Date of Graduation


Document Type


Degree Type



Eberly College of Arts and Sciences



Committee Chair

Julie Hicks Patrick

Committee Member

Constance Toffle

Committee Member

Amy E. Fiske


Research shows that physical and mental health are closely linked (Ohrnberger, Fichera, & Sutton, 2017). Further, social role theory states that holding and enacting valued roles, such as grandparenting, can buffer the negative effects of health on depression (Reitzes & Mutran, 2004). Using data from 247 grandparents (Mean age = 66.5; range 42 to 90 years; 46.2% grandfathers), we examined whether grandparent role centrality and engagement with grandchildren altered the effects of physical health on depression. We then explored model differences between 164 custodial and 83 traditional grandparents. We found that for all grandparents the model was of sufficient fit: X 2 (DF = 12, N = 247) = 39.15, p < .001; R2 = .236; CMIN/DF = 3.26, TLI = .955, and RMSEA = .068. Inspection of the individual regression paths showed that among the full sample, significant main effects for Role Centrality (𝛽 = .699***) and Activities (𝛽 = -.768***) on Depression were detected. However, significant main effects for Subjective Health (𝛽 = -.034) did not emerge. All interactions were significant in predicting depression, the three-way interaction term Subjective Health, Role Centrality, and Activities additionally accounted for depression variance (𝛽 = -1.062***). We then examined whether the paths were moderated by custodial status. The model accounted for 25.7% of depression among traditional grandparents, however, no paths emerged as significant. The model accounted for 36.8% of depression among custodial grandparents, and all paths were significant. Among custodial grandparents, there were significant effects on Depression from Subjective Health (𝛽 = .439***), with main effects for Role Centrality (𝛽 = - .635**) and Activities (𝛽 = -.711**). Moreover, each interaction term also emerged as significant with the 3-way interaction of Health, Role Centrality, and Activities accounting for additional variance in depression (𝛽 = -.445*). Custodial grandparents in poorer health who valued the grandparent role and those in poorer health who engaged with their grandchildren experienced fewer depressive symptoms than grandparents who do not engage with grandchildren. Results are discussed in terms of the need to examine the differences of family/social contexts in grandparent populations.