Effects of Construal Level on Responses to Ambiguous Health Information about Alcohol Consumption

ABSTRACT

Information about the health effects of alcohol consumption can be ambiguous (i.e., lacking in reliability, credibility, or adequacy) and thus may promote maladaptive health behavior. Guided by Construal Level Theory and a conceptual taxonomy of uncertainty in health care, we tested the hypothesis that manipulating construal level would promote adaptive responses to ambiguous health information. We examined the effects of ambiguous health information about alcohol on health cognitions, message responses, and intentions, as well as whether manipulating construal moderated these effects. Alcohol users (n = 135, M_age = 20.15, 68.9% female) were randomly assigned to either a high-level or low-level construal task and then to read either an ambiguous or unambiguous health communication about the health effects of alcohol. Participants responded similarly to ambiguous health information as they did to unambiguous health information and participants in a high-level construal did not generally report differences compared with those in a low-level construal. Findings suggest that ambiguous health information might not always lead to maladaptive effects. More research is needed to examine moderators of the relationship between ambiguous health information and health outcomes, as well as to understand how and when using construal manipulations are effective in different health contexts.

Notes

1. Outliers were scores that were three standard deviations above or below the variable mean. As pre-determined in the pre-registered analytic plan, analyses were conducted both with and without any outliers to determine how outliers influence results while considering the full range of responses that might be collected across variables. When two outliers on the variable of age were removed, the results were consistent across all analyses except for one: there was a main effect of Health Communication on reactance that was counter to hypotheses, F(1,129) = 4.299, p = .040, ηp² = .032, such that participants who received an ambiguous health communication reported lower reactance (M = 1.61, SD = .043) than participants in the control condition (M = 1.80, SD = 0.60).