Virtually all of such research has involved cross-sectional studies that have used convenience samples without standardized measures, triangulation of sources, or suitable controls. This means that the research can yield only cautious hypotheses regarding the effects of imprisonment of inmates' partners and children. For prisoners' partners, the imprisonment can mean the practical difficulties of loss of income, social isolation, difficulties in maintaining the relationship, and extra burdens of child care, along with the emotional adjustment to the loss of the psychological benefits of the relationship. Experiences of partners will differ according to the quality of the relationship, offense types, social support systems, and other sociodemographic factors. The assumption that parental imprisonment causes psychosocial difficulties for children is pervasive in the literature; however, studies have lacked the methodological sophistication required to distinguish the effects of parental imprisonment from the effects of other influences on children. The author argues that four types of factors should be included in a model of parental imprisonment and child adjustment: selection effects that precede the imprisonment, and direct, mediating, and moderating effects following the imprisonment. To date, there is insufficient evidence on the moderators of the effects on children of imprisonment of a parent. This is partly because of the difficulties of conducting prospective studies of prisoners' families. To test the effect of parent-child contact on children during parental imprisonment, future studies should measure child well-being by using different sources, controlling for parent-child relations prior to imprisonment, and devising sensitive measures of different types of contact during imprisonment. 6 notes and 67 references