我到底有多好？－－社會比較與時序比較對自我評價的影響

Am I Really Good?–The Impact of Social Comparison and Temporal Comparison on Self-Evaluation

本研究目的在探討社會比較與時序比較對自我評價的影響。在實際生活中，他人表現常影響個體評價自我，但自我評價也會受到個體自身由昨至今的變化所影響。我們提出一調節模式，假設時序比較影響效能相關信念，從而調節社會比較對自我評價的影響，以此說明兩類比較如何共同作用影響自我評價。實驗一藉由測驗回饋分數操弄不同比較訊息，92名參與者隨機分派至3（時序比較：進步／持平／退步）× 2（社會比較：向上／向下）各情境，結果顯示不同趨勢的時序比較訊息顯著影響個體的效能相關信念，而在自我評價上，除了主要效果達顯著之外，交互作用亦達顯著。實驗二嘗試釐清造成此調節效果的主因是基於時序比較訊息中所隱含的何種資訊內涵（趨勢變化或平均水準）所致，因此運用拮抗反差式的分數設計，75名參與者隨機分派至3（時序比較：進步─低平均／持平─中平均／退步─高平均）× 2（社會比較：向上／向下）各情境，結果呈現與實驗一相同的調節型態，揭示趨勢變化是首要關鍵因素。整體結果支持調節模式：社會比較訊息提供個體一個比較的相對「位置」，而時序比較訊息影響個體的效能相關信念，賦予這個「位置」不同的個人意涵，兩類比較共同影響最終的自我評價。

How individuals evaluate themselves? In the present study we proposed that the self-evaluation should be influenced by both social comparison and temporal comparison. To be exact, we suggested that different temporal performance trends would affect one's self-efficacy, and then moderated the impact of social comparison on selfevaluation. Two experiments were conducted to examine our hypotheses. In Experiment 1, ninety-two participants were randomly assigned to a 3 (temporal comparison: improvement/same/deterioration) × 2 (social comparison: upward/ downward) between-subjects design. A series of false performance feedbacks of an ability tests were used to manipulate different trends of temporal comparison information and others' average performance was also provided at the end as social comparison information. The results showed that, the trend of temporal performance feedback did affect one's sense of self-efficacy, controllability, and future expectation. As predicted, a significant interaction effect of social comparison and temporal comparison on self-evaluation was found. The Experiment 2 tried to replicate the moderate effect of temporal comparison and further investigated the importance of the temporal performance trend. Seventy-five participants were randomly assigned to a 3 (temporal comparison: improvement with low average/same with moderate average/deterioration with high average) × 2 (social comparison: upward/downward) between-subjects design. The results also showed a similar interaction effect as Experiment 1. Besides, the result demonstrated that it was the temporal performance trend, not performance average, moderated the impact of social comparison.