Abstract
Introduction
Literature Review
Methods
Results
Discussion
Conclusion
References
Full Text PDF

The purpose of this study was two fold. The first part of the study was to examine the credibility, emotional tone and depiction of the military revealed in milblogs by performing a content analysis. The second part of the study was to perform an experiment to observe the elicited credibility, emotional tone and perceptions of the military produced by milblogs versus traditional media forms.  The study was undertaken to answer some fundamental questions regarding the use of blogs as a means of communication, particularly about military events and people.  As the Internet increasingly grows as an information source it is imperative that communicators attempt to understand who uses blogs as a means of communication and what types of information are passed through blogs.  The experimental phases of the project sought to conduct analysis between the breadth and accuracy of information from three distinct types of internet news sources as a means of evaluating their mass media impact.
 Content Analysis
            Procedures. A content analysis was conducted of blogs written by military personnel.  Military personnel are defined as active duty military, reserve military, family members and retirees. The analysis focused on milblogs which were determined to be the most frequently mentioned as military blogs by other bloggers and readers.  This was determined by a cross reference check of milblogs that were recognized during the 2004 Military Weblogs Awards, by the website milblogging.com and those mentioned as sources during reporting on military blogs by Army Times Publishing Company, USA Today and Wired Magazine.  The 10 most frequently mentioned sites by these three sources were chosen for study.
The unit of analysis was each single post to the milblog about a person or event. The posts were chosen to be analyzed if they fell in the dates dictated by the study. 
It was determined to analyze milblogs across a six month period to ensure the information analyzed was consistent, as opposed to in response to one particular news event. The samples of Internet based news stories were stratified by month and week. The final week of analysis coincided with the Iraqi Constitutional Referendum vote.
Six DoD public affairs personnel, who were enrolled in the Joint Communication Course
at a Midwestern university, conducted the content analysis. Coding norms were established during supervised training sessions conducted using a representative sample of complete military blog enteries for two key weeks. During training, coders established a high degree of standardization resulting in inter-coder reliabilities of (a=.97), based on the Effective Intercoder reliability scale (Rosenthal, 1987).
Variables coded. The investigation examined four variables.  Overall tone of coverage toward the military was assessed with a global attitude measure adapted from Burgoon, Cohen, Miller, and Montgomery (1978).  It consisted of six 7-interval scales (a=.95), including: good/bad, positive/negative, wise/foolish, valuable/worthless, favorable/unfavorable, and acceptable/unacceptable. Depiction of military personnel in the milblog was assessed using the Individualized Trust Scale (ITS), developed by Van Lear and Trujillo (1986) based on four, 7-interval items including trusting/untrusting, candid/deceptive, and sincere/insincere, and honest/dishonest (a=.98).
            Multiple item indicators were used to evaluate emotion. The emotion scale was based on the previous work of Dillard and colleagues (Dillard, Plotnick, Godbold, Freimuth & Edgar, 1996; Smith & Dillard, 1997). Featured emotions and alpha reliabilities were: anger (angry, irritated, and annoyed), a = .93;  surprise (surprise, astonished, and amazed ), a = .90; puzzled, (puzzled, bewildered and confused), a = .90; sad, (sad, dreary and dismal), a =.85; fear, (fearful, afraid, and scared), a = .95; happy, (happy and cheerful), a = 84.; contentment, (mellow, tranquil, peaceful and contented), a = .89; pride, (dignity, honor, and gratification) a = .91; and humor, (funny, witty, and amusing),  a = .86. The dimensions of pride and humor were added to Dillard’s measures for the purpose of this particular study.           
Credibility of the milblog was measured using the news credibility scale (Gaziano & McGrath, 1986). The scale consists of  12-item indicators on a semantic differential scale bounded by 1 for more negative words and 7 for more positive words. The word pairs included: unfair and fair, biased and unbiased, doesn’t tell the whole story and tells the whole story, is inaccurate and is accurate, invades people’s privacy and respects people’s privacy, does not watch after readers’ interest and does watch after reader’s interests, is not concerned about the community’s well being and is concerned about the community’s well-being, does not separate fact from fiction and does separate fact from fiction, cannot be trusted and can be trusted, is opinionated and is factual, is concerned about profit and is concerned about public and reflects a poorly trained reporter or well trained reporter. The effective intercoder reliability for all measures was a = .97.  The total n = 528.
Experiment
The study sought to compare the effects of three versions of the same news story about these issues. The versions were: a story from a milblog, a story from a military news agency and a story from a traditional on-line media source such as the New York Times, Washington Post, or L.A. Times. The topics covered in the stories were the Iraqi constitution, Iraqi military or explosive devices.
These three topics were chosen as important news events that parallel stated U.S. and coalition goals in Iraq.  The three topics chosen included: the writing and ratification of the Iraqi constitution, the training of the Iraqi military, and reports on  improvised explosive devices (IEDs). IEDs were chosen as the third topic due to their representation of the insurgency in Iraq as being contrary to coalition goals.  Word counts were edited to be of comparable length to reduce bias. Word counts of the three news stories were 400 words each.
Participants.  Participants in the study were drawn from the pool of communication students at a Midwestern university and students were awarded extra credit for participating in the study.The study was conducted in one phase. A total of 326 participants reported for the study and completed an optional questionnaire. The questionnaire recorded demographic information on research participants to include gender (170 males, 156 females), age (mean 20.37 years), and year in school (63 freshman, 111 sophomores, 86 juniors and 69 seniors). Additionally, an exposure and attention measure of blogs, newspaper, and TV, was also collected to determine whether participants were familiar with the type of information that is found on these Internet based sources and to gauge their perception of the credibility of these news sources.  Finally, participants were asked about their attitudes regarding U.S. military presence in Iraq and the importance of continued U.S. military presence in Iraq.
            Next, participants were randomly assigned to one of three events: Iraqi constitution, Iraqi military or explosive devices. Participants were then assigned to read a story from either a milblog, the military news source, or the traditional newspaper source. Subjects were assigned randomly with the exception that care was taken to ensure conditions were relatively balanced in regards to initial involvement.  The participant’s attitudes toward the military were record prior to their exposure to the story.
            After reading the story, research participants were again asked about their attitude toward the U.S. military presence in Iraq as well as the importance of continued U.S. military presence in Iraq.  Participants were also asked their perceptions of the credibility of the story and their emotional response to the story they read.  Finally, they were asked about their uses of blogs, and their attention to newspaper news, television news, and blogs and if they had discussions with parents, friends, or in classes about the information they read on the blogs.
Measures. Overall attitude about U.S. military presence in Iraq was assessed before and after message exposure with a global attitude measure adapted from Burgoon, Cohen, Miller, and Montgomery (1978). The measure’s six-point bipolar adjective scale. Multiple item indicators were used to include; unacceptable/acceptable, foolish/wise, unfavorable/favorable, negative/positive, bad/good, and wrong/right (before exposure, a =.96; after exposure a = .97 ). The importance of continued U.S. military presence in Iraq was measured before and after message exposure using a six-point bipolar adjective scale (Zaichkowski, 1985). The scale included; unimportant/important, or no concern/of much concern, means nothing/means a lot, doesn’t matter/matters to me, insignificant/significant, and irrelevant/relevant (attitude before, a = .95, attitude after, a = .96).
Credibility of the milblog was measured using the news credibility scale (Gaziano & McGrath, 1986). The scale was a 12-item indicators on a semantic differential scale bounded by 1 for more negative words, and 7 for more positive words. The word pairs included: unfair and fair, biased and unbiased, doesn’t tell the whole story and tells the whole story, is inaccurate and is accurate, invades people’s privacy and respects people’s privacy, does not watch after readers’ interest and does watch after reader’s interests, is not concerned about the community’s well being and is concerned about the community’s well-being, does not separate fact from fiction and does separate fact from fiction, cannot be trusted and can be trusted, is opinionated and is factual, is concerned about profit and is concerned about public and reflects a poorly trained reporter or well trained reporter,  a = .87.
Multiple item indicators were used to evaluate emotion. The emotion scale was based on the previous work of Dillard and colleagues (Dillard, et. al., 1996; Smith & Dillard, 1997). Featured emotions included anger (angry, irritated, and annoyed) a = ,93 ., surprise (surprise, astonished, and amazed) a = .90 ., puzzled (puzzled, bewildered and confused) a =  .90 , sad (sad, dreary and dismal) a = .85 ., fear (fearful, afraid, and scared) a = .95, happy (happy and cheerful) a =.84, contentment (mellow, tranquil, peaceful and contented) a = .89, and pride (dignity, honor, and gratification) a = .91 and humor (humorous, funny and witty) a = .86. The dimensions of pride and humor were added to this measure for the purpose of this particular study.