The U.S. Department of Defense employs both uniformed military personnel and non-uniformed civilian employees as information sources. The objectives of this study was to determine whether students, acting in the role of journalists, attributed greater credibility to uniformed or non-uniformed spokespersons and whether a difference in attribution could be measured when the topic being briefed was more specifically related to the military mission.Seventy undergraduate journalism students were randomly assigned to four groups and exposed to one of four videotaped press briefings. Two briefings announced the outbreak of hostilities involving U.S. forces or award of a major construction contract. Each of the announcements was delivered by a uniformed military public affairs officer or by a spokesperson in civilian business suit.Following the briefings, students evaluated the source using semantic differentials first developed by Berlo, Lemert and Mertz (1969) and prepared questions exactly as they would ask them following the spokesperson's prepared statement. The semantic differentials were analyzed using ANOVA. The follow-on questions were coded using methodology similar to that used by Einsiedel (1974) and evaluated using the "Coefficient of Imbalance" proposed by Janis and Fadner (1949). This second method was employed to determine whether data obtained and analyzed using the Coefficient of Imbalance would validate results obtained through the use of more traditional semantic differentials.Neither method resulted in findings which would suggest a statistically significant difference in the credibility attributed to the defense source by the student-journalists in any of the four treatments. / Department of Journalism