Figure 1.

Modeling indirect effects. These panels present a hypothetical example of a study in which the effect of acculturation status upon health varies as a function of neighborhood ethnic composition, and the effect of neighborhood ethnic composition on health is transmitted via social support. A1, B1, and C1 panel diagrams explicate what mediation and moderation models practically mean, with arrows representing binary conceptualization of variables. Color gradients reflect the continuous nature of each variable. A2, B2, and C2 panels present conceptual diagrams of indirect effects in the tradition of statistics. In A2 and C2, the arrows directly linking predictor and outcome indicate that the mediating variable does not necessarily account for the entirety of the main effect. Altogether, these diagrams reveal the importance of considering the many complex interactions involved in the relationship between acculturation and health. For example, previous authors have asserted that for certain immigrant groups, heritage-culture orientation may be associated with health benefits due to more available social support (A). If this correlation is attributable to the communalistic values in the heritage culture, then the relationship may vary as a function of the neighborhood proportion of coethnics (C). A heritage culture oriented individual who resides in an ethnic enclave may exhibit health benefits due to high availability of social support that offsets the detriments of poverty, compared to a heritage culture oriented individual who resides in a neighborhood dominated by the majority ethnocultural group (B).