Testing statistical moderation in research on home-school partnerships: Establishing the boundary conditions

Parent academic involvement, as one of the typical forms of home-school partnerships, is generally defined as parents' work with schools and with their children to benefit their children's educational outcomes (Hill et al., 2004). Forms of parent academic involvement include help with homework, conversations with children about school, communication with teachers, attendance at school events, volunteer activities at school, and discussions with children about the value of education and their educational plans and aspirations (Epstein and Sanders, 2002; Fan and Chen, 2001; Wong and Hughes, 2006). Although most forms of parent academic involvement are associated with positive outcomes for children's academic and behavioral functioning, a number of family and child factors moderate the effects of parent academic involvement on children's outcomes (for review, see Hill and Tyson, 2009). Statistical moderation refers to the finding of variations in the strength or direction of the effects of some (focal) variables on an outcome variable across subgroups of students who differ on some demographic, behavioral, or other characteristic. A finding that parents' help with homework is more strongly predictive of their children's math achievement for parents with higher versus lower levels of educational attainment would be an example of statistical moderation of the effect of parent academic involvement (or home-school relationship) on students' achievement. By examining "for whom" a particular type of parent academic involvement bestows benefits, statistical moderation establishes the "boundary" conditions for the effect, thus providing a more nuanced understanding of the effects of home-school relationships. In this chapter, we provide a tutorial on how to test for moderated effects of home-school relationships using different statistical approaches. First, we provide a brief review of current research on moderated effects of home-school relationships to illustrate the theoretical and practical value of statistical testing of moderated effects. Next, with a detailed example, we demonstrate statistical methods to test the moderating effect of a child characteristic (i.e., ADHD symptoms) on home-school relationship. We also discuss some related issues on testing moderation effects such as centering variables and handling missing data to provide guidance to researchers.