ISSN 0439-755X
CN 11-1911/B
主办:中国心理学会
   中国科学院心理研究所
出版:科学出版社

心理学报 ›› 2024, Vol. 56 ›› Issue (10): 1462-1470.doi: 10.3724/SP.J.1041.2024.01462

• 研究方法 • 上一篇    

变量之间的影响关系和多重影响因素的共同作用类型

温忠麟, 王一帆, 马鹏, 孟进   

  1. 华南师范大学心理应用研究中心/心理学院, 广州 510631
  • 收稿日期:2024-02-18 发布日期:2024-07-10 出版日期:2024-10-25
  • 通讯作者: 温忠麟, E-mail: wenzl@scnu.edu.cn
  • 基金资助:
    国家自然科学基金项目(32171091)、教育部人文社会科学重点研究基地重大项目(22JJD190006)资助

The influence relationship among variables and types of multiple influence factors working together

WEN Zhonglin, WANG Yifan, MA Peng, MENG Jin   

  1. Center for Studies of Psychological Application/School of Psychology, South China Normal University, Guangzhou 510631, China
  • Received:2024-02-18 Online:2024-07-10 Published:2024-10-25

摘要: 探讨研究变量之间的关系是心理学和其他社科领域实证研究的主要工作。基于问卷调查的实证文章多数涉及变量之间的“影响关系”, 但这个概念未有明确界定, 经常被人理解为相关关系或者因果关系, 无论哪种情况都会带来问题, 尤其在中介效应研究中更为突出。本文将影响关系定义为有方向性的相关关系, 从外延和内涵上揭示了相关、影响和因果三者之间的关系; 提供了多种途径为影响关系建模找到理据; 讨论了一个结果变量的多重影响因素的共同作用类型。本文将为问卷研究中的变量之间关系研究提供理论支撑。

关键词: 影响关系, 相关关系, 因果关系, 影响因素, 中介变量, 调节变量

Abstract: The investigation of relationships among variables is the main focus of empirical research in psychology and other social science disciplines. Many empirical studies based on questionnaire surveys involve the influence relationship between variables. However, the lack of a universally accepted definition for this concept has led to ambiguity, and it is often conflated with causal or correlational relationships, which leads to problematic, especially in the studies on mediating effects.
This article defines the influence relationship as a directional correlation, elucidating relations between correlation, influence, and causation in terms of denotation and connotation. Risk factors and protective factors are both influence factors of a negative outcome, and the impact increases with the level of a risk factor and decreases with that of a protective factor.
We summarize several ways to find evidence for modeling the influence relationship: (1) establishing directionality based on the temporal sequencing of variable occurrences; (2) testing the explanatory power by reversing the order of variables; (3) following the rule that object variables tend to influence subject variables; (4) considering attributes of variables (e.g., essential vs. state attributes, long-term vs. temporary attributes; stable vs. unstable attributes) to predict their influence; (5) gaining evidence from theoretical or empirical literature; (6) obtaining support from life experience and common sense; (7) reasoning through analogies; (8) applying principles of cross-lag analysis to identifying dominant factors.
Furthermore, we categorize multiple influence factors working together. These include independent effects, overlapping effects, two types of proxy effects, two types of mediating variables, and three types of moderating variables. These distinctions clarify how different influence factors work together to shape outcomes.
Some influence relationships exhibit characteristics akin to causal relationships, while others resemble correlation patterns. The degree to which different studies capture influence relationships close to causation may differ, affecting the quality of the research. Closer proximity to causal relationships enhances the informational value and significance of findings.

Key words: influence relationship, correlation relationship, causal relationship, influence factor, mediator, moderator

中图分类号: