Results

The purpose of this study was to examine the impact of broadcast news visuals of war. Specifically, it employed a content analysis to determine the overall tone of war coverage to describe what is covered in television news about the war in Iraq. In addition, it features an experiment to determine the impact of combat footage on viewers. Finally, it examines whether inoculation can protect viewers’ attitudes from the impact of broadcast news footage.

Descriptive Statistics

Research Question 1 examined the content of broadcast news coverage about Iraq. To examine this research question, the dates were broken down into three equal time frames. The first time period was the time period from January to July 2004. The second time period was October 2004 through June 2005, and the last time period was from July 2005 through March 2006. The coding categories were collapsed into five specific content focuses: progress in Iraq, the number of casualties, casualty personality, U.S. politics, and Iraqi politics. Table 1 indicates the proportion of coverage devoted to each content area across the three time intervals.

There was little coverage of progress in Iraq, which never exceeded 2% of coverage of Iraq. Progress in Iraq was covered slightly more often between October 2004 through June 2005 (M = 1.28; s.d. = 4.54 Time 2; M = 1.11; s.d. = 4.90 Time 3; M = 1.06; s.d. = 4.50 Time 1). Coverage of casualties was more frequent, especially after Summer 2004. The number of casualties was discussed most frequently between July 2005 and March 2006 (M = 10.89; s.d. = 13.94 Time 3; M = 7.27; s.d. = 11.03 Time 2; M = 6.90; s.d. = 10.96 Time 1). Casualty personality (e.g., Fallen Heroes stories) was discussed frequently between October 2004 through June 2005 (M = 13.41; s.d. = 22.20 Time 2; M = 11.67; s.d. = 21.51 Time 3; M = 6.94; s.d. = 17.41 Time 1). U.S. politics were not the focus of much coverage. U.S. politics regarding the war in Iraq was discussed most frequently from July 2005 through March 2006 (M = 2.22; s.d. = 8.46 Time 3; M = 1.90; s.d. = 6.73 Time 2; M = 1.89; s.d. = 7.69 Time 1). However, Iraqi politics was much more a focus of coverage, especially in the first phase from January to July 2004. Iraqi politics was discussed most frequently from January to July 2004 (M = 15.69; s.d. = 19.62 Time 1; M = 12.44; s.d. = 15.47 Time 2; M = 12.25; s.d. = 17.98 Time 3).

Research Question 2 asked how objective news stories about Iraq were over time. Overall, coverage was fairly neutral or balanced with few differences across time. More opinionated news coverage about the war in Iraq appeared during October 2004 through June 2005 (M = 4.56; s.d. = 1.98 Time 2; M = 4.67; s.d. = 2.23 Time 3; M = 4.83; s.d. = 1.53 Time 1).

Research Question 3 asked what the overall tone was of broadcast news coverage about the war in Iraq. On balance, tone of coverage was neutral. Tone of coverage became more positive from Time 1 to Time 2 and then leveled off (M = 4.61; s.d. = 1.06 Time 2; M = 4.57; s.d. = 1.55 Time 3; M = 4.17; s.d. = .99 Time 1). A Scheffe post-hoc test revealed that there was a significant change in tone from Time 1 to Time 2, t(114)4.11, p < .01.

Research Question 4 asked about the frequency of casualties featured in broadcast news coverage about the war in Iraq. Forty-four percent of the broadcast news stories coded depicted U.S. military casualties. Forty percent of the stories depicted Iraqi civilian casualties. Wounded Iraqi civilians were covered most frequently (53.6%) followed by wounded U.S. military (35.7%).

Research Question 5 asked if public opinion was also discussed when featuring reports about U.S. casualties about the war in Iraq included in the stories. Overall, coverage did not stress public opinion. Indeed, the only period it was emphasized was during October 2004 through June 2005 (M = 5.11; s.d. = 19.86).

Research Question 6 asked what type of frame would be employed when discussing military operations in Iraq. Framing was measured using a scale from 1 to 7 (Pfau, et al., 2004). Overall, broadcast news stories employed a slightly more thematic frame (M = 3.82; s.d. = 2.21 Time 2; M = 3.61; s.d. = 1.76 Time 1; M = 3.30; s.d. = 2.37 Time 3). Broadcast news stories were somewhat less thematic at Time 2 between October 2004 and June 2005.

Experiment Results

To assess Research Question 7 and Hypothesis 1, a one-way MANCOVA was computed for the independent variables of casualties (U.S. or Iraqi) and the dependent variables of post-attitude, post-issue involvement, emotional response, experiential processing, and rational processing, as well as the emotions anger, puzzlement, sadness, fear, and pride. Two covariates (initial attitude and sex) were also employed in this analysis. An omnibus test revealed significant differences for the covariate initial attitude Wilks’ λ F1,129  =  51.09, p < .001, partial eta2 .81.

Subsequent univariate tests indicated significant effects for initial attitude on the dependent variables post attitude F(1, 129) = 571.11, p < 001, partial eta2 .80; post-issue involvement F(1, 129) = 90.10, p < 001, partial eta2 .39; emotional response F(1, 129) = 21.98, p < 001, partial eta2 .14; anger F(1, 129) = 13.05, p < 001, partial eta2 .09; puzzlement F(1, 129) = 7.41, p < 01, partial eta2 .05; sadness F(1, 129) = 14.19, p < 001, partial eta2 .09; and pride F(1, 129) = 36.44, p < 001, partial eta2 .21. For initial attitude toward continued U.S. military presence in Iraq and elicited pride, the betas were positive, indicating that stronger initial attitudes were related to greater pride. Also, initial attitude was positively related to post-issue involvement. Betas were negative when examining initial attitude and the emotions anger, puzzlement, sadness, and emotional responses, indicating that those with a more positive attitude were less likely to experience negative emotions or emotional responses to broadcast new coverage.

 Omnibus tests also revealed significant differences for the covariate sex on the dependent variables emotional response F(1, 129) = 8.86, p < 01, partial eta2 .06; puzzlement F(1, 129) = 8.90, p < 01, partial eta2 .06; sadness F(1, 129) = 6.70, p < 05, partial eta2 .05; and fear F(1, 129) = 15.98, p < 001, partial eta2 .10. All betas were positive, indicating broadcast news coverage about the war in Iraq elicited more emotion from females.

Research Question 7 asked how broadcast news stories about U.S. versus Iraqi combat news stories were processed differently. Table 3 depicts a pattern of means that demonstrates that the focus on U.S. versus Iraqi combat resulted in little differences in how one processes television news stories.

Hypothesis 1 predicted that stories with footage of U.S. combat would elicit more negative affect than stories of Iraqi combat. There was no significant main effect for casualty focus. An omnibus test revealed no significant differences for the independent variable of the casualty condition Wilks’ λ F1,129  =  1.45, p  =  1.61, partial eta2 .11. Therefore, Hypothesis 1 was not supported.

To examine Hypothesis 2 and 3 a 2 x 2 MANCOVA was computed with the independent variables of participant sex (male or female) and visual condition (no picture of casualties or picture of casualties) and the dependent variables post-attitude, post-issue involvement, emotional response, experiential processing, and rational processing, as well as the emotions anger, puzzlement, sadness, fear, and pride. An omnibus test revealed significant differences for the covariate initial attitude Wilks’ λ F1,129  =  49.29, p < .001, partial eta2 .81. Subsequent univariate tests revealed significant differences for the covariate initial attitude and the dependent variables post-attitude F(1, 129) = 555.22, p < 001, partial eta2 .80; post-issue involvement F(1, 129) = 89.55, p < 001, partial eta2 .39; emotional responses F(1, 129) = 22.04, p < 001, partial eta2 .14; anger F(1, 129) = 12.57, p < 01, partial eta2 .08; puzzlement F(1, 129) = 6.72, p < 05, partial eta2 .05; sadness F(1, 129) = 13.86, p < 001, partial eta2 .09; and pride F(1, 129) = 35.83, p < 001, partial eta2 .21. For initial attitude toward continued U.S. military presence in Iraq and elicited pride, the betas were positive, indicating that stronger initial attitudes were related to more positive subsequent attitudes and greater pride. Also, attitude was positively related to post-issue involvement. Betas were negative when examining initial attitudes and the emotions anger, puzzlement, sadness, and emotional responses, indicating that those with a more positive attitude were less likely to experience negative emotions or emotional responses to broadcast new coverage.

An omnibus test revealed significant differences for the independent variable participant’s sex, Wilks’ λ F1,129  =  2.29, p < .05, partial eta2 .17. Subsequent univariate tests revealed significant differences for the independent variable sex and the dependent variables emotional responses F(1, 129) = 6.08, p < 05, partial eta2 .04; experientially processing F(1, 129) = 3.86, p < 05, partial eta2 .03; and the emotions puzzlement F(1, 129) = 7.30, p < 01, partial eta2 .05; sadness F(1, 129) = 5.90, p < 05, partial eta2 .04; and fear F(1, 129) = 15.66, p < 001, partial eta2 .10. Examination of the pattern of means revealed that females processes television news stories with more emotional responses and more experiential processing. Table 4 illustrates the pattern of means.

Hypothesis 2 predicted that television news stories featuring visual footage of combat would be processed more experientially. The results failed to support this hypothesis. There was no difference in processing for those viewing stories with visual footage of combat and those not seeing combat footage.

Hypothesis 3 posited that television news stories featuring visual footage of combat exert negative influence or attitudes about continued U.S. military presence in Iraq compared to stories without combat footage. There was no main effect for experimental condition. However, when theory posits a particular effect, subsequent planned comparison tests are warranted (Huberty & Morris, 1989). Planned comparisons of the two viewing conditions indicated that news stories with visual footage of combat significantly undermined support for continued U.S. military presence in Iraq compared to stories without footage of combat, F (1, 138)  =  16.76, p < .01, h2  =  02. This supports Hypothesis 1.

Hypothesis 5 predicted that inoculating with print messages plus photo would produce better effects than inoculating with only a print message. An omnibus test revealed significant differences for the covariate initial attitude, Wilks’ λ F11, 80  =  39.09, p < .001, partial eta2 .84. Subsequent univariate tests revealed significant differences for the covariate initial attitude and the dependent variables post – attitude F(1, 95)  =  438.69, p < 001, partial eta2 .83; emotional responses F(1, 95)  =  16.54, p < 001, partial eta2 .16; and the emotions anger F(1, 95)  =  7.72, p < 01, partial eta2 .08; sadness F(1, 95)  =  6.92, p < 05, partial eta2 .08; and pride F(1, 129)  =  25.00, p < 001, partial eta2 .22. For initial attitude toward continued U.S. military presence in Iraq and elicited pride, the betas were positive, indicating that stronger initial attitudes were related to more positive subsequent attitudes and greater pride. Betas were negative when examining initial attitudes and the emotions anger, sadness, and emotional responses, indicating that those with a more positive attitude were less likely to experience negative emotions or emotional responses to broadcast new coverage.

There were no main effects for the independent variables of inoculation condition or gender, therefore Hypothesis 5 was not supported.

Hypothesis 4 predicted that inoculation reduces negative affective and attitudinal influence. More data needs to be collected in the control condition to determine if this is actually true. Every participant who was inoculated also saw footage of casualties. Those in the current control condition did not see footage of casualties, only generic footage about the war in Iraq. The data to be collected will expose participants to the footage of casualties without inoculating them prior to seeing the television news broadcasts.