Extensive research documents that family experience, particularly family structure, is a risk (or protective) factor for a range of adverse physical and mental health outcomes in children and parents. However, the extent to which - or whether - family life causes unwanted outcomes is open to question. Neither parents nor children experience family life at random;rather, individuals are """"""""selected"""""""" into family experiences based on known (and unknown) characteristics. Selection into the major family experiences that are our focus - teen pregnancy, divorce, stable marriage, and religious participation - is known to occur at the demographic level. Selection also is likely to occur based on unmeasured background experiences and individual characteristics, including genetically mediated selection effects such personality (e.g., antisocial behavior) or physical characteristics (e.g., age at first menarche). Behavior genetic researchers call this the gene- environment correlation, and suggest that the confound renders existing social science research on family life uninterpretable. We take this concern seriously, and have conducted genetically-informed twin studies of divorce and children's adjustment, step-father presence and early menarche, harsh parenting and child well- being, and parental conflict and child adjustment. Twin studies allow for complete (in co-twin control designs) or partial (in children of twin designs) control for genetic selection. Twin studies also control for the shared environment, because twins share family experiences in addition to genetic background. Thus, twin studies afford control for genetic background, measured experience, and unmeasured shared experience. We propose to use the Australian Twin Registry, AddHealth, and the Virginia 30,000 to distinguish genetic and shared environmental selection effects from the likely causal effects of teen pregnancy, divorce, stable marriage, and religious participation on the emotional and physical well being of parents and children. Public Health Relevance: Family difficulties are linked with adverse physical and mental health outcomes in children and parents. Because family experience is not random, we do not know whether family troubles cause these problems. Twin studies uniquely control for genetic and shared environmental selection effects, thus refining distinctions between outcomes merely correlated with teen pregnancy, divorce, stable marriage, and religious participation from outcomes caused (or prevented) by these major family experiences.
Showing the most recent 10 out of 32 publications
