ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展 ›› 2019, Vol. 27 ›› Issue (5): 773-788.doi: 10.3724/SP.J.1042.2019.00773

• 元分析 • 上一篇    下一篇

内疚与亲社会行为的关系:来自元分析的证据

汤明, 李伟强, 刘福会, 袁博()   

  1. 宁波大学心理学系暨研究所, 浙江 宁波 315211
  • 收稿日期:2018-05-18 出版日期:2019-05-15 发布日期:2019-03-20
  • 通讯作者: 袁博 E-mail:yuanbopsy@gmail.com
  • 基金资助:
    * 国家自然科学基金(31600914);教育部人文社会科学基金(15YJC190026);天津市哲学社会科学研究规划项目资助(TJJX16-021)

The association between guilt and prosocial behavior: A systematic review and meta-analysis

TANG Ming, LI Weiqiang, LIU Fuhui, YUAN Bo()   

  1. Department of Psychology, Ningbo University, Ningbo 315211, China
  • Received:2018-05-18 Online:2019-05-15 Published:2019-03-20
  • Contact: YUAN Bo E-mail:yuanbopsy@gmail.com

摘要:

内疚作为一种典型的道德情绪, 被认为具有亲社会作用, 但很多研究却发现内疚并不总能促进亲社会行为。为了明确内疚对亲社会行为的作用, 分析造成结论分歧的可能原因, 本研究采用元分析方法探讨了特质内疚与亲社会行为的关系以及状态内疚对亲社会行为的影响。共有46篇文献92个独立样本纳入元分析(N = 17248)。元分析结果表明:(1)特质内疚与亲社会行为之间存在中等程度的正相关, 二者之间的关系受到亲社会行为类型的调节, 相比较捐赠、助人、环保行为等, 特质内疚与补偿之间的相关更强; (2)启动内疚状态能显著提升个体的亲社会行为, 但两者之间的关联呈较小的效应量, 亲社会行为对象在其中起到调节作用, 感到内疚的个体更愿意对受害方做出亲社会行为; (3) p曲线(p-curve)分析发现, 两个元分析研究的p曲线均呈显著右偏态, 表明特质内疚与亲社会行为的关系以及状态内疚对亲社会行为的影响均存在真实的效应, 而不是出版偏倚或者p hacking导致。

关键词: 内疚, 亲社会行为, 元分析, 调节效应, p曲线

Abstract:

A large body of research has examined the relationship between guilt and prosocial behavior - yet the findings are inconclusive. A meta-analysis was conducted to explore the effect of trait and state guilt on prosocial behavior, as well as potential moderators of this effect. A literature search yielded 46 qualified papers with 92 effect sizes and 17248 participants. Results showed that trait guilt was significantly associated with prosocial behavior (r = 0.36, p < 0.001), and this relationship was moderated by the type of prosocial behavior. The induction of state guilt also significantly impacted prosocial behavior, with a small effect size (d = 0.24, p < 0.001) - and this relationship was moderated by whether the recipient of the prosocial behavior was also the victim of the guilt-evoking act. No other moderators were found. The p-curve analysis showed that the p-curves of the two meta-analyzes were right skewed, indicating that the relationship between the guilt proneness and prosocial behavior and the effect of state guilt on the prosocial behavior had a real effect, not caused by publication bias or p hacking.

Key words: guilt, prosocial behavior, meta-analysis, moderate effect, p-curve

中图分类号: