Research into the outcomes for children who have or have not experienced family disruption and the accompanying debate is complex, contentious and controversial. On the one hand there are those wary of the quality of the data available and what can safely be said about children’s outcomes as a result. On the other hand, there is the sense that outcomes must evidently be poorer for children who experience family disruption and lone parenthood (than for those who do not) because of their distress and frequently poorer socio­economic circumstances.

In between, a number of researchers feel confident about the poorer outcomes where they have been found for children and young people who have experienced family disruption, because of similar such findings from a number of studies (and cross-nationally). The differences between the outcomes, they acknowledge, may not be large and there is less certainty about accounting for them; the complex web of causation has yet to be untangled. Doing so is thought to be neither simple nor straightforward. Why, for example, are there children experiencing family disruption who fare ‘well’ and others in two-parent intact families who fare ‘badly’? Neither experience of family life will be ‘all good’ or ‘all bad’; such a model is too simplistic a view of the realities and dynamics of family life. Conflict and poor socio-economic circumstances, after all, may and do occur in two-parent families.

Where children who experience family disruption are likely to differ from those who do not is in facing more and more difficult circumstances, events and changes in their lives. But such an analysis also allows that the reverse will sometimes be the case, and that the breakup of a family may make relations more harmonious.

The fact that some children fare worse than others is likely to reflect the interrelationship between the particular pathway of these circumstances, family events, the stage of development and maturity of the children as they occur, as well as their relationship with their parents. The relationship between their parents will have its effect too and there is no given path that it will travel as families part.

It is possible, therefore, that the apparently poorer outcomes for children from step-families, for example, may be due in part at least to the greater number of such potentially difficult transitions that they have to make (Capaldi and Patterson 1991; Capaldi 1992). For many this will have been their second family change— for example, from living with their own natural parents, to living with one parent alone, and then to living with their own parent and a step-parent.

Such an interpretation is thought to be in keeping with stress theories, which assume that the effects of family disruption arise from the transitions and changes in relationships and not the family setting per se, generally lone parenthood, that follows (McLanahan and Bumpass 1988). For while it is possible that every transition might be done carefully, none the less each might still be difficult and may require psychological adjustments to new family relationships and social circumstances. Clearly the relationships between individuals within step-families are complex and the psychological adjustment required greater (Robinson and Smith 1993). Moreover, just as new relationships may be gained (as is often said about boys and stepfathers), others may be ‘lost’ (as girls are said to feel sometimes about their relationship with their mother).

Researchers seem increasingly of the view that there is no single explanation, whether psychological, sociological or economic, for the generally poorer outcomes that children who have experienced family disruption may display compared with those from intact families. There could be no single explanation that would fit the different experiences of family change. But that is not to say that some sorts of explanation might be more likely for some types of change or disruption, or at different times or of different intensities. Some researchers suggest the need for a multi-dimensional model based on the concepts of ‘resources’, which provide opportunities to develop social and cognitive competence to help deal with ‘stressors’, that is stressful life events such as marital dissolution and the psychological, social and economic difficulties that it entails (Amato 1993). Others suggest, perhaps in similar vein, the need to consider not so much whether a particular factor is present or absent, but the total ‘configuration’ of these resources and stressors. Not only, therefore, might one resource compensate for the lack of another, but pre-divorce stressors could exceed those post-divorce (Demo 1993).

The discovery of pre-divorce effects on children’s outcomes might add weight to the theory that the ‘cause’ of children’s outcomes is the relationship between the dynamics of the marriage itself (in turn affected by partner pre-selection) and the (changing) social and economic circumstances and events that families experience. Divorce is just one social event (albeit an important one) within this process (Richards 1993).

It is perhaps important to reflect, however, that there will be children for whom the separation and divorce of their parents will not be a damaging long-term experience. Nor do we know whether children for whom it is a damaging experience would have been better off had their parents stayed together. We do not know what the outcomes for these children would have been had they done so. But nor can we be certain that where the relationship between their parents is marked by discord, children will always be better off if they part. To do so is to assume that such conflict would have continued. We do not know what will happen to either group of children in the future. We can never assume that families will stay as they are either in the nature of their relationships or in their structure.