9. Hayes Process Macro SPSS - Model 7 - Moderated Mediation Analysis
Based on Research With Fawad's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.
Role ambiguity moderates the organizational commitment → organizational learning relationship, not the organizational learning → organizational performance link directly.
Briefing
Moderated mediation analysis in SPSS (Hayes PROCESS Model 7) shows that role ambiguity changes how organizational commitment translates into organizational performance through organizational learning—so the indirect effect is not constant across levels of the moderator. In this setup, organizational commitment (IV) affects organizational learning (mediator), which in turn affects organizational performance (DV). Role ambiguity (moderator) specifically moderates the strength of the IV→mediator path, meaning the commitment-to-learning link—and therefore the commitment→performance pathway via learning—weakens as role ambiguity rises.
The analysis begins with the moderated path to the mediator. Organizational commitment significantly predicts organizational learning, and the interaction between organizational commitment and role ambiguity is also significant. The moderation term corresponds to a measurable change in organizational learning (reported as a 3.1% change in the criterion variable). Role ambiguity carries a negative influence on the organizational commitment→organizational learning relationship, indicating that higher role ambiguity undermines the effectiveness of commitment in producing learning.
Johnson–Neyman intervals and simple slope results clarify where the moderation is statistically meaningful. At low and average levels of role ambiguity, the organizational commitment→organizational learning relationship is significant and relatively stronger. As role ambiguity increases, the slope remains significant but becomes much weaker; the relationship “flattens” at higher role ambiguity levels. Johnson–Neyman output identifies a threshold region: beyond a certain moderator value, further increases in role ambiguity no longer produce a statistically significant change in the commitment→learning effect. A plotted visualization reinforces this pattern: the blue (low) and red (average) lines show steeper gradients than the green (high) line.
Turning to the outcome side, organizational learning significantly predicts organizational performance, and organizational commitment also shows a significant direct effect on organizational performance even after accounting for the mediator. However, the moderator does not appear in the direct-effect portion for organizational performance; the key question becomes whether role ambiguity changes the indirect effect. The indirect effect of organizational commitment on organizational performance through organizational learning is significant at low, average, and high role ambiguity levels, but its magnitude declines as role ambiguity increases (strongest at low role ambiguity, smaller at average, and smallest at high).
The decisive test is the index of moderated mediation. The index is negative (reported as −0.0564) and its confidence interval does not include zero, confirming moderated mediation. That result supports the hypothesis that the indirect effect of organizational commitment on organizational performance via organizational learning depends on role ambiguity. In practical terms: organizational commitment can improve performance through learning, but that pathway is most potent when role ambiguity is low, and it weakens as role ambiguity rises.
Cornell Notes
The analysis uses Hayes PROCESS Model 7 to test moderated mediation: role ambiguity changes the indirect effect of organizational commitment on organizational performance through organizational learning. Organizational commitment significantly predicts organizational learning, and the interaction between organizational commitment and role ambiguity is significant, indicating moderation on the IV→mediator path. Johnson–Neyman results show the commitment→learning relationship is stronger at low and average role ambiguity and weakens at higher levels, with a threshold after which additional increases no longer change significance. The indirect effect (commitment → learning → performance) is significant at low, average, and high role ambiguity, but it decreases as role ambiguity increases. The index of moderated mediation (−0.0564) has a confidence interval that excludes zero, confirming that the indirect effect is truly moderated by role ambiguity.
What exactly is being moderated in this model—an indirect effect or a direct path?
How do Johnson–Neyman intervals determine where the moderation matters?
Why is it possible for the indirect effect to be significant at all moderator levels while still being moderated?
What does the index of moderated mediation test, and how is it interpreted?
What roles do the direct effects play in the interpretation?
Review Questions
- How would you describe the difference between moderation of the IV→mediator path and moderation of the indirect effect in this analysis?
- What does it mean if the indirect effect is significant at low, average, and high moderator values, but the index of moderated mediation is still significant?
- Using the reported pattern of slopes, what practical conclusion follows about organizational commitment when role ambiguity is high versus low?
Key Points
- 1
Role ambiguity moderates the organizational commitment → organizational learning relationship, not the organizational learning → organizational performance link directly.
- 2
The organizational commitment × role ambiguity interaction is significant, indicating role ambiguity changes the strength of the path leading into the mediator.
- 3
Johnson–Neyman intervals show the commitment→learning effect is stronger at low and average role ambiguity and weakens at higher levels, with a threshold beyond which additional increases stop changing significance.
- 4
Organizational learning significantly predicts organizational performance, and organizational commitment also has a significant direct effect on performance after accounting for learning.
- 5
The indirect effect of organizational commitment on organizational performance through organizational learning is significant at low, average, and high role ambiguity, but its magnitude declines as role ambiguity increases.
- 6
The index of moderated mediation (−0.0564) has a confidence interval excluding zero, confirming that the indirect effect is moderated by role ambiguity.
- 7
The supported hypothesis states that the indirect effect of organizational commitment on organizational performance via organizational learning depends on role ambiguity levels.