websites that had some relation to each other. Kawaura et al. (1998) noted that personal websites in Japan included diaries or similar content more frequently than those in other countries and suggested that cultural differences were the basis for these differences in self-disclosure on the web. Thus far, however, there seems to have been no empirical cross-cultural study of these differences.

Systems and tools usually progress along with the needs of their users. Online diarists have always had a need for both self-description and interaction with others. Starting in about 1998, a new type of website called weblogs (blogs) began to be posted (cf. Blood, 2002) and has gradually merged with online diaries (Herring, Kouper, et al., 2004). A blog is a kind of web-based system that uses a variety of tools to facilitate self-description and interaction with others.

Blogs effectively facilitate self-description, in that all authors have to do is write (type) a title and subsequent sentences in plain text in order to create or update their blog entries. It is not necessary to add HTML codes or upload files to a website. Using a blog system, authors can dramatically simplify the procedures required for online self-description. Additionally, blog systems allow authors to label their individual entries with thematic categories (or tags) and classify their entries according to their entry date and their subject. This function enables blog authors to review their self-description both by history and by subject.

Blog systems provide various tools for supporting interaction with others; the most distinctive are comments and TrackBack. Readers can leave a comment on posted entries and authors can answer it with another comment or by posting a subsequent or revised entry. TrackBack is a mechanism for communicating across blogs. If a blog author writes a new entry commenting on, or referring to, an entry found on another blog, and both blog tools support the TrackBack protocol, then the commenting blog author can notify the other blog with a “TrackBack ping.” The receiving blog will typically display summaries of and links to all the comment entries below the original entry. This allows for conversations spanning several blogs that readers can follow. Both tools help blog authors and readers to interact online in more diverse ways. The extended community of blog authors and readers has come to be known as the “blogosphere,” the web environment in which blog authors “form connections with others while progressing along their own paths” (Badger, 2004, n.p.).

As a result of the affordances of blog software, the process of publishing online diaries became feasible for a much larger, less technically oriented population. In Japan, the number of blog users has increased exponentially since 2003. This dramatic increase, which was labeled the “blog boom,” occurred with the development of the Japanese version of certain popular blog tools and the start of free blog services around 2002-2003. Recent research on Internet use estimates that 8,680,000 Japanese had registered with blog services as of the end of March 2006 (Ministry of Internal Affairs and Communications, 2006).

However, while many blogs are started, a considerable proportion are abandoned after a short period of time (Perseus, 2003). This study is interested in bloggers who persist in blogging. Specifically, the goal of the study is to clarify the psychological and social processes motivating blog authors to continue to write their personal blogs. In order to do so, we conducted a questionnaire survey of Japanese blog authors and examined two hypothesized models of psychological and internal factors (personal traits/benefits/satisfaction) and social and external factors (feedback from readers) using structural equation modeling. The hypothesized models are confirmed, albeit weakly so, suggesting that in addition to the hypothesized variables, other unmeasured motivational variables may have influenced the intentions of blog authors.


Previous Studies of Online Diaries and Personal Blogs

Many researchers have focused on the characteristics of the online diary and personal blog (e.g., Herring, Scheidt, Kouper, & Wright, 2004, 2006; Langellier & Peterson, 2004; Yamashita & Fujinami, 2000). One might think of personal blogs as media primarily used for self-description. In reality, personal blogs often contain information that authors would be unlikely to disclose in face-to-face communication, even with close friends and family. This suggests that blogs have characteristics in common with conventional diaries, such as contents concealed in closed spaces. However, releasing such text as web content means that it can be viewed anonymously by many people and implies that the author wants the contents to be known. Thus, blogs are multi-faceted, with both reader- and self-oriented features (Kawaura, Yamashita, & Kawakami, 1999).

Previous studies have classified online diaries or blogs based on their contents. For example, Kawaura et al. (1998) classified online diaries into four types based on “the direction of diarists’ consciousness” and “expression content:” 1) records of facts oriented toward the self (memoirs), 2) records of facts oriented toward readers (journals), 3) expressions of sentiments oriented toward the self (narrowly defined diaries), and 4) expressions of sentiments oriented toward readers (open diary). Krishnamurthy (2000) proposed classifying blogs into four basic types based on whether they are personal vs. topical and individual vs. communal. The two dimensions proposed by Krishnamurthy resemble the classification by Kawaura et al. (1998) of online diaries.

Blood (2002) claimed that the free-form interface of blogs caused a shift from a filter-style (link-based, externally focused) blog to a journal style, with both styles coexisting on the web. Following Blood (2002), Herring, Scheidt, Bonus, and Wright (2004) analyzed the contents of 357 random blogs and identified three sub-genres according to the purpose of the blog: personal journals, filters, and k(nowledge)-logs; personal journals were found to be overwhelmingly most common. Along with this classification, Herring and Paolillo (2006) conducted a multivariate analysis of entries from random blogs in a sample balanced for author gender and two of the three blog sub-genres (diary or filter) and found that authors of diary blogs write differently from authors of filter blogs; diary style correlates with hypothesized female authorial style, and filter style correlates with hypothesized male authorial style.

Nardi, Schiano, Gumbrecht, and Swartz (2004) investigated the various motivations that drive individuals to create and maintain blogs. They took account of a handful of alpha blog authors who were very popular and widely read but also paid special attention to ordinary authors with much smaller audiences. Based on in-depth interviews with 23 blog authors in the U.S., they identified and categorized five major motivations for blogging: documenting one’s life, providing commentary and opinions, expressing deeply felt emotions, articulating ideas through writing, and forming and maintaining community forums.

Kawaura et al. (1999) conducted a questionnaire survey of online diarists in Japan and empirically explored the psychological and social process of online diary writing behavior. The authors formulated a causal model of why diarists continued to write their online diaries. In their model, a psychological and social causal relationship was hypothesized: Some benefits of writing diaries satisfy authors, and this satisfaction compels them to continue writing. Furthermore, Kawaura et al. hypothesized that the self-consciousness of authors and feedback from readers would mediate this causal relationship. Their results revealed that online diarists continued to write their diaries because they had discovered the function of self-disclosure through the diaries. They felt satisfied that they had expressed themselves to others and always envisioned the presence of readers when they wrote (Kawaura et al., 1999). In addition, feedback from readers affected all factors that influence the intention to continue. These results demonstrated that online diaries in Japan were not merely a form of self-description but also involved active communication with others.

Kawaura et al.’s survey was conducted in 1997, before the word “blog” was known in Japan. It seems plausible that their conclusions can be extended to personal journal style blogs, as well.

Psychological and Social Processes Supporting the Intention to Continue Blogging

Individuals have less capacity to transmit information than do the mass media, yet they can transmit information freely over the Internet by publishing blogs. Blogs, together with tools such as social networking systems and wikis, put the public on notice that it can play a central role in the next-generation web, that is, Web 2.0. If blogs actually can make the Internet a space with extensive and varied accumulated information, each blog author should provide as much and as diverse information as possible in their blogs. In order for this vision to be achieved, it is crucial that blog authors continue to write their blogs without interruption. The longer they maintain their blog, the richer the information that accumulates in the blog. In reality, however, most blog authors make only one or a few entries in their blog and then stop writing in it. According to a U.S. survey (Perseus, 2003), 66.0% of 4.12 million surveyed blogs had not been updated in two months, representing 2.72 million blogs that have been either permanently or temporarily abandoned. What are the mechanisms that support personal blog authors’ intentions to persist in writing their blogs?

Generally speaking, a blog is a diary-like website that is updated on a regular basis. Through frequent updating, the contents of a blog become richer and more meaningful. Therefore, it may be reasonable to assume that most blog authors start blogging with the intention of continuing to write as long as possible. However, as the Perseus (2003) study reported, some blog authors abandon writing blogs in a few days or weeks, whereas others continue over several years or (potentially) all their life. What is the difference between these two groups? To investigate this, we assumed that both psychological and social factors form the mechanism of their intention to continue writing blogs. A causal model was hypothesized that includes both these factors as independent variables and the individual’s intention to continue writing blogs as a dependent variable.

When we determine whether to start a new action or continue an action that we have started already, we consider what benefits those actions will bring us in the future. If those actions are expected to provide us some benefits, we will undertake or continue them. If not, or if it seems the actions will be disadvantageous to us, we will not start them, or else we will abandon them. If the same is true for blog writing, people who expect benefits from blog writing will set up their own blogs, and those who benefit from blog writing and derive satisfaction from it will be more likely to continue blogging than those who do not.

As mentioned above, a blog both provides information about its author and has features that support mutual communication with others. Kawaura et al. (1999) focused on the communication function of online diaries (e.g., email forms, bulletin boards), suggesting that it caused authors to self-disclose and further promoted their intention to continue writing. Characterizing online diary writing as a kind of self-disclosing behavior, the authors pointed out two benefits of self-disclosure for online diary writing based on a previous study about the appropriateness of self-disclosure (Derlega & Grezelak, 1979). One is the benefit to self, and the other is the benefit to relationships with others.

Benefit to Self

Most articles written in a personal blog are based on the personal experience of its author. A survey conducted by the Pew Internet & American Life Project (Lenhart & Fox, 2006) revealed that the most popular topic among blog authors was their life and experiences. As many previous psychological studies (e.g., Pennebaker & Beall, 1986) point out, writing about our personal experience can help us to understand ourselves more deeply and mitigate major problems or conflicts. Therefore, it might be expected that blog authors can deepen their understanding of themselves through expressing or disclosing themselves in a blog. The more they benefit this way, the more they should feel satisfaction with blog writing. This should reinforce their intention to continue writing.

Benefit to Relationships with Others

Blogs also have a communication function; when someone transmits information on their own blog, they create opportunities for others to participate in their behavior, for example by leaving comments or pinging TrackBacks. If readers exploit this opportunity, the blog author may create or maintain closer relationships with them. This may also help the authors identify themselves as having a social existence. As blog writing behavior is directly connected with belonging to an interactive community through a blog, it might be expected that blog authors can benefit from activating communication with, and consequently gaining acceptance from, others. The more authors benefit continuously in this way, the more they should feel satisfaction with blogging, which should reinforce their intention to continue writing.

The effect of these benefits on the intention to continue writing was previously examined and confirmed based on a survey of online diary authors by Kawaura et al. (1999). This study examines whether it will be confirmed also in the case of blog authors. In the current study, we propose a benefit in addition to these two benefits of blog writing and develop a causal model to explain continued blog writing that includes all three benefits.

Benefit to Information Handling Skills

Another benefit that may be expected from blog writing is for authors to improve their information handling skills. In writing their blog and participating in a blogosphere, blog authors have repeated experience in appropriately transmitting information, leaving comments on other’s blogs, pinging TrackBacks, and obtaining new information from other blogs listed on their blogrolls. Through such blog activities, authors may benefit by improving their information handling skills; for example, they may develop an enhanced sensitivity to information and an improved ability to retrieve information.

As previously described, blog authors can manage their information efficiently by making use of various functions of the blog system such as RSS and topic-dependent categorization. Although they need to acquire a certain level of proficiency to utilize these functions fully, it is reasonable to assume that authors gradually learn the skills for handling information through their actual blog writing. Once a blog author acquires the skills for handling information, he or she will have opportunities to transmit and obtain information more effectively through blogging. The more authors benefit continuously in this way, the more satisfied they should be with blog writing, which should reinforce their intention to continue writing.

The central idea of our causal model for continuing blog writing is summarized as follows: The most important factors supporting the intention to continue blog writing are the degree to which blog authors are satisfied with the three benefits to (1) self, (2) relationship with others, and (3) information handling skills. When blog authors are conscious of, and accordingly feel satisfied with, these benefits of their blog writing, their intention to continue blog writing is enhanced.

We also consider factors related to these benefits and satisfaction: psychological factors such as personality traits of blog authors related to how much they are conscious of these benefits and social factors such as frequency of feedback from others related to how much they communicate with others via their blogs.

Psychological Factors

Individuals who are highly conscious of their self-expression in their blogs seem to be keen to know who they are. We assume private self-consciousness to be a personality trait related to the benefit to self. Private self-consciousness is the tendency to be aware of covert, private aspects of self that cannot be externally observed by others (Fenigstein, Scheier, & Buss, 1975). Individuals with high private self-consciousness tend to attach a higher value to personal than social identity (Cheek & Briggs, 1982) and behave in accordance with their own feelings, attitudes, and perceived rights (e.g., Scheier, 1976). In blog writing, high private self-consciousness authors may tend to start keeping a blog with an introspective motive and may more clearly anticipate a benefit to self, especially the ability to understand themselves more deeply.

Individuals who are highly conscious of being understood by others seem to be keen to verify it. A personality trait related to the benefit to relationships with others is reassurance seeking. Reassurance seeking is a tendency to demand significant contact with others excessively in order to reassure oneself of one’s worth, and it is a relatively stable pattern of motivation and behavior with one’s significant others (Joiner, Metalsky, Katz, & Beach, 1999; Katsuya, 2004). It is supposed that high reassurance-seeking individuals are likely to seek to confirm frequently whether they are loved of or value to others through various kinds of explicit behavior. In blog writing, it may be that high reassurance-seeking authors tend to start blogging with the motive of confirming their value to others, and they may expect blogging to benefit their relationships with others who read their blog and to enable them to attain deep mutual understanding with others.

We propose that the benefit to information handling skills is related to the degree of need for information. Information need consists of one’s need both to provide and acquire information (Kawaura, 1998). Blogs can be seen as one of the most suitable tools for satisfying this kind of need because blog authors can not only provide information easily to their readers, but they can also acquire information from their readers through comments and TrackBacks in their blogs. Therefore, those who have a strong need either to transmit their own information or to obtain information from others as much and as precisely as possible would be likely to expect a benefit to information handling skills from blog writing.

Social Factors

As discussed above, blog writing is a social behavior that enables mutual communication with others based on posted articles. Given this, hypothesizing a causal model for persisting in blog writing based only on individual psychological factors would be short sighted. Blog authors have frequent opportunities to receive feedback from readers through standard blog functions such as comments and TrackBacks. The amount of feedback they receive should have a considerable impact on blog authors’ satisfaction with their blogs. If some blog authors receive a lot of positive feedback (e.g, sympathy, support, encouragement), they should feel more satisfied and motivated to continue writing. Other blog authors who receive negative feedback (e.g., criticism, complaints, and quibbles) should feel less satisfied and less motivated to continue writing.

Hypothesized Causal Model of Intention to Continue Blog Writing

Summing up the discussion in the previous sections, we hypothesized two causal models for continuing to write a blog; these are displayed in Figures 1 and 2. The model in Figure 1 is related to benefits to self and relationships with others corresponding to the model previously confirmed by the survey of Japanese online diarists done by Kawaura et al. (1999). The other, in Figure 2, relates to information handling skill, as examined in the present study through a survey of Japanese blog authors. To test the validity of these models, a questionnaire survey of blog authors who were engaged in ongoing blog writing at the time was conducted.

Figure 1.

Hypothesized model of benefits to self and relationships with others

Figure 2.

Hypothesized model of benefit to information handling skills



The participants in this survey were individuals who had kept and published some sort of blog on the Internet. All were users of the Internet service Hatena (, which since March 2003 has provided a free blog service called the “Hatena diary” ( and which had the largest number of registered users in Japan at the time of this study. In Hatena Diary, 28,541 IDs (reflecting the actual number of blogs created) had been issued at the time the questionnaire was disseminated on March 1, 2004.

Hatena provides various content services, such as manpower search (an online forum in which Internet users exchange questions and answers without charge), web content update information antenna (a tool that tracks a webpage to determine whether or not it has been updated), web groups, and social bookmarks. In Hatena Diary, blog contents are updated from web pages that only individuals with IDs and passwords can access. When a certain amount of text is posted under a date, an HTML file is automatically created, making it possible to publish easily without any technical knowledge. Multiple topics can be written about on the same day, and, as categories can be specified for each topic, readers can also browse entries by category. In addition, services such as comment and TrackBack functions make it possible to use diaries to communicate, making Hatena Diary as a whole a large “blogosphere.” A typical example of a blog in Hatena Diary is illustrated in Figure 3.

Figure 3.

Screenshot of a typical blog in Hatena Diary


With the collaboration of Hatena Co., Ltd., we sent emails to all of the registrants of Hatena Diary on March 1, 2004, asking them to take part in our survey. They were asked to access the website address listed in the email and fill in the survey form posted there ( As only those who accepted our request participated in the survey, the survey was based on a self-selected sample. The replies were automatically stored in our CSV-format data files. The website for the survey was accessible from March 1 to 14, 2004.


The questionnaire consisted of items based on Likert-type or multiple-choice methods. The participants were asked to answer a questionnaire made up of four web pages.

To measure the degree of need to be understood by others, three items using a five-point scale were prepared that referred to the Japanese revised version of the reassurance-seeking scale (Katsuya, 2001, 2004). In the following three scales, the same questionnaire items that Kawaura et al. (1999) prepared (using a five-point scale) were used. There were questions on private self-consciousness (three items), the benefits to self from blogging (three items), and the benefits to relationships with others from blogging (five items). To determine degrees of satisfaction, a single item (five-point scale), similar to that used by Kawaura et al. (1999), was prepared, which questioned the extent to which self expression/understanding by others had been achieved through blog writing. The statement about satisfaction derived from self-expression and disclosure was “I can actually express myself sufficiently in my blog.” The statement about satisfaction with understanding from others was “I can actually be sufficiently understood by others through my blog writing.”

To measure the need to acquire information and the need to provide information, nine items on a five-point scale were prepared based on Kawaura’s (1998) scale of informational behavior. In addition to the seven original items in Kawaura’s scale, we prepared two new items. The new items were “I like to monopolize information” (reversed item) and “I am willing to provide any information I have.” To measure the benefits of blog writing for information handling skills, six items on a five-point scale were prepared that included skills related to information acquisition and provision. A single item (five-point scale) was prepared that asked about the degree to which users were satisfied with their ability to provide or acquire useful information using blogs. The statement about satisfaction with information acquisition was “I can actually acquire useful information by writing my blog.” The statement about satisfaction with information provision was “I can actually provide useful information by writing my blog.” Reader feedback was categorized into positive and negative feelings, three items for each. Feedback with positive feelings included sympathy, support, and encouragement. Feedback with negative feelings included criticism, complaints, and quibbles. For each of these six feedback items, the frequency was assessed using a four-point scale.

Finally, a single item (five-point scale) that directly questioned the intention to continue blogging was prepared as an item for the final target variable of the causal model. The statement was “I would like to keep writing my blog.”


Demographic Traits of Participants

There were 1,434 responses to the survey. The IP addresses of all the responses were recorded, and accesses from computers with the same IP addresses were viewed as accesses from the same responder. Only answers with the latest time stamps were taken, and earlier answers were deleted. Responses with missing data were also excluded from analysis. As a result, we had 1,142 responses for analysis.

The demographic traits of the responders were as follows: 783 males (68.6%) and 359 females (31.4%); 119 teenagers (10.4%), 602 people in their 20s (52.7%), 339 people in their 30s (29.7%), 67 people in their 40s (5.9%), and 15 people in their 50s (1.3%). Five hundred and sixty-four people (49.4%) had already started blogging in 2002, before the “blog boom” came to Japan, and 578 people (50.6%) started in 2003 or later. Therefore, the response sample had almost equal numbers of responders from the online diary generation and the post-blog-arrival generation.

Reliability of Scales

Before evaluating the hypothesized causal model, we estimated reliability of the scales included in the model. The items and the reliability index (Cronbach’s coefficient alpha) of each scale are shown in Tables 1 to 3.

Table 1.  Reliability of scales for benefits and personality traits
ItemsCronbach’s alpha
Benefit to self
I can sort out my feelings.0.73
I can know my true feelings. 
I can get over my complaints or conflicts. 
Benefit to relationships with others
I can get to know sympathetic others.0.81
I can communicate with others honestly. 
I can not only let others know who I am but also let myself know who others are. 
I feel relieved when others understand me. 
I can receive comments about my strengths/weaknesses from others. 
Private self-consciousness
I am always trying to figure myself out.0.72
I reflect about myself a lot. 
I am alert to changes in my mood. 
Reassuarance seeking
It is important for me to receive positive comments from the people I feel close to.0.77
It is important for me to always be accepted by the people I feel close to. 
It is important for me to receive consideration from the people I feel close to. 
Table 2.  Factor loading and reliability of scale for benefit to information handling skills
ItemsFactor LoadingCronbach’s alpha
Benefit to information handling skill
I can discriminate truth from error of information.0.730.83
I don’t find information search a heavy burden.0.72 
I can get information in a short time.0.71 
I can increase my knowledge by collecting information.0.69 
I can heighten my social sensitivity.0.65 
Extracted item
I can feel my social efficacy.0.37 
Table 3.  Factor loading and reliability of scale for information need
ItemsFactor 1Factor 2Cronbach’s alpha
Need for information acquisition
I spare no effort in obtaining information.0.71−0.120.77
I like to know about something in detail.0.76−0.09 
I always gather information before doing anything.0.480.04 
Need for information provision
I am willing to provide any information I have.0.100.690.79
I like to tell others any knowledge I have.0.370.45 
I like to monopolize information. (reversed item)0.180.47 
Extracted items
I feel ashamed if I don’t know things others generally know.0.200.08 
I like to have some genre I know better than anyone else does.0.360.13 
I like to express my opinion actively.0.280.34 
Inter-Factor Corr: −0.36

The scales for benefit to self, relationship with others, and information handling skills consisted of three, five, and six items, respectively. The reliability of all items included in the two antecedent scales was sufficiently high, as shown in Table 1. However, based on the results of factor analysis using the maximum likelihood method, shown in Table 2, one item representing social efficacy in the scale of benefit to information handling skills was not homogeneous with the other five items. The item was excluded from further analysis.

The scales of psychological factors related to benefit and satisfaction, private self-consciousness, reassurance-seeking, and information need were estimated using 3, 3, and 9 items, respectively. The reliability of all items included in the two antecedent scales was sufficiently high, as shown in Table 1. Factor analysis using the maximum likelihood method and promax rotation on the scale of information need was conducted, and two factors were extracted, as shown in Table 3. As three of the nine items did not have sufficient factor loadings with both factors, they were excluded from further analysis. The remaining six items were divided into the two factors. Factor 1 is need for information acquisition and factor 2 for information provision. Reliability of each factor was sufficiently high.

Structural Equation Modeling

We hypothesized the two causal models of intention to continue blogging shown in Figures 1 and 2, both of which consist of latent and measured variables, and applied the models to the actual survey data from 1,142 participants to confirm their validity. We confirmed them by applying structural equation modeling. Each of the factors included in the models is represented by a latent variable (ellipses in the figures) composed of multiple measured variables, or a measured variable (rectangles in the figures) consisting of a single item. The measured variables that constituted the latent variable were selected based on the analysis of reliability described above. For feedback-related latent variables, the measured variable for each of the assumed two types of feedback (positive feeling, negative feeling) was prepared. As shown in Table 4, the measured variables of satisfaction in each causal model had significant but low or moderate correlation coefficients.

Table 4.  Correlation matrix of measured variables of satisfaction
Variables of satisfactionSelf-understandingAcceptance from othersInformation provisionInformation acquisition
Acceptance from others
Information provision 1.000.38
Information acquisition 1.00

We first analyzed each hypothesized initial model. To search for the optimal final models, we then deleted the hypothesized paths within a reasonable range that referred to the goodness-of-fit indices, modification indices, and the results of significance tests of the set path.

Model Related to Self-Description and Interaction with Others

There was no path with an insignificant coefficient in the hypothesized initial model, and the goodness-of-fit indices were sufficiently high (GFI = .94, AGFI = .92, CFI = .93, RMSEA = .05). The final model, the same as the initial one, with estimated standardized coefficient values is shown in Figure 4. All paths shown in the figure were statistically significant (p < .05).

Figure 4.

Final model of benefits to self and relationships with others

Model Related to Information Handling Skills

Based on the significance tests of the set path, there were two paths that had insignificant coefficients in the hypothesized initial model: negative feedback to satisfaction with information provision (beta = .04, p = .13) and acquisition (beta =−.05, p = .27). The modified model deleting these two paths was calculated. The goodness-of-fit of this modified model was sufficiently high (GFI = .93, AGFI = .92, CFI = .92, RMSEA = .05), as shown in Figure 5, with estimated standardized coefficient values that are all statistically significant (p < .05).

Figure 5.

Final model of benefit to information handling skills

As indicated in Figures 4 and 5, the coefficients of all four satisfaction variables for intention to continue blog writing, although relatively low, were significant. There are two possibilities that may explain this weak level of prediction. One is that there were other unmeasured motivational variables that more strongly influence the intentions of blog authors. Another possible explanation is that the intention to continue blog writing for the blog authors in our sample was generally high and had small variance (M = 3.82, SD = 0.65). In interpreting the meaning of these results, such possibilities need to be taken into consideration.


Interpretation of the Two Causal Models

To clarify the psychological and social processes motivating authors to continue to write personal blogs, we conducted a questionnaire survey of Japanese blog authors and examined two hypothesized models using structural equation modeling. As a result, two final models with good fit were obtained. Both models confirmed that being satisfied with the hypothesized benefits had a significant positive effect on the intention to continue blog writing. The psychological traits supposed to make blog authors conscious of the benefits also confirmed their significant positive effect. In contrast, different patterns of the effects of feedbacks from readers were observed in the two models. One commonality was that positive feedback from readers had a positive effect on all kinds of satisfaction. This suggests that positive feedback from readers, for example sympathy, support, or encouragement, worked as a strong emotional social support on the behavior of publishing a personal blog and motivated a person to continue to be an author. However, negative feedback, for example criticism, complaints, or quibbles, did not have significant effects on satisfaction with information provision and acquisition, although it did have significant negative effects on satisfaction with self-understanding and acceptance from others. Unlike subjective events related to self or interpersonal relationships wedded to self, with objective event-related information, we suggest that the authors tended to attribute negative feedback not to themselves but to the quality of their information. Consequently, negative feedback had an insignificant effect on their satisfaction.

In this study, we confirmed the psychological mechanism supporting the mentality of blog writing and the social effects of feedback on it. The most obvious difference between writing a diary privately and online is the possibility of having a large global audience. Needless to say, that is only a possibility; often a blog author has a limited number of readers in reality. Satisfaction from being accepted by others had the strongest effect on the intention to continue blog writing. This might suggest that there is additional significance in blog writing for authors beyond the merely personal act of diary writing.

We should also note that our respondents were exclusively Japanese who were living in Japan. There should be both commonalities and differences across cultures. The basic framework of causal models of the psychological process in blog writing is similar to that in the motivation study by Nardi et al. (2004) and may be common across cultures. In contrast, the finding that negative feedback from readers had a negative effect on satisfaction with self-understanding and acceptance from others may be culturally specific to Japan.

In the context of cultural psychology, people in Western cultures have independent self-construals and tend to think of themselves as autonomous individuals separate from others. In contrast, those in non-Western, primarily Asian, cultures have interdependent self-construals and are more likely to think of themselves in the context of the larger social world, tending to define themselves in terms of their group memberships and relationships with others (Markus & Kitayama, 1991). Such cultural differences in self-construals have a strong influence on behavior in daily life (Kitayama, Markus, Matsumoto, & Norasakkunkit, 1997). Negative feedback from others would have a stronger meaning for one’s emotions and behavior in interdependent Japanese culture, where there is a great need to cooperate with others, than in an independent, individualistic culture. As a consequence, individual behavior, such as self-expression in blogs, would be more strongly impacted. Thus the negative effect of negative feedback on blog writing may be a result that is specific to Japanese culture, although this speculation has not yet been subjected to empirical study. To clarify whether there are cultural differences in social influences on blog writing, a cross-cultural survey needs to be done.

Limitations and Directions for Future Research

Although its results are useful in understanding the motivations of Japanese blog authors, this study has known limitations. The most obvious limitation is that our survey used a self-selected convenience sample; this point must be taken into account in generalizing the results. As previously described, although there are many people who start blogs, relatively few continue actively (Perseus, 2005). It is possible that blog authors who began blogs but soon lost interest in them did not read our request to complete the survey or did not have any interest in the survey, even if they read it. Therefore, the survey sample of this study might have been biased toward the minority of blog authors who were actively updating their blogs at the time of the survey. In estimating the motivational factors of blog writing, it is also important to consider why some blog authors stop blogging. To correct this bias, a survey with random or systematic sampling should be done in the future.

In our study, by conducting a horizontal survey of blog authors, we constructed and tested two causal models of psychological and internal factors (personal traits/benefits/satisfaction) and social and external factors (feedback from readers), which were hypothesized to motivate authors to continue blog writing. However, as noted above, since our questionnaire focused on authors who are currently blogging, we obtained no data from users who had never set up blogs or from those who had set them up in the past but had already stopped writing. Another survey with more appropriate sampling and a detailed case study targeting a wider range of users is needed to determine the sources of web behaviors, such as wanting to write blogs in the first place, or wanting to stop blog writing.

The hypothesized and confirmed causal models in this study included the psychological and social factors related only to the particular behavior of blog writing. For further theoretical development, it is necessary to consider the effects of blog writing on the social lives of the authors. A well-known study of the negative effects of Internet use on traditional social life claimed that there was an Internet Paradox: The more frequently people use the Internet, the more they reduce their social involvement and psychological well-being (Kraut, Patterson, Lundmark, Kiesler, Mukopadhyay, & Sherlis, 1998). This claim sparked a controversy, and many have contested the study’s results. For example, McKenna and Bargh (2000) argued that the participants in Kraut et al.’s study did not constitute a truly random sample, possibly biasing their results. It is still unclear whether the use of the Internet has an overall more positive or negative effect on users’ social lives. In light of our results for blog writing, it might be said that continuing to blog has a positive effect on the social life of authors, because their blog writing means (at least the cognition of) deeper self-understanding and acceptance by others. It can reasonably be predicted that such cognition would improve personal psychological well-being and reduce loneliness. To test this prediction, further research is needed.

Furthermore, when various kinds of information are accumulated and categorized by topics in personal blogs and shared with many blog authors in the blogosphere, both our online and offline lives can be enriched by being exposed to these kinds of information. For peoples’ internal lives, for interpersonal relationships, and for the social aspect of information databases, a major advantage in continuing to write blogs is the continued accumulation of their logs on the Internet. Changes or development over time in either authors or their environments should have a significant influence on this process. Fully understanding the characteristics of blogs and the significance of persisting in writing them will be possible only after including additional findings obtained from approaches featuring a vertical perspective, for example a panel survey, in which it is possible to determine time-series changes and evolutionary trends.

About the Authors

  1. Asako Miura is an associate professor in the Department of Psychology at Kobe Gakuin University, Japan. Her research interests are online communication, human-computer interaction, and group creativity.Address: 518 Arise Ikawadan-cho Nishi-ku Kobe, 651–2180, Japan

  2. Kiyomi Yamashita is a professor in the School of Network and Information at Senshu University, Japan. Her research interests are self-expression in CMC, the network society, and the family.Address: 2–1–1 Higashi-mita Tama-ku Kawasaki, Kanagawa, 214–8580, Japan