FAMILIES AS NATURAL EXPERIMENTS - A PROCEDURE FOR ESTIMATING THE POTENTIALLY BIASING INFLUENCE OF FAMILIES ON RELATIONSHIPS BETWEEN VARIABLES

In this article, the author argues that data on siblings provide a way to account for the impact of unmeasured, omitted variables on relatio nships of interest. This is possible because families form a sort of n atural experiment. Family members are likely to have many shared exper iences, as well as a common genetic heritage, but relationships betwee n variables defined as differences between family members cannot be at tributed to these shared family characteristics. Although fixed- and r andom-effects models are discussed as one means to make use of informa tion on siblings from the same family, the author proposes a latent-va riable structural equation approach to the problem. This model provide s estimates of both within- and between-family relationships, and it a ccounts for the impact of measurement error.