Dear PLS community,
I am struggling with the following issue: In my PLS model, I want to analyze the interaction effect between strategy and context on performance. However, I am unable to specify the classical predictor/moderator relationship, as neither strategy nor context are assumed to have a direct relationship with performance. Rather, it is merely the fit between strategy and context which is assumed to lead to higher performance.
Can anybody give me advise on how to implement this setting in PLS? Is it still feasible (from a methodological point of view) to analyze the classical moderation effect, without having a theoretical foundation for the direct effects between strategy and context on performance? Venkatraman (1989) introduces the concept of "fit as matching" as an alternative to "fit as moderation". In this case, "fit is specified without reference to a criterion variable, although, subsequently, its effect on a set of criterion variables could be examined" (p. 430). I'm wondering whether this is more appropriate in my setting and how it would be implemented in PLS.
Any advise? Your help would be greatly appreciated!
Thanks a lot,
Dirk