本研究探討在高、低涉入的情況下，來自內、外團體高共識的說服訊息對態度改 變的影響。 本研究採 2 （涉入程度：高╱低）× 2 （訊息來源：內╱外團體）× 2 （論 點品質：強╱弱）三因子實驗設計，所有實驗組皆呈現高共識的相反立場說服訊息。結果顯 示，高涉入情況的說服效果明顯小於低涉入。同時，在高涉入情況下，強弱論點的說服效果 沒有顯著差異，然而，在低涉入情況下，強論點的說服效果明顯大於弱論點。這種結果顯示 ，在高涉入情況下，受試者有抗拒說服的現象，而低涉入情況較不會引起抗拒，受試者會比 較客觀地處理訊息。徑路分析的結果顯示，相對於低涉入，高涉入造成個體的正向認知反應 比較少、情感反應較為負向、對訊息來源會有較不好的評價，因而造成抗拒說服的現象。而 訊息來源與論點品質只有經由對訊息來源的評價這個中介變項影響說服效果。外團體的訊息 來源使得受試者對其評價較差，也較沒有說服效果。訊息來自內團體時，強論點使得受試者 對訊息來源有較好的評價，也較具有說服效果，而弱論點則否。徑路分析的結果說明本研究 所操弄的三個自變項對說服的影響，不僅是經由認知反應的作用，還有經由受試者的情感反 應與對訊息來源的評價所導致。

　　　　　This study investigated the persuasive effects of in-group and out-group high consensus messages with varying argument strength under high or low issue involvement conditions. A 2 (high or low issue involvement) × 2 (in-group or out-group source) × 2 (strong or weak arguments) factorial design was used, and participants were exposed to a counterattitudinal message representing the majority viewpoint of in-group or out-group. Results showed that participants in high, relative to those in low, issue involvement conditions were less persuaded. In addition, participants agreed more with the strong message than with the weak one in low issue involvement conditions. However, the arguments did not differ significantly in their persuasibility under conditions of high issue involvement. These results suggest that participants in high issue involvement conditions were more resistant to the message, while those in low issue involvement conditions were less resistant to the persuasive appeal and processed the persuasive message in an objective fashion. Path analyses revealed that high (vs. low) issue involvement led to fewer positive cognitive elaborations, more negative affect, lower evaluations of the source, thus resulted in resistance to persuasion. The persuasive impacts of the source and argument quality were mediated by evaluations of the source. Out-group source was evaluated less favorably, and had little persuasive impact. Strong arguments from in-group source proved more persuasive than the weak one as a result of evaluating the source more favorably. Results of path analyses indicate that the persuasive effects examined were mediated not only by cognitive responses, but also through affective responses and evaluations of the source.