Does Matching Versus Mismatching Message Content to Attitude Functions Lead to Biased Processing? The Role of Message Ambiguity

Abstract

Previous research on the question of whether matching message content to the functional basis of people's attitudes may lead to biased message processing has been inconclusive. In particular, existing evidence is open to reinterpretation such that matched strong arguments led to more attitudinal agreement because they were scrutinized more effortfully than mismatched strong arguments. The present study was conducted to examine the hypothesis that matching the message to attitude functions may lead to biased processing only given an ambiguous (vs. an unambiguous strong or weak) message. High and low self-monitors were presented with a matched message (i.e., a quality appeal for low self-monitors and an image appeal for high self-monitors) or a mismatched message (opposite combinations). The message content was strong, weak, or ambiguous. As predicted, only given an ambiguous message did biased processing lead to more agreement when the appeal matched (versus mismatched) attitude functions. In contrast, a strong message led to more agreement than a weak message regardless of functional matching (unbiased processing).

ACKNOWLEDGEMENTS

The authors would like to thank Franziska Drogla, Mandy Raumschüssel and Katharina Wais for their help in conducting this research.

Notes

¹Note that the persuasive messages themselves were not specifically designed to match or mismatch attitude functions. Rather, the messages were a mix of functionally relevant and irrelevant arguments.

²See Ziegler and Diehl (2003) for a discussion regarding the operationalization of an ambiguous message in terms of moderately persuasive arguments rather than a mix of strong and weak arguments.

³A preliminary Message type by Message strength analysis of variance did not reveal any effects on self-monitoring scores.

⁴That is, each dependent variable was regressed on the following predictor variables: functional argument matching (FAM), message appeal (MA), message contrast 1 (MC1), message contrast 2 (MC2), FAM × MA, FAM × MC1, FAM × MC2, MA × MC1, MA × MC2, FAM × MA × MC1, FAM × MA × MC2.

⁵Means and standard deviations are reported in the case of main effects and interaction effects of the categorical independent variables Type of Appeal and Message Strength. In the case of effects involving the continuous variable Match, predicted values are reported.

⁶The two-way interaction of functional match and message ambiguity in our analysis is identical to a three-way interaction of self-monitoring propensity, message appeal and the contrast of the ambiguous message versus the strong and weak messages in the case of a multiple regression analysis based on the three factors measured (self-monitoring) and manipulated (message appeal and message ambiguity), but this is not as meaningful conceptually.