Moderator and mediator effects in hospitality research

Investigation of moderator or mediator effect allows researchers to explain a more precise description of the relationship between independent and outcome variables. Although moderator and mediator are often incorporated in the hospitality research, unclear differences between the two variables and other third variables, and technical complexity in statistical testing seem to confuse researchers in adopting appropriate conceptualizations and analyses. The goals of this article are to (a) summarize the conceptual differences between moderators and mediators, (b) provide nontechnical descriptions of how to examine each type of effect in statistical analysis via regression and SEM approaches, (c) differentiate mediated-moderation and moderated-mediation effects, and (d) discuss other third variable effects of confusion in the moderator and mediator studies. Efforts are made to translate existing technical literature on the moderator and mediator testing. (C) 2011 Elsevier Ltd. All rights reserved.