Elsevier

Social Science & Medicine

Volume 55, Issue 5, September 2002, Pages 863-880
Social Science & Medicine

Social inequality in educational achievement and psychosocial adjustment throughout childhood: magnitude and mechanisms

https://doi.org/10.1016/S0277-9536(01)00228-3Get rights and content

Abstract

The paper examines the hypothesis that social inequalities in children's developmental resources level off during adolescence against an alternative hypothesis that they continue to exert their influence throughout all of childhood. Using data from the National Child Development Study, the study applies two models. Both are premised on the understanding that the social and physical environments in which children are raised affects their resources in the domains of educational achievement and psychosocial adjustment. A ‘class inequalities’ model seeks to determine the extent of social class inequalities at three key stages in children's development: the transition from infant to junior schooling at age 7, from primary to secondary education at age 11 and from compulsory education to further education or work at age 16. The second model is a contextual-systems model which seeks to expand our understanding of the pathways from family social class to children's educational achievement and psychosocial adjustment through some more proximal determinants of these resources: material deprivation, school composition, parental involvement and aspirations. Social class inequalities in educational achievement were found to be greater than inequalities in psychosocial adjustment. The same developmental pattern was observed for both outcomes: inequalities increased from age 7 to age 11 and then remained at the same level at 16 yr. The contextual-systems models showed that when social inequalities are interpreted more broadly than a narrow class based definition, they continue to widen in adolescence. In particular, family influences, indicated by parental involvement become less important and social contexts beyond the family, reflected in material conditions and school composition, become more important. At age 16, material deprivation was the strongest determinant of psychosocial adjustment while school composition was most strongly related to educational achievement. The contextual-systems model provides a more complete account of social inequalities in children's educational achievement and psychosocial adjustment than simple estimates of social class effects.

Introduction

Social class effects operating early in life have long-term effects on inequalities later in life including social destinations (Lauder & Hughes, 1990), income (Kuh & Wadsworth, 1991; Power & Matthews, 1997), qualifications (Lauder & Hughes, 1990; Power & Matthews, 1997), mortality (Davey Smith, Hart, Blane, Gillis, & Hawthorne, 1997), and physical and mental health (Power & Hertzman, 1999). The two most consistent childhood determinants of social inequalities in adulthood, measured at the individual level, have been educational achievement and psychosocial adjustment (Montgomery & Schoon, 1997; Roberts, 1997; Wadsworth, 1997), both of which are socially patterned (Fogelman, 1983; Wadsworth, 1996; Marmot, Ryff, Bumpass, & Shipley, 1997).

Power and colleagues observe that “[educational attainment] is an excellent marker of the ‘healthfulness’ of accumulated childhood experience” (Power & Hertzman, 1999, pp. 47). Keating and Hertzman (1999) use the term ‘developmental health’ to refer to the full range of outcomes related to human well-being, including educational achievement and psychosocial adjustment, while Dahl (1996) similarly coins the term ‘latent health potential’. The connections are made between inequality, achievement and adjustment in childhood and health in adulthood. Moreover, it has been suggested that education is the key to escaping the negative long-term effects of psychosocial problems (Wadsworth, 1997). But exactly how social class during childhood exerts its influence on educational achievement and psychosocial adjustment remains unclear (Kaplan & Lynch, 1997). It has been noted that the extensive knowledge about the magnitude of social inequalities in health is not matched by a comparable understanding of the pathways through which social class influences physical and emotional health (Brooks-Gunn, Duncan, & Britto, 1999). The mechanisms relating childhood circumstances to adult health inequalities may be unclear precisely because we lack an understanding of the ways by which social disadvantage affects educational and psychosocial resources during childhood.

West (1991) has shown that social class inequalities in health during adolescence are equalised and many may infer from this that all socioeconomic effects are minimised at this time. If social inequalities in educational achievement and psychosocial adjustment are also found to reduce or level off in adolescence, then the most appropriate model for understanding health inequalities would be in terms of latency effects. Within this paradigm, it is not social inequality during adolescence which is important for understanding adult health inequalities, rather it is the effect of social circumstances earlier in childhood which sets children on course for differential adult outcomes. Alternatively, if social inequalities in educational achievement and psychosocial adjustment are found to continue to increase as children grow up, then the preference would be for a pathways or cumulative risk model (Hertzman, 1999). Here it is argued that it is the differential aggregation of these risks throughout all of childhood and adolescence that contributes to health inequalities in adulthood. A first step in our aim to understand social inequalities during childhood must, therefore, be to quantify the magnitude of social inequalities in educational achievement and psychosocial adjustment at different stages of children's development.

Schneewind (1981), outlines seven disadvantaging conditions which contribute to inequalities in children's potential for healthy functioning. These are poverty, nutrition and health care, physical and home environments, the socio-cultural system, structural changes in the family, parenting behaviour and individual differences of child and family members. So both material and psychosocial conditions are represented in equal measure. In the British context, material conditions indicated by such factors as overcrowding, housing tenure, receipt of benefits, lack of amenities for basic requirements such as hygiene and cooking as well as a lack of amenities for work and play have all been implicated in the perpetuation of social inequalities in children's development (Macintyre, Watt, West, & Ecob, 1991; Coggon, Barker, Inskip, & Wield, 1993; Stitt, Griffiths, & Grant, 1994; Power & Matthews, 1997; Baker, Taylor, & Henderson, 1998). There is also widespread acceptance of the importance of the family environment, with time spent on joint family activities, reading to the child, interest in education and parental aspirations found to be associated with both psychosocial adjustment and educational achievement in school (Astone & McLanahan, 1991; Holahan, Valentiner & Moos, 1994; Teachman & Paasch, 1998). For children, the school is a dominant social context besides that of the family. In a contentious US study (Jencks et al., 1973), school characteristics such as funding and effectiveness were shown to contribute little to social class differences in educational achievement, but other factors relating to the school environment have been implicated, in particular, the socio-economic mix of the school and the educational expectations within the school (Mays, 1980; Roberts, 1980; Brutsaert, 1998). This body of research provides us with a starting point for fulfilling the second part of our aim: to understand the ways by which social disadvantage affects educational and psychosocial resources during childhood.

We propose a model of how social class affects educational achievement and psychosocial adjustment through its effects on the material and psychosocial conditions discussed above. This work borrows from Bronfenbrenner's work on child development to provide a theoretical framework for the model. Bronfenbrenner's (1979) contextual-systems model of psychological development suggests that understanding children's development cannot take place without consideration of proximal and distal social processes which affect how individuals interact with their environment. The developing child is portrayed as being in the centre of an interconnected set of contexts with some directly relating with the child and others affecting the child indirectly, these being mediated by contexts in direct contact with the child (Tudge, Gray, & Hogan, 1997). Directly acting on or being affected by the child are contexts such as the immediate social and material setting in which the child is situated. The family, school and home are examples of these proximal systems. Parental aspirations and involvement are often taken as markers of cultural influences operating at the family level, while the social mix of schools may reflect different cultural norms amongst the peer group (Mays, 1980; Roberts, 1980). The social mix and the educational diversity within schools may also act as indicators of social capital emanating from the educational system (Brooks-Gunn, 1995). Distal systems are those which indirectly impinge on the child. As well as occupational characteristics and material living standards, social class encapsulates cultural norms and customs (Bourdieu, 1984). Social class norms may be thought to influence young children via their parent's practices and attitudes. Social class is a thus a distal system that relates to children's development indirectly, mediated by the material resources available to the family and the emotional resources of parents which may affect the quality of the relationship with their children. Due to its relationship to income and residential location, social class also affects the educational opportunities open to children. The model also takes account of time, recognising that as children develop, they move through a changing set of contextual effects at all levels.

Thus, this paper has two aims. First to examine the magnitude and patterning of social inequalities in educational achievement and psychosocial adjustment at key stages in children's development. In particular, we ask whether social variation in children's resources narrows over time from mid-childhood to late adolescence in common with the health related findings. A ‘class inequalities’ model is outlined below which allows for an estimation of this social class effect. Second, to explore the mechanisms underlying the effects of social inequalities on children's educational achievement and psychosocial adjustment through the use of a contextual-systems model. Data from the 1958 National Child Development Study (NCDS) enable us to examine how far material circumstances, school characteristics, and parental involvement with and aspirations for their child mediate the observed relationship between family social class and children's educational and psychosocial resources. Information gathered when the NCDS cohort were 7 yr (the age for transition from infant to junior school), 11 yr (transition from primary to secondary education) and 16 yr (transition from compulsory education to further education or work) allows for a comparison of possible mechanisms underlying social inequalities at three key stages in children's development.

The class inequalities model is shown in Fig. 1. It is a schematic diagram of a simple hypothesis that family social class influences a child's developmental resources in the domains of educational achievement and psychosocial adjustment. By quantifying the relationships between the 3 constructs, we are able to answer the questions: Does social class influence both educational achievement and psychosocial adjustment? Do social class effects change as children grow older? Does the relationship between educational achievement and psychosocial adjustment change over time?

The contextual-systems model shown in Fig. 2 is a diagrammatic representation of the hypothesised pathways whereby social class affects children's educational achievement and psychosocial adjustment. This model aims to move beyond the simple quantification of the relationship between social class and children's developmental resources to a fuller understanding of how the relationship comes about. The focus of the model is the child; specifically, his or her resources for successfully interacting with the current social environment and for moving on to a healthy and fulfilled adult life. Two forms of developmental resources are considered in the model, namely educational achievement and psychosocial adjustment. At the first level are the proximal contextual systems which are hypothesised to be directly related to the child. The proximal effects are determined in terms of the social composition of the school, material deprivation suffered by the child at home, parental involvement with and parental aspirations for the child. It is predicted that attending a school with a ‘better’ social composition and having involved parents are associated with higher levels of developmental resources in the child. Material deprivation is hypothesised to be associated with lower developmental resources. The direction of the influences between child and context may also flow the other way. Children's educational achievement is hypothesised to affect their parent's aspirations for their future. In turn, higher parental aspirations are predicted to be related to better psychosocial adjustment.

Relationships between the proximal systems of family and school are also included in the model. It is suggested that material deprivation will affect parental involvement and aspirations in the child because those living in financial and material hardship may have fewer resources of their own for interacting with and investing in their children. Parental involvement is hypothesised to be related to their aspirations: parents with higher aspirations are more involved with their children's education and social development.

At the second level of the model are the distal systems. Due to our focus on inequalities in children's development, we are modelling the effects of a single distal contextual system, that of social class. In the model, social class operates mainly through the more proximal contexts. Thus, there are pathways in the diagram from family social class to school social composition, material deprivation, parental involvement and parental aspirations. In addition, a direct path is shown from family social class to the child's educational achievement and psychosocial adjustment. These paths represents the relationship between family social class and the child's resources which are mediated by unknown factors not made explicit in the model. Finally, we are considering temporal contextual effects, which are not shown on the diagram but are taken into account in the analyses by the application of the models to the children when they were aged 7-, 11- and 16-yr old.

Section snippets

The National Child Development Study

The NCDS originated in the Perinatal Mortality Survey (Butler & Bonham, 1963), and has followed 98% of all births in England, Scotland and Wales born during the week 3–9 March 1958. The cohort members, numbering some 17,414 individuals, have been followed-up at ages 7, 11, 16, 23 and 33 yr, with the sample being augmented by immigrants to the UK who were born in the study week. This paper is based on data collected when the cohort members were 7-, 11- and 16-yr old. The total achieved sample

Results

The distribution of the contextual variables is shown in Table 2. On average, material conditions improved over time. There was less overcrowding, with more families owning their homes, living in a house with sole use of a bathroom, hot water and indoor toilet. More parents in the NCDS appeared to have been interested in their 11-yr-old child's education than when the child was 7 or 16, despite age 16 also being an educational transition stage. Parents’ aspirations were also very different at

Discussion

We have taken educational achievement and psychosocial adjustment as markers of ‘health potential’ or ‘developmental health’ to see if the class patterning of these factors during childhood mirrors those seen for physical health. Are educational and psychosocial outcomes determined early in childhood or do they continue to be influenced by social class throughout the life course? The answer depends on the level of analysis: a simple class inequalities model or a contextual-systems model.

The

Limitations

This work involves a comparison of social inequalities in children's educational and psychosocial resources at three stages in their development. The use of a latent variable modelling approach attempts to deal with the problem of different forms of measurement of the same variables of interest collected on three separate occasions. Relationships between latent variables have been compared, rather than the relationship between observed variables. However, it is possible that estimates of these

Acknowledgements

This work was funded by the UK Economic and Social Research Council Grant no. R00429334394. Data from the National Child Development Study were supplied by the ESRC Data Archive. Those who carried out the original collection and analysis of the data bear no responsibility for its further analysis and interpretation.

References (70)

  • N.M Astone et al.

    Family structure, parental practices and high school completion

    American Sociological Review

    (1991)
  • D Baker et al.

    Inequality in infant morbidityCauses and consequencies in England in the 1990s

    Journal of Epidemiology and Community Health

    (1998)
  • P.M Bentler

    Comparative fit indices in structural models

    Psychological Bulletin

    (1990)
  • R Boudin

    The unintended consequences of social action

    (1982)
  • P Bourdieu

    Distinction

    (1984)
  • U Bronfenbrenner

    The ecology of human developmentExperiments by nature and design

    (1979)
  • Brooks-Gunn, J. (1995). Children in families in communities: Risk and intervention in the Bronfenbrenner tradition. In:...
  • J Brooks-Gunn et al.

    Are socioeconomic gradients for children similar to those for adults?

  • H Brutsaert

    Home and school influences on academic performanceState and catholic elementary schools in belgium compared

    Educational Review

    (1998)
  • N.R Butler et al.

    Perinatal mortality

    (1963)
  • D Coggon et al.

    Housing in early life and later mortality

    Journal of Epidemiology and Community Health

    (1993)
  • J Cohen et al.

    Applied multiple regression/correlation analysis for the behavioral sciences

    (1983)
  • E Dahl

    Social mobility and healthCause or effect?

    British Medical Journal

    (1996)
  • G Davey Smith et al.

    Lifetime socioeconomic position and mortalityProspective observational study

    British Medical Journal

    (1997)
  • R Davie et al.

    From birth to sevenThe second report of the national child development study (1958 Cohort)

    (1972)
  • L Dillner

    Poverty spreads among children in Britain

    British Medical Journal

    (1996)
  • C.V Dolan et al.

    Testing specific hypotheses concerning latent group differences in multi-group covariance structure analysis with structured means

    Multivariate Behavioral Research

    (1994)
  • O.D Duncan

    A socioeconomic index for all occupations

  • J Elander et al.

    An update on the status of the Rutter Parents’ and Teachers’ Scales

    Child Psychology and Psychiatry Review

    (1996)
  • J Elander et al.

    Use and development of the Rutter Parents’ and Teachers’ Scales

    International Journal of Methods in Psychiatric Research

    (1996)
  • Fergusson, D.M., Lynskey, M.T., J, H. L. (1996). Factors associated with continuity and changes in disruptive behavior...
  • Fogelman, K. (Ed.). (1976). Britain's 16-year-olds. London: National Children's...
  • Fogelman, K. (Ed.). (1983). Growing up in Great Britain: Papers from the National Child Development Study. London:...
  • L.A Hayduk

    Structural equation modeling with lisrelEssentials and advances

    (1987)
  • C Hertzman

    Population health and human development

  • Cited by (116)

    View all citing articles on Scopus
    View full text