Covid 19 impact on children′s social and emotional development

British Journal of Social Work (2000)30, 703–720

Ecological Influences on Parenting and Child Development

Gordon Jack

Gordon Jack began his career in social work in 1974, working in three local authority social services departments in the north of England. He has always worked with children and families— eleven years as a practitioner and five years as a manager—with special emphases on adoption and fostering, child protection and direct work with children. Since 1991 he has been lecturing in social work at Exeter University, where he is now Head of Social Work Studies. He is involved in qualifying and post-qualifying training, as well as writing and research into child-care issues.

Correspondence to Gordon Jack, University of Exeter, Department of Social Work Studies, Richards Building, St. Lukes Campus, Heavitree Road, Exeter EX1 2LU, UK.

Summary

The development of children and young people and their subsequent life chances in adulthood are the product of a complex set of interacting factors, at individual, family and community levels. An ecological framework is used to examine the mutual influ- ences between the different systems that are important during childhood. The links between poverty and health and the influence of social exclusion and structural inequalities on family and community functioning are considered in some depth, utiliz- ing findings from a wide range of research disciplines, in different parts of the world. The identification, development and protection of sources of informal social support, available to families living in impoverished circumstances, are identified as key tasks for social workers and other welfare professionals. Together with features of resiliency, found amongst children and families coping successfully in high-risk environments, this knowledge is used to point the way towards features of effective, ecological assess- ment and intervention strategies.

Within the context of a series of high profile child abuse inquiries and a hostile political and media environment, the main focus of social work with children and families in the UK over the last two decades has been the development of systems to identify and protect children from ‘significant harm’. Unfortunately, this restricted focus has deprived other areas of social work with children and families of both resources and professional status, and created barriers of mistrust and suspicion between disadvantaged families and professional workers (Jack, 1997).

More recently, in the light of legislative changes and research findings, an effort to re-focus services, locating child protection in the wider context of family support

 2000 British Association of Social Workers

704 Gordon Jack

provision, has been made. However, insufficient attention continues to be given, within social work policies and practices, to the structural and environmental factors that are at the root of most of the problems experienced by families (Blaxter, 1990; Kumar, 1993; Nazroo, 1997; Wilkinson, 1996). A proper understanding of the pro- cesses of mutual influence between individuals and their social and economic envir- onments requires an ecological perspective. This will be considered in some depth and the implications of the ecological model for assessing families and designing appropriate interventions will be highlighted. In particular, the major structural and environmental influences on children’s health and development will be examined, together with the roles of different sources of support and resilience.

The social ecology of families

What aspects of children’s environments have the most significant effects on their development? What factors influence parents in the way that they bring up their children? Answers to these questions began to emerge within what became known as the ecological perspective in the 1970s.

Researchers investigated the way in which social relationships outside the family can influence parental behaviour and child development (Cochran and Brassard, 1979; Belsky, 1984), as well as physical and psychological illnesses (e.g. Cassell, 1974; Cobb, 1976; Tolsdorf, 1976) and reactions to stress (Coyne and Downey, 1991). Meanwhile, other studies looked at the impact of inequalities and environ- mental factors in the community on outcomes for children and families (Garbarino and Kostelny, 1992; Rutter and Madge, 1976; Rutter and Quinton, 1977; Rutter and Giller, 1983). Uri Bronfenbrenner developed a systems model for analysing the social ecology of families, in which parenting capacity to meet children’s develop- mental needs is related to such external factors as the parents’ work patterns, the adequacy of available childcare, the level of social support provided by friends and neighbours, the quality of local welfare services and the safety of the area in which they live (Bronfenbrenner, 1979).

The ecological model first outlined by Bronfenbrenner in relation to families and children, focuses on the balance of stresses and supports in the family environment and the interactions between them. A detailed consideration of these factors will form the main body of this article. The model also places an emphasis on people’s subjective perceptions of their own circumstances as a significant determinant of outcomes. The implications of this phenomenological approach will also be consid- ered for social work assessments and interventions.

Stresses on families—the effects of inequalities

The links between inequality and personal health and development are well-known and clearly demonstrated by a large body of empirical evidence from around the world. People living in poverty are more likely to suffer ill health and premature

Ecological Influences on Parenting and Child Development 705

death than their more affluent counterparts. Equally, poor health and disability are recognized as significant factors contributing towards the risk of experiencing pov- erty (Blackburn, 1991; Bywaters and McLeod, 1996; Kumar, 1993). Here we see a perfect illustration of the ecological model, with the mutual interactions of indi- viduals and their environments demonstrated in study after study.

A good example of these processes is provided by the British Health and Lifestyle Survey (Blaxter, 1990), which showed how different aspects of physical and psycho- social health all tend to demonstrate a social class gradient, with poorer outcomes the lower down the social class ladder you go. Similar health associations were also found for other indicators of social disadvantage, such as unemployment and lone parenthood. The specific influence of unemployment on health was also demon- strated in a study of eight hundred adults living in South London, where it increased the risk of suffering psychiatric problems by a factor of three, for both men and women (Bebbingtonet al., 1981).

Numerous studies have investigated the links between inequality and mortality rates. For example, a study in the Northern Health Region of England found death rates for adults were four times higher in the poorest Wards than in the most affluent Wards (Phillimoreet al., 1994). The same associations are also found in relation to children, with road traffic accidents and fatal accidents in the home both showing clear social class gradients (Spencer, 1996). There are also close and consistent relationships between deprivation and educational achievement at all ages (Kumar, 1993) and between poverty and rates of recorded child maltreatment and crime (Garbarino and Sherman, 1980; Rutter and Giller, 1983). Children growing up in poverty are also at much greater risk of developing conduct and emotional disorders and chronic illnesses and of not reaching their full cognitive potential (Rosset al., 1994; Woodroffeet al., 1993; Steinhauer, 1998; Duncanet al., 1994).

Health and developmental inequalities of this nature have risen sharply in the UK over the last twenty years, in line with increasing inequalities in income distribution and a rising proportion of the population living below the poverty line (most com- monly defined as half of average earnings, after housing costs). Among developed countries only New Zealand has experienced a larger growth in inequality of income distribution during this period (Joseph Rowntree Foundation, 1995). New Zealand has also witnessed sharp rises in social and psychological malaise during this period, with youth suicides having doubled and the crime rate having soared to the point where both of these measures are now the highest in the industrialized world (Steinhauer, 1998). Although the pace of change slowed in the UK during the latter part of the 1990s, it remained ‘top’ of the child poverty league in the EU, with more than one third of its children living below the poverty line, compared to an EU average of 20 per cent and only 5 per cent in Denmark (Department of Social Security, 1998;Guardian, 1997).

Ethnic minority and lone-parent families are disproportionately affected by this process of growing impoverishment because they are over-represented among the unemployed, the low-paid and those households dependent on diminishing state benefits (Kumar, 1993; Platt and Noble, 1999). For example, the proportion of lone parents (mainly women) living in poverty in the UK rose from just over a quarter

706 Gordon Jack

in 1979 to nearly three-quarters during the 1990s (Department of Social Security, 1994). Children living in ethnic minority or lone-parent families are therefore at increased risk of health or developmental problems (Smaje, 1995).

The illustrations of the close and consistent associations between inequality and health are particularly significant in the light of emerging evidence that, in the developed world, it is the degree of inequality within a population, rather than its overall level of prosperity, which has the major influence on levels of health (Marmot and Davey Smith, 1989; Davey Smithet al., 1990; Hertzmanet al., 1990; Wilkinson, 1996). The explanation for this conclusion is considered in the next section, which focuses on social support and its ability to both enhance quality of life in general and offer protection from the potentially harmful effects of stressful life events and circumstances.

Protective factors—social support, resilience and social capital Personal social support networks

Personal social networks are the webs of relationships that exist between individuals and a wide range of people, including relatives, friends, neighbours, work colleagues and professionals. The structure of these networks can be analysed along a number of different dimensions, including size, proximity, stability, frequency of contact and density. Network relationships can be sources of both support and stress, so it is important to examine the nature and the content of relationships within networks, to understand whether they are liable to assist or undermine family functioning.

Network relationships that providesocial supporthave consistently been found to be associated with positive influences on families, parents and children (Dunstet al., 1988a). The most important functions of social support, identified in numerous empirical studies, appear to be the provision ofemotional support, instrumental help and information and advice(Cochran and Brassard, 1979; Sarasonet al., 1990; Crockenberg, 1988).

There is evidence to indicate that social support’s main influences on child devel- opment and parent–child interactions are indirect, via more direct effects on parents’ health and well-being and overall family functioning. However, within the ecological model, all aspects of the child’s environment would be expected to have mutual influences upon one another (Dunstet al., 1997; Crnic and Stormshak, 1997).

It should be noted, at this point, that much of the research in this area, although it often refers to ‘families’ and ‘parents’, has actually been conducted with mothers and young children, particularly focusing on disadvantaged groups, such as single parents, teenagers and those living in poverty. There is much less research evidence about fathers and older children.

A consistent finding within the research that has been undertaken is the import- ance of a close, confiding adult relationship for mothers raising young children. This is well illustrated by a study of the social support networks of mothers with young

Ecological Influences on Parenting and Child Development 707

children in South London (Brown and Harris, 1978). There was a clear association between lower social class and increased incidence of depression for these mothers, which could be accounted for by four ‘vulnerability factors’. The most important of these factors was lack of emotional support from the mother’s husband or boyfriend. A subsequent study revealed that this ‘core support’, which could also be provided by a close friend, was an important protective factor against depression. Mothers with either conflictual or unreliable intimate relationships were found to be at the greatest risk (Brownet al., 1986; see also, Kotchet al., 1997; Lacharite´ et al., 1996). Other important sources of social support, which can assist personal functioning, in their most usual order of influence, include close relatives outside the household, friends, neighbours and, finally, lay or professional helpers (Gibbons, 1990; Bronfen- brenner, 1986). The important message here is thatinformal sources of social support tend to be more effective in enhancing personal functioning thanformal sources (Beckman, 1991).

Some of the beneficial effects of social support operate even when it is not actually provided. In other words, theperceptionthat support will be available, if it is needed, is important in some situations. While social support (e.g. help with childcare) actually provided is generally beneficial to all parents, perceived availabil- ity of support, especially from a close confidant, acts as a protective factor against stress, particularly for mothers living in impoverished circumstances (Sheppard, 1994; Hashima and Amato, 1994).

Another illustration of the positive potential of social support is provided by studies that demonstrate the negative effects ofsocial isolation. A number of studies have found social isolation to be a feature of families in which child neglect occurs, with mothers in these families generally having smaller social networks and fewer social contacts than non-neglecting control groups (Coohey, 1996; Thompson, 1995). More widely, social isolation has been shown to be a clear risk factor for poorer health and, conversely,social integrationis strongly associated with good health, especially psychosocial well-being (Blaxter, 1990).

Before moving on to consider some other supportive mechanisms, it is worth reiterating that network relationships can also be sources of stress and conflict. We should not make the mistake of assuming that personal networks are synonymous with support (Gibbons, 1990; Smaleet al., 1994). For example, a study comparing the networks of single and married mothers found that, although both groups received more support than stress from these relationships, the single mothers experi- enced more stress from their network relationships than the married mothers, often in the form of criticisms, and this was associated with poorer mother–child interactions (Brassard, 1982). This potential for network relationships to undermine parenting can also be a problem for parents of children with disabilities (Dunstet al., 1997).

The concept of resilience

The main threat to healthy child development lies in the accumulation of adversities that they face, rather than any isolated individual event. (Garmezy, 1994). Whilst

708 Gordon Jack

the presence of one or two risk factors in a child’s environment may not result in significant harm, a constellation of five or six factors is likely to seriously undermine healthy development (Rutter, 1979). There are, none the less, many children and families who display varying degrees ofresiliencein the face of apparently over- whelming odds. The components of such resilience will be important features in any ecological assessment and may provide important clues to interventions that can help to protect children growing up in high-risk environments.

One of the most consistent findings from this area of research is that children growing up in high-risk environments can be protected by an enduring relationship with a special person outside of their household, like a teacher or a grandparent (Cohen and Wills, 1985; Jenkins and Smith, 1990; Werner and Smith, 1992). It has also been found that a source of recognition and achievement outside the home, perhaps through educational or sporting success or involvement in a church or youth group, can lead to improved life chances in adulthood (Werner and Smith, 1992). For instance, the adult outcomes for children raised in institutional care were signi- ficantly enhanced by positive experiences at school, which were associated with better planning and choice of such things as a career and a marital partner (Rutter et al., 1990).

Within the home, positive sibling relationships and parent-child interactions, the availability of toys and safe play space and appropriate parental expectations are all protective for young children (Bradleyet al., 1994). It may be necessary for parents living in particularly high-risk environments (e.g. impoverished, high-crime, inner- city areas) to place extra restrictions on their children’s freedom to reduce their exposure to the serious risks that exist in the environment (Baldwinet al., 1990). For example, it is known that parental supervision is one of the most important factors influencing children’s offending behaviour in their teenage years (Riley and Shaw, 1985).

The role of social capital

The final component in the range of protective factors to be considered here, and arguably the most important, is the influence of the wider community environment on children’s development. We have already considered the way in which countries with large inequalities in income distribution, like the UK, tend to produce similar inequalities in health, whereas countries with more equal allocations of income (e.g. Sweden) do not show such associations between social class and health (Vagero and Lundberg, 1989; Leonet al., 1992). These differences between countries are not explained by variations in either the levels of health services or the behaviour of individuals within the populations being compared. What emerges as the most likely factor behind these differences is the level of what is referred to as either social cohesion orsocial capital(Wilkinson, 1996; Kawachiet al., 1997).

Social capital consists of a wide range of community-level interactions, both informal and formal, between members of a particular population. It is a way of expressing the summation of all the personal social networks of a particular com-

Ecological Influences on Parenting and Child Development 709

munity, the day-to-day exchanges between friends, relatives and neighbours, work colleagues, church members, political and social groups, community activities and official organizations. It is developed and maintained by repeated and varied exchanges between relatively equal members of a community, which fosters a col- lective sense of trust and mutual respect. However, it can be fatally undermined by inequalities and divisions or exclusions within a society, which act as barriers to the open and reciprocal interactions on which social cohesion thrives (Jack and Jordan, 1999).

Social capital interacts in important ways with both income distribution and health. This is nicely illustrated by a study of the development of regional govern- ment in Italy. Areas with more democratic and egalitarian structures had higher levels of social capitaland better child health, whereas less egalitarian regions were characterized by lower levels of participation in public and social life and poorer child health (Puttnamet al., 1993). It seems likely that the effects of social capital on health will contain both direct and indirect components, helping to improve the quality of life in general (Mitchellet al., 1982) and protecting vulnerable individuals from the negative impact of stressful life events and circumstances (Brownell and Shumaker, 1984; Wills, 1985).

It is also worth drawing attention to the role that different aspects of the physical environment can play in social interactions within particular geographical locations. For example, some locations, such as local shops and community centres can provide particularly rich ‘behaviour settings’ (Barker, 1978) which are open to a wide spec- trum of residents and permit or encourage a range of social behaviours and activities, whereas other settings are more restrictive and exclude certain groups. Thus, whilst schools represent major potential resources within any community, there are enorm- ous variations in the extent to which that potential is opened up (beyond the children taught there) to include parents and other members of the community (Ball, 1998; Donnison, 1998). Finally, the availability of safe play areas for children, protected from traffic, is another important feature of the physical environment (Spencer, 1996). The development of ‘home zones’ (Guardian, 1999a) is one approach to this issue, which has been developed in Continental Europe and is now being tested in the UK.

Having considered the major sources of stress and support that affect families, it is now time to turn to the influences that these factors can have upon one another.

Interactions between stresses and supports

As well as the separate effects of stresses and supports on family functioning already considered, the ecological model indicates that there will also be significant interac- tions between individual, structural and community variables and different compon- ents of social support. It is to some of these influences that we will now turn, before trying to extract some of the most important messages from the ecological model for social work with children and families.

710 Gordon Jack

Individual and structural influences on social support networks

We have already considered the association often found between social isolation and the occurrence of child neglect. A number of researchers have highlighted the way in which parental characteristicsmay influence the social networks available to different families. Norman Polansky and his colleagues found that ‘neglectful’ mothers tend to perceive their neighbourhoods as less friendly and supportive and engage in less neighbourly exchanges than non-neglecting mothers from the same neighbourhoods. They have argued that these differences are due to inherent person- ality traits in ‘neglecting’ mothers, who use ‘social distancing’ as a psychological defence mechanism, thereby contributing significantly to their own isolation (Polanskyet al., 1985a, 1985b). Along somewhat similar lines, Patricia Crittenden developed the theory of ‘internalized working models’ of relationships, primarily formed in childhood, to explain the different behaviour and social network relation- ships of ‘abusing’, ‘neglecting’ and ‘adequate’ mothers who had been matched for socio-economic status. ‘Neglecting’ mothers tended toavoid relationships, dis- playing a sense of hopelessness and despair, whereas ‘abusing’ mothers tended to offendpotential network members, by engaging in non-reciprocal interactions aimed at coercing others to meet their needs. Only the ‘adequate’ group entered intoco- operative and reciprocalrelationships, involving satisfaction for themselves and empathy towards others (Crittenden, 1985). These patterns were common to all of their relationships, including those they formed with their children.

Educationhas also been shown to have a significant impact on social networks, with higher educational achievement associated with larger networks, consisting of more intimate relationships with people spread over a wider geographical area (Fischer, 1982; Werner, 1995). When education is held constant,household income emerges as an important variable, with members of higher earning households including more non-kin in their networks and reporting more satisfaction with the support available from network members (Fischer, 1982). It is worth pointing out here that most of the research upon which these findings are based used cross- sectional, rather than longitudinal designs. This means that causal mechanisms can only be suggested, rather than proved, leaving it open to debate in this instance whether higher levels of education provide the skills and resources necessary for network building and maintenance or, alternatively, whether certain innate personal- ity traits favour academic achievement, career success and social relationships. This latter proposition is the ‘selection’ model favoured by Polansky and Crittenden, while other researchers have emphasized the role of social competence in determin- ing both network membership and utility (Seagull, 1987).

Significantgender differencesalso exist in social support networks, with women generally displaying more stable, confiding network relationships, while men tend to rely on a combination of marital and more distant network ties, connected to work or shared leisure interests (Leavy, 1983; Wills, 1985). However, gender differences are, in their turn, influenced by such factors asmarital status, social classand culture. In general, middle-class mothers report larger networks than working class mothers, while single mothers generally have smaller networks than mothers in two-

Ecological Influences on Parenting and Child Development 711

parent families and mothers in the USA have been found to have larger networks than similar mothers in Sweden, Germany and Wales (Gunnarsson and Cochran, 1990; Bell and Ribbens, 1994). People from ethnic minorities also appear to have more restricted social networks, tending to rely on local-kin and having fewer dis- persed-kin or friends in their networks (Cross, 1990).

Community-level influences on social support and parenting

The social ecology of families and the developmental outcomes for children are also strongly influenced by the communities in which they are located. Most of the vari- ation between different communities in levels of recorded crime and child maltreat- ment can be explained by differences in a number of socio-economic and demo- graphic measures (Coultonet al., 1995; Sampsonet al., 1997). For example, in a comparative study of four areas of Chicago, with differing rates of officially recog- nized child maltreatment, 79 per cent of the variations found could be explained by nine such measures, including poverty, unemployment, family structure, over- crowding, ethnic origin, educational level and stability of residence (Garbarino and Kostelny, 1992).

In the UK, Michael Rutter and his colleagues have studied community variations in the rates of mental disorder found among children, which were much higher in Inner London than on the Isle of Wight (Rutteret al., 1975; Rutter and Quinton, 1977). Socio-economic status was found to be a much stronger predictor of mental disorder in Inner London than on the Isle of Wight (Rutter and Madge, 1976; Quinton, 1980) and this, in turn, was associated with a constellation of family prob- lems which were much more prevalent in the inner-city environment, including mar- ital discord, family breakdown, parental psychiatric history or criminal record and large family size. They concluded that the stresses of city life adversely affected families in ways that then had a negative impact on children’s mental health. Similar indirect mechanisms appear to be instrumental in community-level effects on rates of juvenile crime (Rutter and Giller, 1983), although a number of different studies have shown that schools can also have direct effects on rates of delinquency among their pupils (Rutter, 1979; Sylva, 1994; Maughan, 1994).

Work by James Garbarino and his colleagues in the United States provides further valuable insights into differences in the social ecology of parenting in what at first sight appear to be similar, stressful urban environments. In an early study of two socio-economically matched neighbourhoods, with different levels of officially recognized child maltreatment, families living in the ‘high-risk’ area reported a lack of mutually supportive relationships and informal social networks in their neighbour- hood. By contrast, families in the ‘low-risk’ area engaged more frequently in a range of neighbourhood exchanges (Garbarino and Sherman, 1980). As we have already noted, poor urban communities generally reveal restricted ties between neighbours and a greater reliance on family relationships (Coultonet al., 1995), as well as fewer neighbourhood organizations (Furstenberg, 1993; Dunstet al., 1988b) and lower

712 Gordon Jack

levels of participation in social, political and community activities (Howarthet al., 1998).

However, what Garbarino’s work has revealed is differences in the social capital of similarly poor and disadvantaged communities, which, in turn, are associated with variations in parenting capacity and developmental outcomes for children. He therefore lends support to the conclusions reached by Puttnam and his colleagues, described earlier (Puttnamet al., 1993), indicating that social capital has an import- ant influence on measures of health and well-being in the community. This is par- ticularly well illustrated by the Chicago study already cited (Garbarino and Kostelny, 1992). Two of the four matched areas studied, which had similar levels of recorded child maltreatment in 1980, were tracked over a period of six years. ‘North’, which started as a slightly ‘low-risk’ area, became a decidedly ‘high-risk’ area, whereas ‘West’, which began as a higher-risk area than ‘North’, moved in the opposite direc- tion, to become a much safer environment for children. Interviews with community leaders revealed that ‘North’ was perceived to be unstable, characterized by low levels of community participation and interactions between neighbours, whereas ‘West’ was perceived to have strong informal and formal social support systems and was described as a ‘poor but decent place to live’. Similar findings have emerged in a study conducted in Western Sydney, Australia (Vinsonet al., 1996).

The Labour government’s New Deal for Communities, unveiled in 1998, is a recognition of the need to try to rebuild social capital and regenerate run-down neighbourhoods in Britain. A report by the Social Exclusion Unit (1998) highlights the features of their estates disliked most by residents living in run-down areas. These include lack of shops, vandalism, fear of violence (and crime in general), unsupervised youngsters, poor public transport, danger and pollution from traffic, drug dealing, a lack of community spirit and conflict with neighbours. One example of neighbourhood regeneration is provided by Hyde Park in Leeds, the scene of street riots in 1995. A Community Safety Project, evaluated by NACRO over a three-year period, brought together statutory agencies, churches, student unions and local residents, with the aim of reducing crime and improving the general atmo- sphere of the area. Through changes in policing, the introduction of various crime prevention strategies and improved information services, residents’ perceptions of the area have changed significantly over the period of the study. Awareness of local activities has risen, along with the feeling that the area is improving and that com- munity spirit is growing. These improvements in social capital have been accompan- ied by a fall of nearly a half in recorded crime and the perception among residents that problems such as racial harassment, drug dealing, burglary, car-theft and young people ‘hanging-around’ are beginning to be reduced (Guardian, 1999b).

Unfortunately, schemes of this nature are likely to have only a limited or short- term impact unless they are accompanied by successful strategies to tackle the ser- ious income inequalities that exist in the UK. At the present levels of expenditure, the New Deal for Communities can only reach a tiny proportion of the deprived areas that require urgent regeneration. Even in the targeted neighbourhoods, the government’s emphasis on trying to enforce the social inclusion of marginalized groups through the formal labour market, rather than working with local people and

Ecological Influences on Parenting and Child Development 713

groups to build on their existing patterns of social interaction and strategies for survival, is likely to be counter-productive (Jack and Jordan, 1999).

Conclusion: Implications for practice

The ecological approach to social work with children and families, which has been described here, is not something which can merely be added to the social worker’s ‘tool-kit’ of skills and techniques, to be used selectively, as and when appropriate. Rather, it should be thought of as the tool-kit itself, out of which the various methods of assessment and intervention can be selected. It is the cultural environment within which all other polices and practices should be developed. If it is simply ‘bolted-on’ to existing organizational structures, it is likely to have only limited and short-term effects. So what are the essential components of the ecological model that have emerged from the review of the available evidence that has been presented here?

First, the ecological approach highlights the fundamental role played by poverty and inequality in influencing the health and development of children, mainly through its negative impact on family functioning. The clear message is that children living in impoverished and disadvantaged circumstances should be recognized as ‘children in need’, entitled to the provision of statutory services (under s.17 of the 1989 Children Act), in just the same way as children who suffer ‘significant harm’ or are disabled (Aldgate and Tunstill, 1995). However, it has also been clearly demon- strated that poverty is not an isolated phenomenon—it is intimately connected to a whole range of other factors, including education, employment, housing and health care. The implication is therefore that social workers and their managers and employing agencies, in both the statutory and the voluntary/independent sectors, need to actively engage with all of these issues. This should be driven, at the local level, by the development of Children’s Services Plans, based on detailed knowledge of the social demography of the areas being served and consultation and partnerships with local people and all the relevant agencies (Department of Health/Department for Education and Employment, 1996). It is no longer good enough for social welfare agencies, primary health care teams or schools to respond to the set of interacting factors which seriously disadvantage the children they are responsible for by burying their heads in the sand or trying to pass the responsibility on to someone else. There is a growing research-based literature which can assist all agencies involved in promoting and protecting children’s welfare to develop more effective responses to these complex problems (see, for example, Coulton, 1996; Weil, 1996; Cochran, 1993; Audit Commission, 1994; Bywaters and McLeod, 1996; Blackburn, 1991; Donnison, 1998).

Secondly, this approach highlights the powerful influence of the social interac- tions that take place between members of a community. Their frequency, mutual- ity, inclusiveness, availability and supportiveness affect individual and collective well-being and health. Communities displaying high levels of social cohesion tend to have lower levels of crime and child maltreatment and their individual

714 Gordon Jack

members are less likely to suffer from mental disorders in both childhood and adulthood (Sampsonet al., 1997; Garbarino and Kostelny, 1992). However, social capital and the personal social networks on which it is based, are undermined or restricted by inequalities and divisions within society (Jack and Jordan, 1999). People’s perceptions of their situations, relative to others who they see on a day-to-day basis (in their neighbourhood, at the shops, travelling to work or at the job centre) has a powerful influence on their sense of worth and on their healthy development. Often, the groups or individuals who are in the greatest need of additional, more reliable, or more satisfying sources of social support, are those who are least likely to have access to these things. The temptation for welfare professionals might be to respond by seeking to plug these deficiencies with formal, professionally provided or arranged services. However, the ecological literature suggests thatinformal sources of support, which involve enduring per- sonal relationships of equality, mutual trust and reciprocity, and which place control over the help accepted and offered in the hands of the individuals concerned, tend to be far more effective in enhancing personal functioning (Dunst et al., 1997). The implications are two-fold; first professionals trying to help families and children with unmet needs should direct their efforts towards strengthening natural support systems, involving relatives, friends, neighbours and community organizations, rather than trying to create new systems. Secondly, where there is no alternative to creating or utilizing professionally controlled or organized formal support systems, they should try to mirror those aspects of the informal support systems, outlined above, which appear to be important for the most beneficial outcomes (see, for example, Ball, 1996; Oldset al., 1986; Olds et al., 1994; Johnsonet al., 1993; MacMillan et al., 1994).

Thirdly, it is clear from the evidence presented here that the most effective ways of promoting or protecting the health and development of children often involve ensuring adequate social support is available to their parents. In this connection, the central importance for mothers (especially those living in poverty) of a supportive and reliable relationship with a partner or close friend and support (rather than additional stress) from their relatives should be recognized. Anything that can be done to strengthen the support available from these sources will ultimately have beneficial effects for children, through the effects which such network relationships can have on parent and family functioning. As we have seen, theperceptionthat such support is available is a powerful protective factor against stress and conflictual or unreliable relationships are a significant source of increased risk (Brown and Harris, 1978).

A fourth area of importance is the way that the ecological model establishes a clear picture of the balance of stresses experienced by a family and the supportive resources available to them. The accumulation of different risk factors in a child’s environment, rather than individual, isolated stressful life events, are usually most harmful to children’s development (Garmezy, 1994). Accurate and effective assessments must therefore collect and organize information in partner- ship with parents and children about individual characteristics, family and com- munity sources of stress and support and the interactions between these different

Ecological Influences on Parenting and Child Development 715

influences on children’s well-being. This requires that those undertaking assess- ments have a good understanding of the social ecology of families and, in particular, that they develop a clear picture both of each family member’s per- sonal social support networks and the resources potentially available in the wider community. There are a number of useful guides to assessing sources of social support and the capacity of community resources to provide support to children and families, which can be utilized (e.g. Nelson, 1997; Dunst and Trivette, 1990; Hawtin et al., 1994; Warren and Warren, 1977, pp. 167–96; McKnight, 1987; Kretzmann and McKnight, 1993). The philosophy is always to build on existing cultural practices and strengths and to develop and maintain informal systems of support wherever possible (Trivetteet al., 1997).

The final component of the ecological model highlighted is the importance of utilizing the messages for practice provided by the study ofresilience. Many children and families living in high-risk environments of one sort or another are protected by certain innate characteristics or particular features of their family circumstances. Amongst those described above, the significance for a child of an enduring relation- ship with an adult, or involvement in esteem- or skill-enhancing activities outside the family, should be borne in mind. For example, evaluation of what are described as ‘urban sanctuaries’ has revealed the features of after-school programmes which are particularly successful in protecting and enhancing the development of inner-city teenagers in the United States (McLaughlinet al., 1994).

It is also worth pointing out that some of the features of resilience can be taught or developed in families and children. Many early intervention schemes aim to do just this (see Macdonald and Roberts, 1995, for a summary of evidence-based approaches), either through parenting programmes, home visitation schemes, family centres, or wider community-based programmes (Grimshaw and McGuire, 1998; Olds et al., 1986; Oldset al., 1994; Jack, 1997, 1998).

Accepted: October 1999

References

Aldgate, J. and Tunstill, J. (1995)Making Sense of Section 17. Implementing Services for Children in Need within the 1989 Children Act, London, HMSO.

Audit Commission (1994)Seen but Not Heard: Co-Ordinating Community Child Health and Social Services for Children in Need, London, HMSO.

Baldwin, A. L., Baldwin, C. and Cole, R. E. (1990) ‘Stress-resistant families and stress- resistant children’, in Rolf, J., Masten, A., Cicchetti, D., Neuchterlein, K. and Wintraub, S. (eds),Risk and Protective Factors in the Development of Psychopathology, Cambridge, Cambridge University Press, pp. 257–80.

Ball, M. (1996) ‘Round and Round the Circle’. Support for Isolated Families through the Department of Health’s Childcare Circles Initiative, London, Department of Health.

Ball, M. (1998) School Inclusion: the School, the Family and the Community, York, Joseph Rowntree Foundation (Findings, June).

716 Gordon Jack

Barker, R. (1978)Habitats, Environments and Human Behaviour, San Francisco, Jossey-Bass. Bebbington, P., Hurry, J., Tennant, C., Stuart, E. and Wing, J. (1981) ‘Epidemiology of mental

disorders in Camberwell’,Psychological Medicine, 11, pp. 561–79. Beckman, P. J. (1991) ‘Comparison of mothers’ and fathers’ perceptions of the effect of young

children with and without disabilities’,American Journal on Mental Retardation, 95, pp. 585–95.

Bell, L. and Ribbens, J. (1994) ‘Isolated housewives and complex maternal worlds—the signi- ficance of social contacts between women with young children in industrial societies’,The Sociological Review, 42(2), pp. 227–62.

Belsky, J. (1984) ‘The determinants of parenting: a process model’,Child Development, 55, pp. 83–96.

Blackburn, C. (1991)Poverty and Health: Working with Families, Buckingham, Open Univer- sity Press.

Blaxter, M. (1990)Health and Lifestyles, London, Routledge. Bradley, R. H., Whiteside, L., Mundfrom, D. J., Casey, P. H., Kelleher, K. J. and Pope, S. K.

(1994) ‘Early indications of resilience and their relation to experiences in the home envir- onments of low birthweight, premature children living in poverty’,Child Development, 65, 346–60.

Brassard, J. (1982) ‘Beyond Family Structure: Mother–Child Interaction and Personal Social Networks’, Unpublished doctoral dissertation, Ithaca, NY, Cornell University.

Bronfenbrenner, U. (1979)The Ecology of Human Development, Cambridge, MA., Harvard University Press.

Bronfenbrenner, U. (1986) ‘Ecology of the family as a context for human development: research perspectives’,Developmental Psychology, 22(6), pp. 723–42.

Brown, G. W. and Harris, T. (1978)Social Origins of Depression, London, Tavistock. Brown, G. W., Andrews, B., Harris, T. O., Adler, Z. and Bridge, I. (1986) ‘Social support,

self-esteem and depression’,Psychological Medicine, 16, pp. 813–31. Brownell, A. and Shumaker, S. A. (1984) ‘Social support: an introduction to a complex phe-

nomena’,Journal of Social Issues, 40, pp. 1–9. Bywaters, P. and McLeod, E. (eds) (1996)Working for Equality in Health, London, Routledge. Cassell, J. (1974) ‘Psychosocial processes and >ODQ<stress’’: theoretical formulations’,Inter-

national Journal of Health Services, 4, pp. 471–82. Cobb, S. (1976) ‘Social support as a moderator of life stress’,Psychosomatic Medicine, 38,

pp. 300–14. Cochran, M. (1993) ‘Parenting and personal social networks’, in Luster, T.and Okagaki, L.

(eds),Parenting: an Ecological Perspective, Hillsdale, NJ, Lawrence Erlbaum Associates, pp. 149–78.

Cochran, M. M. and Brassard, J. A. (1979) ‘Child development and personal social networks’, Child Development, 50, pp. 601–15.

Cohen, S. and Wills, T. A. (1985) ‘Stress, social support and the buffering hypothesis’,Psycho- logical Review, 98(2), pp. 310–57.

Coohey, C. (1996) ‘Child maltreatment: testing the social isolation hypothesis’,Child Abuse and Neglect, 20(3), pp. 241–54.

Coulton, C. J. (1996) ‘Poverty, work and community: a research agenda for an era of dimin- ishing federal responsibility’,Social Work, 41(5), pp. 509–20.

Coulton, C. J., Korbin, J. E., Su, M. and Chow, J. (1995) ‘Community level factors and child maltreatment rates’,Child Development, 66, pp. 1262–76.

Coyne, J. C. and Downey, G. (1991) ‘Social factors and psychopathology: stress, social support and coping processes’,Annual Review of Psychology, 42, pp. 401–25.

Crittenden, P. M. (1985) ‘Social networks, quality of child rearing and child development’, Child Development, 56, pp. 1299–313.

Ecological Influences on Parenting and Child Development 717

Crnic, K. A. and Stormshak, E. (1997) ‘The effectiveness of providing social support for families of children at risk, in Guralnick, M. J. (ed.),The Effectiveness of Early Interven- tion, Baltimore, MA, Paul H. Brookes, pp. 209–25.

Crockenberg, S. (1988) ‘Social support and parenting’, in Fitzgerald, H., Lester, B. and Yogman, M. (eds),Theory and Research in Behavioural Paediatrics, (Vol.4), New York, Plenum, pp. 67–92.

Cross, W. (1990) ‘Race and ethnicity: effects on social networks’, in Cochran, M., Larner, M., Riley, D., Gunnarson, L. and Henderson, C., Jr. (eds),Extending Families: The Social Networks of Parents and their Children, London, Cambridge University Press, pp. 67–85.

Davey Smith, G., Shipley, M. J. and Rose, G. (1990) ‘Magnitude and causes of socioeconomic differentials in mortality: further evidence from the Whitehall study’,Journal of Epidemi- ology and Community Health, 44, pp. 265–70.

Department of Health and Department for Education and Employment (1996)Children’s Ser- vices Planning: Guidance, London, HMSO.

Department of Social Security (1994)Households Below Average Income: A Statistical Ana- lysis 1979–1991/2, London, HMSO.

Department of Social Security (1998)A New Contract for Welfare, London, The Stationery Office.

Donnison, D. (1998)Policies for a Just Society, Basingstoke, Macmillan. Duncan, G. J., Brooks-Gunn, J. and Klebanov, P. K. (1994) ‘Economic deprivation and early

childhood development’,Child Development, 65, pp. 296–318. Dunst, C. J. and Trivette, C. M. (1990) ‘Assessment of social support in early intervention

programs’, in Meisels, S. J. and Shonkoff, J. P. (eds),Handbook of Early Childhood Education, New York, Cambridge University Press.

Dunst, C. J., Leet, H. E. and Trivette, C. M. (1988a) ‘Family resources, personal wellbeing and early intervention’,Journal of Special Education, 22, pp. 108–16.

Dunst, C. J., Snyder, S. and Mankinen, M. (1988b) ‘Efficacy of early intervention’, in Wang, M., Reynolds, M.and Walberg, H. (eds),Handbook of Special Education: Research and Practice: Vol. 3 Low Incidence Conditions, Oxford, Pergamon Press, pp. 259–94.

Dunst, C. J., Trivette, C. M. and Jodry, W. (1997) ‘Influences of social support on children with disabilities and their families’, in Guralnick, M. J. (ed.),The Effectiveness of Early Intervention, Baltimore MA, Paul H. Brookes, pp. 499–522.

Fischer, C. (1982)To Dwell among Friends: Personal Networks in Town and City, Chicago, University of Chicago Press.

Furstenberg, F. F., Jr. (1993) ‘How families manage risk and opportunity in dangerous neigh- bourhoods’, in Wilson, W. (ed.)Sociology and the Public Agenda, Newbury Park, CA, Sage, pp. 231–58.

Garbarino, J. and Kostelny, K. (1992) ‘Child maltreatment as a community problem’,Child Abuse and Neglect, 16, pp. 455–64.

Garbarino, J. and Sherman, D. (1980) ‘High-risk neighbourhoods and high-risk families: the human ecology of child maltreatment’,Child Development, 51, pp. 188–98.

Garmezy, N. (1994) ‘Reflections and commentary on risk, resilience and development’, in Haggerty, R. J., Sherrod, L. R., Garmezy, N. and Rutter, M. (eds),Stress, Risk and Resili- ence in Children and Adolescents, Cambridge, Cambridge University Press.

Gibbons, J. (1990)Family Support and Prevention: Studies in Local Areas, London, HMSO. Grimshaw, R. and McGuire, C. (1998)Evaluating Parenting Programmes: A Study of Stake-

holders’ Views, London, National Children’s Bureau. Guardian, The (1997) ‘Children in poverty: Britain tops the European league’, 28 April, p. 1. Guardian, The (1999a) ‘Street lives’, 3 March, Society Section, pp. 6–7. Guardian, The (1999b) ‘Past is binned’, 17 March, Society Section, pp. 6–7.

718 Gordon Jack

Gunnarsson, L. and Cochran, M. (1990) ‘The social networks of single parents: Sweden and the United States’, in Cochran, M., Larner, M., Riley, D., Gunnarsson, L. and Henderson, C., Jr., (eds),Extending Families: The Social Networks of Parents and their Children, Cambridge, Cambridge University Press, pp. 105–18.

Hashima, P. Y. and Amato, P. R. (1994) ‘Poverty, social support and parental behaviour’, Child Development, 65, pp. 394–403.

Hawtin, M., Hughes, G. and Percy-Smith, J. (1994)Community Profiling: Auditing Social Needs, Buckingham, Open University Press.

Hertzmann, C., Frank, J. and Evans, R. (1990)Heterogeneities in Health Status. CIAR Internal Document No. 3, Toronto, Canadian Institute for Advanced Research.

Howarth, C., Kennedy, P., Palmer, G. and Street, C. (1998)Monitoring Poverty and Social Exclusion: Labour’s Inheritance, York, Joseph Rowntree Foundation (Findings, December 1998).

Jack, G. (1997) ‘Discourses of child protection and child welfare’,British Journal of Social Work, 27, pp. 659–78.

Jack, G. (1998) ‘The social ecology of parents and children: implications for the development of child welfare services in the UK’,International Journal of Child and Family Welfare, 98(1), pp. 78–88.

Jack, G. and Jordan, B. (1999) ‘Social capital and child welfare’,Children and Society, 13, pp. 242–56.

Jenkins, J. M. and Smith, M. A. (1990) ‘Factors protecting children living in disharmonious homes: maternal reports’,Journal of the American Academy of Child and Adolescent Psy- chiatry, 29, pp. 60–9.

Johnson, Z., Howell, F. and Molloy, B. (1993) ‘Community mothers programme: randomised controlled trial of non-professional intervention in parenting’,British Medical Journal, 306, pp. 1449–52.

Joseph Rowntree Foundation (1995)The Joseph Rowntree Inquiry into Income and Wealth, York, Joseph Rowntree Trust.

Kawachi, I., Kennedy, B. P., Lochner, K. and Prothrow-Stith, D. (1997) ‘Social capital, income inequality and mortality’,American Journal of Public Health, 87, pp. 1491–98.

Kotch, J. B., Browne, D. C., Ringwalt, C. L., Dufart, V., Ruina, E., Stewart, P. W. and Jung, J. W. (1997) ‘Stress, social support and substantiated maltreatment in the second and third years of life’, Child Abuse and Neglect, 21(11), pp. 1025–37.

Kretzmann, J. and McKnight, J. (1993)Building Community from the Inside Out, Evanston, IL, North-Western University Center for Urban Affairs and Policy Research.

Kumar, V. (1993) Poverty and Inequality in the UK: The Effects on Children, London, National Children’s Bureau.

Lacharité, C., Ethier, L. and Couture, G. (1996) ‘The influence of partners on parental stress of neglectful mothers’,Child Abuse Review, 5, pp. 18–33.

Leavy, R. L. (1983) ‘Social support and psychological disorder: A review’,Journal of Com- munity Psychology, 11, pp. 3–21.

Leon, D. A., Vagero, D. and Otterblad Olausson, P. (1992) ‘Social class differences in infant mortality in Sweden: a comparison with England and Wales’,British Medical Journal, 305, pp. 687–91.

Macdonald, G. M. and Roberts, H. (1995)What Works in the Early Years?, Barkingside, Barnardos.

MacMillan, H. L., MacMillan, J. H., Offord, D. R., Griffith, L. and MacMillan, A. (1994) ‘Primary prevention of child physical abuse: a critical review, Part I’,Journal of Child Psychology and Psychiatry, 35(5), pp. 835–56.

Marmot, M. B. and Davey Smith, G. (1989) ‘Why are the Japanese living longer?’,British Medical Journal, 299, pp. 1547–51.

Ecological Influences on Parenting and Child Development 719

Maughan, B. (1994) ‘School influences’, in Rutter, M. and Hay, D. (eds),Development through Life: A Handbook for Clinicians, Oxford, Blackwell Scientific.

McKnight, J. (1987) ‘Regenerating community’, Social Policy, Winter, pp. 54–8. McLaughlin, M. W., Irby, M. A. and Langman, J. (1994)Urban Sanctuaries: Neighborhood

Organisations in the Lives and Futures of Inner-City Youth, San Francisco, Jossey-Bass. Mitchell, R. E., Billings, A. G. and Moos, R. H. (1982) ‘Social support and well-being:

implications for prevention programs’,Journal of Primary Prevention, 3, pp. 77–98. Nazroo, J. Y. (1997)The Health of Britain’s Ethnic Minorities, London, Policy Studies Institu-

te. Nelson, D. (1997)The Building Blocks of Neighborhoods that Support Families, Baltimore,

Casey Foundation. Olds, D. L., Henderson, C. R., Chamberlin, R. and Tatelbaum, R. (1986) ‘Preventing child

abuse and neglect: a randomized trial of nurse home visitation’,Pediatrics, 78(1), pp. 65–78.

Olds, D. L., Henderson, C. R. and Kitzman, H. (1994) ‘Does parental and infancy nurse home visitation have enduring effects on qualities of parental caregiving and child health at 25 to 50 months of life?’,Pediatrics, 93(1), pp. 89–98.

Phillimore, P., Beattie, A. and Townsend, P. (1994) ‘The widening gap. Inequality in health in Northern England, 1981–1991’,British Medical Journal, 308, pp. 1125–28.

Platt, L. and Noble, M. (1999)Race, Place and Poverty: Ethnic Groups and Low Income Distributions, York, Joseph Rowntree Foundation (Findings, February).

Polansky, N. A., Ammons, P. W. and Gaudin, J. M. (1985a) ‘Loneliness and isolation in child neglect’,Social Casework, 66, pp. 38–47.

Polansky, N. A., Gaudin, J. M., Ammons, P. W. and David, K. B. (1985b) ‘The psychological ecology of the neglectful mother’,Child Abuse and Neglect,9, pp. 265–75.

Puttnam, R. D., Leonardi, R. and Nanetti, R. Y. (1993)Making Democracy Work: Civic Traditions in Modern Italy, Princeton, NJ, Princeton University Press.

Quinton, D. (1980) ‘Family life in the inner city: myth and reality’, in Marland, M. (ed.), Education for the Inner City, London, Heinemann, pp. 45–67.

Riley, D. and Shaw, M. (1985)Parental Supervision and Juvenile Delinquency, Home Office Research Study 83, HMSO.

Ross, D. P., Shillington, E. R. and Lochhead, C. (1994)The Canadian Fact Book on Poverty— 1994, Ottowa, ON, Canadian Council on Social Development.

Rutter, M. (1979)Fifteen Thousand Hours: Secondary Schools and their Effects on Children, Cambridge, MA, Harvard University Press.

Rutter, M. and Giller, H. J. (1983)Juvenile Delinquency: Trends and Perspectives, New York, Penguin Books.

Rutter, M. and Madge, N. (1976)Cycles of Disadvantage, London, Heinemann. Rutter, M. and Quinton, D. (1977) ‘Psychiatric disorder—ecological factors and concepts of

causation’, in McGurk, H. (ed.),Ecological Factors in Human Development, Amsterdam, North Holland, pp. 173–87.

Rutter, M., Cox, A., Tupling, C., Berger, M. and Youle, W. (1975) ‘Attainment and adjustment in two geographical areas. I. The prevalence of psychiatric disorder’,British Journal of Psychiatry, 126, pp. 493–509.

Rutter, M., Quinton, D. and Hill, J. (1990) ‘Adult outcome of institution-reared children: males and females compared’, in Robins, L. and Rutter, M. (eds),Straight and Devious Pathways from Childhood to Adulthood, Cambridge, Cambridge University Press, pp. 135–57.

Sampson, R. J., Raudenbush, S. W. and Earls, F. (1997) ‘Neighbourhoods and violent crime: a multi-level study of collective efficacy’,Science, 277, pp. 1–7.

Sarason, B. R., Sarason, I. G. and Pierce, G. R. (eds), (1990)Social Support: an Interactional View, New York, Wiley.

720 Gordon Jack

Seagull, E. A. W. (1987) ‘Social support and child maltreatment: a review of the evidence’, Child Abuse and Neglect, 11, pp. 41–52.

Sheppard, M. (1994) ‘Childcare, social support and maternal depression: a review and applica- tion of findings’, British Journal of Social Work, 24, pp. 287–310.

Smaje, C. (1995)Health, ‘Race’ and Ethnicity: Making Sense of the Evidence, London, King’s Fund Institute.

Smale, G., Tuson, G., Ahmed, B., Darvill, G., Domoney, L. and Sainsbury, E. (1994)Negotiat- ing Care in the Community, London, HMSO.

Social Exclusion Unit (1998)Bringing Britain Together: A National Strategy for Neighbour- hood Renewal, London, The Stationery Office.

Spencer, N. (1996) ‘Reducing child health inequalities’, in Bywaters, P.and McLeod, E. (eds), Working for Equality in Health, London, Routledge, pp. 143–60.

Steinhauer, P. D. (1998) ‘Developing resiliency in children from disadvantaged populations’, in Canada Health Action, Building on the Legacy (Vol. 1), Children and Youth, Sainte-Foy, Quebec, Editions Multi-Mondes.

Sylva, K. (1994) ‘School influences on children’s development’,Journal of Child Psychology and Psychiatry, 35, pp. 135–70.

Thompson, R. A. (1995)Preventing Child Maltreatment through Social Support, Thousand Oaks, CA, Sage.

Tolsdorff, C. C. (1976) ‘Social networks, support and coping: an exploratory study’,Family Process, 15, pp. 407–17.

Trivette, C. M., Dunst, C. J. and Deal, A. G. (1997) ‘Resource-based approach to early inter- vention’, in Thurman, S. K., Cornwell, J. R. and Gottwald, S. R. (eds),Contexts of Early Intervention: Systems and Settings, Baltimore, MA, Paul H. Brookes, pp. 73–92.

Vagero, D. and Lundberg, O. (1989) ‘Health inequalities in Britain and Sweden’,The Lancet, 11, pp. 35–6.

Vinson, T., Baldry, W. and Hargreaves, J. (1996) ‘Neighbourhoods, networks and child abuse’, British Journal of Social Work, 26, pp. 523–43.

Warren, R. and Warren, D. (eds) (1977)The Neighborhood Organizer’s Handbook, South Bend, IN, University of Notre Dame Press.

Weil, M. O. (1996) ‘Community building: building community practice’,Social Work, 41(5), pp. 481–500.

Werner, E. (1995) ‘Resilience in development’,Current Directions in Psychological Science, American Psychological Society, 4(5), pp. 81–5.

Werner, E. and Smith, R. (1992)Overcoming the Odds: High-risk Children from Birth to Adulthood, Ithaca, NY, Cornell University Press.

Wilkinson, R. G. (1996)Unhealthy Societies: The Afflictions of Inequality, London Routledge. Wills, T. A. (1985) ‘Supportive functions of inter-personal relationships’, in Cohen, S. and

Syme, S. L. (eds),Social Support and Health, Orlando, FI, Academic Press, pp. 61–82. Woodroffe, C., Glickman, M., Barker, B. and Power, C. (eds) (1993)Children, Teenagers and

Health, Milton Keynes, Open University Press.