Conclusions
7.1.4 Distal factors
Income and family size are important mediators of the effects of parental education.
In some sense teenage motherhood also mediates education effects if one defines
education in terms of early school attainment and engagement. Although family
structure and maternal employment are not strong mediators of education effects,
education does importantly moderate their effects. For example, education may
provide protective capability for families dealing with the income effects following
family break-up (see Blackwell & Bynner, 2002) or support them in assessing
maternal employment rights, good work-life balance or quality pre-school in order to
moderate any effects of employment or child development. More research on these
moderating benefits of education would be particularly valuable.
7.2. Future research
From the theoretical perspectives laid out in this report the proximal family processes
of warmth, discipline and educational behaviour in the home are all important
separate factors in child development. They are strongly influenced by family
characteristics which also play a substantial role in the transmission of educational
advantage. Parents’ cognitions, well-being and resources all have direct effects on
proximal processes and so are major influences on children’s attainments. These
characteristics of the family are in turn influenced by distal factors, particularly
parental education and income.
Education is also strongly related to each of these factors, all of which interact in
important ways. Education not only enhances the developmentally supportive level of
each important, separate factor it also eases the relations between factors and provides
resilience for families when other important elements are absent or where compound
risk factors are excessive.
This theoretical perspective is supported by the evidence but mainly in terms of
particular links in these chains of association rather than in the whole framework.
Much of this evidence is fairly ambiguous and so could be interpreted in a number of
different ways. The interactions between the elements of the framework are complex
and multi-layered so sophisticated modelling techniques are required to test the theory
empirically. Yet these techniques are better suited to establishing pathways of
association than to proving unique one-way causal hypotheses.
There is a clear need, therefore, for research that uses large sample longitudinal data
and simple hypotheses that separates out particular aspects of the overall model and
uses the lag structures of the data to identify elements of the overall causal picture
with clarity.
For example, it would be valuable to establish how changes in parents’ aspirations for
children respond to changes in children’s actual attainments. This would help in
clarifying how much of the association of child attainment and parent aspiration is
due to aspirations being matched by parents to their children’s apparent possibilities
83