Grandparents’ Affectionate Communication toward Grandchildren and Grandchildren’s Mental Health Difficulties: The Moderating Role of Future Time Perspective

ABSTRACT

Given that grandparent-grandchild relationships often last for several decades of grandchildren’s lives, it becomes important to examine how grandparents’ affectionate communication might be associated with their adult grandchildren’s mental well-being. Using a sample of college-aged grandchildren (n = 401), this study examined whether or not grandparents’ affectionate communication is indirectly associated with grandchildren’s loneliness, depressive symptoms, and stress, via shared family identity. It also considered whether or not grandchildren’s assessments of their grandparents’ futures as expansive versus restricted moderated these associations. Affectionate communication was indirectly associated with less loneliness via heightened shared family identity, but only for grandchildren who judged grandparents’ futures as expansive. These findings suggest the importance of healthy grandparents conveying to grandchildren that their lives as older adults are still full of possibilities and open to new growth.

Acknowledgments

The author would like to thank the editor and two anonymous reviewers for their constructive feedback on earlier versions of this paper.

Notes

1. The original sample size was N = 423. However, a total of 22 grandchildren reported on a relationship with a non-biological grandparent (e.g., a relationship with their biological grandparent’s second spouse through remarriage). Given AET’s status as a bioevolutionary theory and past research suggesting notable differences between biological and non-biological GP-GC relationships (e.g., Mansson & Booth-Butterfield, 2011), this study focused on the 401 grandchildren reporting on biological grandparents. All content in the Method and Results sections pertains to the subset of 401 biological grandchildren.

2. Traditional fit indices (i.e., the chi-square test of model fit, comparative fit index [CFI], Tucker-Lewis index [TLI], root mean square error of approximation [RMSEA], and standardized root mean square residual [SRMR]) are not available for the three structural models because the latent interaction factor was specified as the product of two latent factors (i.e., the SFI factor and the FTP factor). When specifying a latent interaction factor as the product of two latent factors, researchers cannot obtain the set of fit indices traditionally associated with structural equation modeling. However, researchers can obtain a different set of fit indices (i.e., the loglikelihood, Akaike information criterion [AIC], Bayesian information criterion [BIC], and sample-size adjusted Bayesian information criterion [aBIC]). Please refer to Muthén (2012) and Muthén and Muthén (2017) for example syntax and a more detailed discussion of why Mplus yields this alternative set of fit indices for moderated mediation models with latent interaction factors.

3. A complete list of covariate path coefficients and standard errors is available upon request.

4. Complete tables of the depressive symptom model’s factor loadings and path coefficients are available upon request.

5. Complete tables of the stress model’s factor loadings and path coefficients are available upon request. The indirect associations from affectionate communication to stress, via SFI, at the five levels of FTP are also available upon request.