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