mediator or moderator variable? I have been given a set of data and I am confused as to whether one of the variables is a moderator or mediator variable.
The hypothesis is that stress and well-being have an inverse relationship. However, this relationship is lessened or cancelled out in individuals with internal locus of control.
In other words, even if an individual is highly stressed, if they have an internal locus of control, stress will have no effect on well-being.
Is locus of control a mediating or moderating variable?
 A: Moderating variables (aka effect modifier, interacting variable, etc.) alter the effect of another variable. In your example increased stress causes decreased well-being, but this effect is altered in the presence of locus of control.
Of course, nothing prevents a moderating variable from having its own direct effect (aka main effect). An example here would be respiratory exposure to tobacco smoke, and respiratory exposure to asbestos: tobacco smoke exposure increases risk of mesothelioma; however, the micro-scale mechanical damage wrought by asbestos exposure actually increases the toxic effect of tobacco smoke (asbestos exposure moderates the effect of tobacco smoke exposure on mesothelioma risk), and asbestos on its own and in the absence of tobacco smoke exposure also increases risk of mesothelioma. So asbestos exposure increases mesothelioma risk, and moderates the effect of tobacco smoke exposure on same.
Mediating effects are different. Mediating effects are "in the middle" of a causal chain from a distal cause to some effect. For example ingesting cyanide causes death (from hypoxia). However, the effect of cyanide ingestion on hypoxia is mediated by the inactivation of cytochrome oxidase. Put another way, ingestion of cyanide has an effect on cytochrome oxidase (making it inactive), and this inactivation of cytochrome oxydase in turn causes hypoxia.
A: Just to add to what the others said already. 
I think your confusion is well founded. There is no clear-cut answer to your question as far as I can see. Answering your question depends on your ability to clearly delineate the role locus of attention plays in the connection between stress and well-being. Hypothetical constructs are often too fuzzy for this. The confusion in your example arises because "locus of control" can act as a moderator or a mediator depending on the context. 
Moderators serve to influence/modify the relationship between two other variables by either amplifying or attenuating their interaction. But it doesn't serve as the necessary causal mechanism for their interaction. In your example, locus of control would be a moderator if it influences the relationship between stress and well-being, but isn't a necessary link to establish that relationship. Stress will probably still have an influence on well-being and there are numerous other psychological factors that could influence the relationship between well-being and stress quite independently of locus of control.  
Mediating variables, on the other hand, are vital in establishing the relationship because they make the interaction possible; take them away and the interaction stops. 
The line is very blurry in your case, but my gut tells me that locus of control masquerades as a mediator because of the phrasing, but it is actually a moderator.
