The goal of the study was to test young adults’ perceptions of the credibility of two different types of blogs institutional blogs versus personal blogs. The results of the study showed that young adults don’t perceive institutional blogs to be more credible than personal blogs. In addition, the findings suggest that young adults view the news information on the Internet and cable television as being nearly equally credible. The findings of the present study as well as previous studies suggest that that audience acceptance and attention to the perceived credibility of blogs and other non-professional, user-created content on the Internet continues to grow rapidly. This type of information content has and continues to become an important component in the marketplace of ideas. Introduction In the last two decades, the Internet has become an important tool in our daily lives for communication and acquiring information (Metzger, 2007). At the same time, the quality and credibility of the information available on the Internet has also become a concern for online information seekers. In recent years, weblogs, a kind of online diary websites, which are becoming an intrinsic part of the Internet (Johnson, Kaye, Bichard & Wong, 2007), has begun to attract more attention of mass communication researchers. Winer (2003) defines a weblog as “a hierarchy of text, images, media objects, and data, arranged chronologically, that can be viewed in an HTML browser” (Ho, 2006). Internet users can read blog content and share opinions with bloggers and other readers. This up-to-date type of media is growing rapidly since its emergence in 1998. More than 110 million blogs currently exist (Chu & Kamal, 2008), and 175,000 new blogs are estimated to be launched every day (Chu & Kamal, 2008). Generally, blogs can be divided into two categories: personal blog and institutional blog. A personal blog is created, owned, and operated by an individual for his or her own interest; usually not for commercial purposes. A citizen-blogger is totally independent of any institution (Mackay & Lowrey, 2007). In contrast, an institutional blog is created and owned by an organization for direct and applied purposes. It can be operated by one or more members of that institution. Institutional bloggers usually don’t tend to include strong personal opinion and the writing mission is part of the job assignment with goals that benefit the institution. Qing & Oyedejj Credibility Perceptions of Weblogs GMJ 2 Blog credibility has become a popular research topic in recent years as blogs have come to challenge traditional media and other types of online sources as an important information source (Johnson & Kaye, 2008). Singer (2006) stated that the impact of blogs has already exceeded its readership given the fact that leading journalists use influential blogs for various kinds of information and story tips. Studies on blog credibility often associate blog credibility with blog reliance, political attitudes, user demographics, type of information on blogs etc. Scholars argue that, “credibility is not a characteristic of a source, medium, or message, but is dependent on the perceptions of the receiver” (Johnson et al., 2007). Studies have found that bloggers consider blogs more credible than do non-bloggers (Banning & Trammell, 2006). Bloggers tend to believe that information provided by blogs is more credible than other media (Johnson & Kaye, 2004); users who relied on blogs often judged them to be more credible, etc. In addition, previous research has discovered that people use different standards to judge the credibility of different websites (Flanagin & Metzger, 2000). For example, people assess personal blogs using different criteria than those used for institutional blogs (Banning & Sweetser, 2007). Little research has been done to compare the relative credibility of personal and institutional blogs. The present study will use an experimental design and survey to explore young adults’ perceptions of the relative credibility of institutional blogs versus personal blogs. There are two working assumptions. First, young adults represent a major component of Internet users and blog users. Second, understanding their perception of blog credibility will help media professionals learn more about the young group they covet before making strategic decisions on adopting different news types and user blogs on their websites. Literature Review Credibility has been examined from many different angles such as message credibility, which focuses on how message characteristics affect the perceptions of a message’s believability or its source’s believability (Metzger, Flanagin, Eyal, Lemus & McCann 2003); and source credibility, which focuses on the believability of a communicator (O’ Keefe, 1990). The main focus of this study is blog credibility which is a subset component of media credibility. Media Credibility There is extensive literature in the field of mass communication that has focused on media credibility. The findings of previous studies suggested that some certain types of factors can influence people’s perception of media credibility. Demographic variables were found to be related to the perception of media credibility. Westley and Severin (1964), who initially compared the channel credibility among different media outlets, found that certain demographic variables such as age, education, and gender could influence media’s perceived credibility. In addition, some scholars suggest that the perception of media credibility is also influenced by media use; specifically, people who often use a certain type of media will perceive the media as more credible than those who don’t use it often (Cobbey, 1980; Greenberg, 1966; Shaw, 1973). Similarly, unfamiliarity with a certain medium has been found to lead to a low level of perceived credibility (Bucy, 2003; Rimmer & Weaver, 1987). However, Wanta and Hu (1994) did not find a significant correlation between the amount of media use and credibility, but they found that medium dependency is related to credibility. Chaffee (1982) argues that interpersonal communication could affect media credibility for the studies that did not find a correlation or found a negative correlation between channel use and credibility. McLeod, Rush, and Friederich (1968) supported these findings and stated that interpersonal communication patterns may explain the limited effects of media use on credibility. Kiousis (2001) found a moderately negative relationship between the amounts of interpersonal communication that respondents devote to discussing news and their perceptions of credibility of television news. However, the author did not find similar results for newspaper and online news credibility. Armstrong and Collins (2009) found that family encouragement could affect perceived local news credibility but not college news credibility. Fall 2011 Global Media Journal Volume 11, Issue 19 3 The perception of media credibility can also be influenced by different technological features. Previous research has found that audiences considered television relatively more accurate and less biased than other media (Gaziano & McGrath, 1986; Wilson & Howard, 1978). One reason is that television’s visual nature helps increase its credibility ratings (Carter & Greenberg, 1965). Gunther (1988) also argued that television makes its news stories appear more objective than newspapers. Furthermore, Chang and Lemert (1968) argue that live reports of breaking news give audiences a greater sense of authority and importance of the news stories. People, therefore, are more likely to view television news as being more credible than newspapers stories. In addition, different criteria used by people to judge news credibility can influence media credibility (Newhagen & Nass, 1989). It was found that people usually judge the credibility of television news by an evaluation of the journalists or anchors who deliver the news, but they determine the credibility of print media by evaluating the entire institution with no human face. Therefore, scholars have found that the credibility of television news tends to be rated higher than that of newspapers (Kiousis, 2001). Internet Credibility In the past two decades, the increasing popularity of the Internet has attracted a lot of attention from scholars in the area of mass communication. Previous studies have examined several different issues related to credibility: website design features (Johnson & Kaye, 1998, 2000; Shon, Marshall & Musen, 2000), source attributions (Sundar & Nass, 2000, 2001), news story design (Lowrey, 2004) and website genre (Flanagin & Metzger, 2007). Moreover, some other credibility studies have explored issues related to users’ characteristics and behaviors like users’ demographic variables such as age, gender, ethnicity, and income (Bucy, 2003); users’ reliance on online information (Johnson & Kaye, 2000, 2002); users’ behaviors of online information verification (Flanagin & Metzger, 2000; Rosenthal, 1971); the relationship between users’ web experiences and their perception of online information credibility (Flanagin & Metzger, 2000; 2007); and the effects of users’ motivations toward certain media and types of information (Mulder, 1980). Many studies have explored the difference between the perceived credibility of web-based information and that of information obtained from traditional media. Previous Internet credibility studies have examined a variety of online information types, and most of them have focused on news information (Kiousis, 2001; Metzger, Flanagin & Zwarun, 2003; Sundar, 1999). The research on Internet credibility to date provides mixed conclusions. Some studies have found that the Internet is considered to be more credible than traditional media. Greer (2003) found that Internet users consider news information from online news sites to be more credible than information from conventional media. At the same time, other studies have found Internet-based news to be less credible than other media types. A Roper study for the Freedom Forum found that respondents viewed leading traditional sources as being less biased than web-based information (John & Kaye, 1998). Other studies have suggested that people typically do not judge online information or traditional media information as being more credible than the other (Flanagin & Metzger, 2000; Johnson & Kaye, 1998; Kiousis, 2001; Metzger et al., 2003). To further examine the difference of the perceived credibility between the Internet and traditional media, the technological characteristics of the two types of media need to be understood. The Internet shares some advantageous features with television and newspapers (Metzger et al., 2003). The Internet can provide both text and visual elements like pictures and videos, which more easily induces users to believe the information provided online. In addition, similar to television, the Internet can update information quickly or bring live news reports. Unlike newspapers, the Internet is not restricted by time and space constraints, so it can provide a large amount of in-depth news reports. Nevertheless, the Internet also shares some disadvantageous features when compared to television and newspapers. Since the Internet Qing & Oyedejj Credibility Perceptions of Weblogs GMJ 4 always provides newly updated information very quickly, the chance of making mistakes also increases (Nadarajan & Ang, 1999). Online information is similar to newspapers in that it is not regulated and people enjoy much freedom to express ideas and post information. As a result, people are more likely to doubt its objectivity (Metzger et al., 2003; Nadarajan & Ang, 1999). To conclude, there are no consistent findings for the relative credibility of web-based information compared with traditional media like television. Blog Credibility As the population of Internet users grows rapidly, blogs are receiving ever-greater attention from researchers. Blogs provide a new platform for users to share information and exchange opinions about current social events and other issues (Kaye, 2005). Blogs often provide more up-to-date news and other information than traditional media (Hamilton, 2003). Moreover, blog readers seek information on various social and political issues which traditional media may not be able to provide (Cristol, 2002). The characteristics of the information on blogs have also been examined. First, users may find that information on blogs is more credible because blogs allow relatively independent expression (Regan, 2003). Though corporate-controlled media may seek to avoid discussion of social issues to avoid offending audience segments, bloggers are free to discuss any issues (Bruns, 2006; Johnson & Kaye, 2004; Wall, 2006). Second, bloggers can also offer stories from anywhere in the world that traditional media may have ignored (Johnson & Kaye, 2004; Sands, 2003). Third, though journalists consider blogs as a source of news leads, they still have criticized bloggers for not abiding by professional or ethical rules and contend that bloggers should improve their skills (Singer, 2006). Some studies have examined the credibility of different blog types. Researchers have suggested that differences exist in the level of credibility among different types of blogs (Johnson & Kaye, 2008; Kaye & Johnson, 2009). Some demonstrated that the perception of the credibility of institutional blogs were lower than that of personal blogs for several certain types of information. For example, Johnson and Kaye (2008) found that war and military blogs created by individuals and soldiers in Iraq were perceived to be the most credible, followed by the blogs of media organizations or journalists. Corporate blogs were considered to be the least credible of these types. Flanagin and Metzger (2000) argue that this is because corporate blogs usually reflect their commercial interests and cannot be trusted to provide impartial information. Kaye and Johnson (2004) suggest that media and journalist blogs trail behind military blogs because the authors of those are usually conservative. Military bloggers may receive the highest credibility rating because they are perceived to have a good understanding of military culture and warrelated events (Johnson & Kaye, 2004; Kaye & Johnson, 2009). On the contrary, some researchers have argued that blog readers might be more interested in media blogs because journalists are more professional and have quick and easy access to news events (Flanagin & Metzger, 2000, 2008). This supports the argument of Flanagin and Metzger (2007) who argue that web-based information provided by an individual on personal websites might be viewed to be narrow and less representative of other people’s opinions; thus, leading to extremely low credibility assessment. However, a recent study conducted by Banning and Sweetser (2007) found no significant differences in credibility ratings among media blogs, personal blogs, online newspapers, and traditional newspapers. The above literature review indicates that many studies of blog credibility have classified blogs based on the information source; thus, a brief review of source credibility is given in the following section. Source Credibility Source credibility has been defined by researchers as “judgments made by a perceiver concerning the believability of a communicator” (O’ Keefe, 1990, pp. 130-131; Wilson & Sherrell, 1993). Scholarly interest in credibility research began in the 1950s with inquiry into the way credibility functions in the persuasion process (Metzger et al., 2003). Previous studies found that higher source prestige increases the likelihood that people will agree with statements (Hovland & Weiss, 1951). Scholars later focused on how source credibility is related to interpersonal Fall 2011 Global Media Journal Volume 11, Issue 19 5 influence, and the way an organizational source’s credibility impacted human behavior (Metzger et al.). Empirical researchers who have studied source credibility defined the concept of credibility in different ways. Hovland and his colleagues defined credibility as a communicator’s expertise and trustworthiness (Hovland, Janis, & Kelley, 1953). Expertise refers to a speaker’s qualifications or capabilities to learn about issues or topics, while trustworthiness refers to the perception of a speaker’s motivation to tell the truth about a topic (Hovland, Janis, & Kelley, 1953). This laid a conceptual foundation for future researchers to use factor analysis as a measure of credibility. Some secondary dimensions of credibility include accuracy, fairness, completeness, knowledge, and dynamism. Beyond these, some scholars identified two variables that are directly related to perceived trustworthiness and expertise: similarity and liking of the source (Aune & Kikuchi, 1993; McCroskey, 1966; O’ Keefe, 1990; Widgery & Webster, 1969). Researchers also explored how the effectiveness of these dimensions or factors influences attitude change (Metzger, 2003). Wilson and Sherrell (1993) found that source expertise is the most influential factor. However, Lui and Standing (1989) found that more trustworthy sources have more credibility than sources with more expertise. With regard to blog credibility, researchers have argued that source attributes are important for evaluating information on the Internet (Sundar & Nass, 2000, 20001). In addition, Flanagin & Metzger (2007) also argued that it is essential to differentiate between different online source types, because the results of credibility evaluation by information receivers may be different depending on which source attributes are salient when they are evaluating the given information. Research Questions and Hypotheses Although the findings of previous studies are not consistent, the research questions and hypotheses for the present study are presented below: Research question 1: What are respondents’ perceptions of the relative credibility of different media types? Little research has been done with regard to examining readers’ perceived credibility of news information on institutional and personal blogs; thus, the goal of the present study is to explore whether participants’ perceptions of the credibility of news information is related to the source. In addition, Banning and Sweetser (2007) have argued that blogs emerged as a popular source for political information since the 2004 U.S. presidential election and the use of blogs for political information searching has been increasing. Moreover, a survey conducted by Blogads network in 2006 broke out blog readers into four categories: politics, celebrity news, mom, and music blogs (Kaye, 2006). These facts showed that politics and entertainment are two main message types on blogs. Therefore, this study focused on two main types of information on blogs: political and entertainment information. Thus, the following hypotheses were proposed: Hypothesis 1: Political messages posted on institutional blogs are perceived to be more credible than those posted on personal blogs. Hypothesis 2: Entertainment messages posted on institutional blogs are perceived to be more credible than those posted on personal blogs. Methodology An experiment was conducted and then a survey was administered to gather the data for this study. The respondents were 253 undergraduate students enrolled in a large southeastern university. Seventy-four point seven percent (n = 189) of the participants were female, while 23.70% (n = 60) were male. A majority (84.6%, n = 214) was identified as Caucasian, followed by Qing & Oyedejj Credibility Perceptions of Weblogs GMJ 6 Asian (5.50%, n = 14), Hispanic (2.80%, n = 7), and African-American (2.40%, n = 6). Their ages ranged from 18 to 32 years, with a mean of 20.27 (SD = 1.47). The survey probed respondents’ perceptions of the media credibility of blogs and other media types, and gathered demographic and psychographic data from the respondents. The present study explored the difference in perceived credibility of two categories of blogs: institutional and personal blog, and of two message types on blogs: politics and entertainment. Thereafter, a 2 (blog type: institutional versus personal) X 2 (news type: political versus entertainment) factorial experiment was conducted to assess participants’ perceptions of the message credibility of similar messages on the two blog types. Two blogs a clone of www.cnn.com (CNN’s label was used on the blog to show it was an institutional blog) and a personal blog of a fictitiously named blogger, Andrew Garson were designed and updated with two news stories. The first story was a political story from CNN’s political ticker, the cable news network’s institutional blog, and the second news story was culled from a blog post on www.debbiewell.com. The political news story was about Vice President Joe Biden’s quip about Sarah Palin’s appeal as a presidential candidate, in which he called her “nice” but opined that she lacked the political knowledge to run for the highest office in the country. The entertainment story, titled “Too Fat to Fly”, was about Hollywood director Kevin Smith, who was kicked out of a Southwest Airline’s plane because of his weight. Careful consideration was given to the two types of stories selected for this study. First, the length of the two stories was moderate so that participants wouldn’t feel impatient before they finished reading the assigned story. Second, the two stories were equally long (both were three or four paragraphs) to ensure that the article’s length wouldn’t become a possible factor that would cause differences of response to the two stories. Third, both stories were easy to read and understand so that every participant would be able to understand the assigned story before they gave responses. Each participant was exposed to one of the blog stories and asked to indicate his or her level of agreement to five questions on the credibility of the news story using variables adapted from Meyer’s (1988) credibility dimensions. Participants’ agreement was measured on a seven – point scale (where 1 = “strongly disagree” to 7 = “strongly agree”). The experiment was a factorial design: each of the four groups of respondents received and responded to a single stimulus. A total of 253 participants completed the experiment. Fifty-eight respondents were randomly assigned to the institutional blog-political news treatment, 59 to the institutional blogentertainment news treatment, 68 to the personal blog-political news treatment, and 68 to the personal blog-entertainment news treatment. Results The first research question asked the respondents their perceptions of the relative credibility of the media types. Based on the literature review, in this study, we focused on five media types: Cable television, the Internet, Personal communication, Institutional blogs and Personal blogs. The descriptive statistics of the credibility scores of the five media types showed that participants perceived news information on institutional blogs (M = 4.78, SD = 1.24, n = 253) to be the most credible among the five media; followed by the Internet (M = 4.55, SD = 1.20, n = 253), cable television(M = 4.53, SD = 1.34, n = 253), and personal communication (M = 3.91, SD = 1.19, n = 253). Those surveyed perceived news information from personal blogs to be the least credible among the five media types (M = 3.19, SD = 1.25, n = 252). In addition, the medians and the modes of the credibility scores of the Internet, cable, and institutional blogs had an equal value of 5.00. The median and the mode of the credibility score of personal communication had an equal value of 4.00; and the media and the mode of the credibility score of personal blogs had an equal value of 3.00 (See Table 1.). The results indicate that the participants of this study consider news information from the Internet, cable, and institutional blogs to be relatively more Fall 2011 Global Media Journal Volume 11, Issue 19 7 credible. They perceived the information obtained from personal blogs to be relatively less credible. They were neutral about news information from regular people. Table 1. A Comparison of Perceived Relative Credibility of Five Media Cable Internet Personal communication Personal blogs Institutional blogs N 253 253 253 252 253 Mean 4.53 4.55 3.91 3.19 4.78 Median 5.00 5.00 4.00 3.00 5.00 Mode 5.00 5.00 4.00 3.00 5.00 Std. Deviation 1.34 1.20 1.19 1.25 1.24 To further explore the first research question, a Wilcoxon test was conducted. The Wilcoxon test, which is a non-parametric test, was chosen because there is some controversy within the statistical literature about whether or not Likert scale data is appropriate for parametric tests. The Wilcoxon test for the pair of variables “Personal communication” and “Internet” was statistically significant, z = -6.72, p < .01. The mean of the ranks in favor of “Internet” was 93.46, while the mean of the ranks in favor of “Personal communication” was 75.92. The results showed that participants perceived online information (M = 4.55, SD = 1.20, n = 253) to be more credible than information obtained from personal communication (M = 3.91, SD = 1.19, n = 253). In addition, the Wilcoxon test for the pair of variables “Personal communication” and “Cable” was also statistically significant, z = -6.17, p <.01. The mean of the ranks in favor of “Cable” was 93.91, while the mean of the ranks in favor of “Personal communication” was 82.93.The results showed that participants perceived cable news (M = 4.53, SD = 1.34, n = 253) to be more credible than information obtained from personal communication (M = 3.91, SD = 1.19, n = 253). However, the Wilcoxon test for the pair of variables “Internet” and “Cable” was not statistically significant, z = -0.09, p = 0.93, suggesting that participants did not consider there to be a significant difference of credibility between “Internet” and “Cable”. Moreover, the Wilcoxon test for the pair of variables “Personal blogs” and “Institutional blogs” was significant, z = -12.11, p < .01. The mean of the ranks in favor of “Institutional blogs” was 104.36, while the mean of the ranks in favor of “Personal blogs” was 50.07, which indicated that participants considered information on institutional blogs (M = 4.78, SD = 1.24, n = 253) to be more credible than information on personal blogs (M = 3.19, SD = 1.25, n = 252). The first hypothesis predicted that political messages posted on institutional personal blogs are perceived to be more credible than those posted on personal blogs. In order to test the hypothesis, the related summary data were generated through the Case Summaries command in SPSS to compare the credibility scores of the political news messages on institutional blogs and personal blogs (See Table 2). Because the data are not symmetrically distributed, the median values were chosen to compare the differences of the credibility scores of the two treatment groups. The median values of the five dimensions of credibility (fairness = 4.00, bias = 3.00, accuracy = 4.00, comprehensiveness = 3.00, and believability = 4.00) are the exact same for the blogs that appeared to be institutional as well as for those that appeared to be personal in relation to the experimental design. The median values of the overall credibility scores (divided the total scores of the five dimensions of credibility by five) of institutional blogs (median = 3.40) and personal blogs (median = 3.80) didn’t have obvious difference. The results of this analysis suggests that there is little difference in participants’ perceptions of the credibility of political Qing & Oyedejj Credibility Perceptions of Weblogs GMJ 8 stories in relation to source. In addition, the relatively apparent difference of the mean values of “comprehensives” between institutional blogs (M = 2.93, SD = 1.42, n = 58) and personal blogs (M = 3.50, SD = 1.68, n = 68) indicated that participants generally considered the news information on personal blogs to be more comprehensive than the information on institutional blogs.