Abstract
Perceived exposure has been offered as the main factor in how people estimate the effects of media messages on others, but a recent study found this did not hold for two prosocial messages. This study (N = 160) shows that demographic cues in anti-drug messages may communicate who the intended targets are, thus influencing perceived exposure and perceived effects for different age, gender, and racial groups. In turn, perceived effects on the self and others predict support for funding anti-drug campaigns. The possible impact of perceived similarity between a group and the “implied audience” on perceived effects is discussed.
Notes
∗p < .05;
∗∗p < .01;
∗∗∗p < .001.
∗∗p < .01;
∗∗∗p < .001.
a Predisposition is perceived attitudes of a group toward drug use.
∧p < .10 (.05 one-tailed);
∗p < .05;
∗∗p < .01;
∗∗∗p < .001.
a Terms in parentheses indicate the higher values of the predictors.
∧p < .10;
∗p < .05;
∗∗p < .01;
∗∗∗p < .001.
1 Correlations between the two items ranged from .42 for 13 year olds to .87 for males. The other groups and their inter-item correlations were as follows: 16 year olds, .45; people your age, .71; parents of teens, .68; females, .83; Whites, .85; African Americans, .69.
2The F ratio and degrees of freedom reported here and in other contrasts used the conservative Greenhouse-Geisser adjustments because sphericity could not be assumed.
3 The message-by-group interaction also is significant at the p < .001 level when age, gender, and racial groups are run in separate ANOVAs. When racial groups are run separately, there is no main effect for group. Message does not have a significant main effect in any case.
4 Thirteen year olds (M = 3.18, SD = 1.72) and parents of teens (M = 1.92, SD = 1.49) were seen as more predisposed against drug use than 16 year olds (M = 5.25, SD = 1.70) and people their own age (M = 6.09, SD = 1.71), all t > 13, p < .001. Females (M = 4.71, SD = 1.49) were seen as less favorable toward drug use than males (M = 6.06, SD = 1.55).
5 Reliabilities were lowered when the indexes included perceived effects items for “Rough Night,” the ad targeted at parents. Its effects were measured as influence on the likelihood of confronting someone else about their drug use. Accordingly, these measures were dropped for this analysis.
6 Despite the relatively high correlation between perceived effects on self and others, diagnostics indicated collinearity was not a serious problem with the regressions predicting support for funding anti-drug messages or enforcement. The lowest tolerance was .39 in these models, well above the common cutoff of .1.