Participants
        The pilot study consisted of 33 adult, non-public affairs respondents from three major U.S. military installations: Yokota Air Base, Japan; Fort Myer, Virginia; and Marine Corps Air Station Beaufort, South Carolina.  Thirteen respondents were from overseas, and 20 were stateside.  Seven of the respondents were in the Air Force; seven were in the Marine Corps; 12 were in the Army; one was in the Navy; and six of the respondents were family members.  Twenty-five of the respondents were male, and eight were female.

      Procedures
        For the pilot study, the survey was distributed to the internal audience, both overseas and stateside.  The authors distributed the survey to bases of current assignment, through co-workers, to non-public affairs individuals.  Admittedly, the convenience sample yields results with limited generalizabilty.  However, the pilot study does give researchers the opportunity to test the survey for reliability.

      Materials and Design
        The self-report survey included 25 questions and was adapted from Abelman, Atkin, and Rand (1997).  The survey has three demographic questions, two viewing habit questions, a general impression question, and 19 motivation questions.  The authors included two additional motivation questions: "I watch the global service newscast because it makes me feel patriotic," and  "I watch the global service newscast because it helps me relate better to the military."  A table of random numbers was used to determine the order of the motivation questions.
        To aid in analysis, the nineteen motivation questions were divided into five subgroups. This also results in a more reliable test because of the composites. A question can test for either goal-driven or ritualized/habitual viewing behaviors.  The survey had 10 goal-driven questions and nine ritualized/habitual questions.  The other three motivation subgroups are: information with four questions, escape with three questions, and moral support with five questions.  The patriotism question, question 18, fit into both information and moral support.

      Analysis
         The information motivation subgroup consisted of questions seven, 15, 20, and 22.  They had a .86 Cronbach alpha reliability coefficient.  The escape motivation subgroup consisted of questions eight, nine, and 24 (r = .68).  The subgroup for moral support motivation was made up of questions 12, 17, 18, 22 and 25 (r = .86)   The goal-driven viewing behavior subgroup contained questions seven, eight, nine, 12, 15, 17, 18, 20, 22, and 25 (r = .85).  Finally, the ritualized/habitual viewing behavior subgroup consisted of questions 10, 11, 13, 14, 16, 19, 21, 23, and 24 (r = .73).   The survey can be found in Appendix A.
         A composite mean score was derived for each subgroup to compare measures of central tendency for each variable.  For further comparisons, the t-test was used to test if there was a significant difference between the composite variables. Demographic data was compiled using frequencies of occurrence.