The analyst who wants to connect the independent and dependent variables through some sequence of intervening variables must begin by speculating about possible meanings of the relations. The new variables used in the process of interpretation differ from those introduced in attempts at explanation in that they are seen as intervening in the causal sequence between church attendance and delinquency rather than as being antecedent to church attendance. Although the statistical configurations of successful attempts at explanation and interpretation are the same, the analyst's purposes in the two cases are different. In the first he decides whether or not to get rid of the original relation; in the second he accepts the original relation as causal and attempts to show how the two variables are connected. Studies dealing with the family status of children incarcerated in institutions show that most of the inmates, especially the girls, come from broken homes.