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: Exploring the relationships among variables is the main focus of empirical research in psychology and other social science disciplines. Most Empirical studies based on questionnaire surveys involve the influence relationship between variables. However, neither research methodologies nor common online encyclopedia entries provide a strict definition for this concept, leading to a lack of clarity in the understanding of the influence relationship. Some equate it with the causal relationship, while others consider it synonymous with the correlation relationship. Both cases can pose problems, particularly in the study of mediating effects. This article defines the influence relationship as a direction correlation relationship, elucidating relations between correlation, influence, and causation from the perspectives of denotation and connotation. It discusses the connection between influencing factors and risk factors, where risk factors are considered a subset of influencing factors. While theoretically defining risk factors as equivalent to influencing factors is feasible, equating risk factors with influencing factors may sometimes be challenging to explain. Usually, factors that may lead to adverse outcomes are considered risk factors. We provide several ways to find evidence for modeling the influence relationship: (1) Determining direction based on the chronological order of variable occurrences; (2)Following the rule that object variables influence on subject variables; (3) Supporting from theory or empirical literature; (4) According to attributes of variables: essential attributes influencing state attributes, long-term attributes influencing temporary attributes, and stable attributes influencing unstable attributes; (5) Reversing the order of two variables to see which one is easier to explain; (6) Reasoning through analogies; (7) Summarizing from life experience and common sense; (8) Using the principles of cross-lag analysis to identify dominant factors. We also discuss types of multiple influencing factors working together. These include independent effects, overlapping effects, two types of proxy effects, two types of mediating variables, and two types of moderating variables. Some influence relationships are closer to the causal relationship, while others are closer to the correlation relationship. The degree to which different studies obtain influence relationships that are close to the causal relationship may differ, affecting the quality of the research. The closer an influence relationship is to causation, the more information it can provide and the more likely it is to be valued.

Key words: influence relationship, correlation relationship, causal relationship, influence factor, mediating variable, moderating variable