table of contents | abstracts


Reducing Risk and Promoting Resilience in Vulnerable Children

Sonia Jackson



Theories of risk and resilience developed by researchers such as Werner, Garmezy, Rutter, Quinton and Fraser can help us to understand the poor outcomes experienced by vulnerable children, especially those in care as a result of abuse and neglect or family breakdown. There are important implications for practice which form the basis of the British "Looking After Children" system, aiming to ensure that all aspects of children and young people's development receive proper attention and appropriate action. Recent research in the UK shows that educational success is a crucial element in determining developmental progress generally and in particular the quality of life in adulthood. In the past, social workers have not given this domain the priority it merits, with very adverse effects on those who spend long periods in public care. This paper suggests that there are many positive steps that can be taken to promote resilience in vulnerable children, and that stimulating educational motivation and providing support for school achievement and continuing education are among the actions most likely to lead to better outcomes.

In this paper I want to consider how social workers may be able to influence the life chances of vulnerable children using an approach built on emerging theories of risk and resilience. By vulnerable children I mean those growing up in adverse conditions that are known to lead to poor outcomes in adolescence and adulthood. In the past a great deal of research effort has gone into identifying what these factors are and attempting to trace causal relationships. Over the past ten years, however, there has been a significant shift in interest. The question we are asking now is not why children with all these negative factors in their lives have problems, but how is it that many do not. Longitudinal studies have consistently found that a third to a half of children from severely disrupted, deviant or abusive families nevertheless grow up to be relatively well-adjusted adults (Rutter, 1985, Werner, 1989).

The concept of resilience has developed in response to this question, and has been progressively elaborated and refined to take account of new research, particularly by the team based at the University of London Institute of Psychiatry, led by Michael Rutter. It is of great importance to social workers because it provides hope that their work with children and youth can really make a difference, but only if it is based on a clear theoretical understanding and not simply on good intentions.

What do we mean by resilience? The most widely used definition is "successful adaptation despite adversity." In the case of children it is usually taken to mean development in line with expected age norms and the absence of psychopathology when poor outcomes might have been predicted on the basis of the child's background and history. In the early days, resilience was often discussed as though it were an unchanging trait, probably genetic, which children either had or did not have. Some children were even described as "invulnerable," as if they remained unaffected by any adversity, however severe (Anthony and Cohler, 1987). Such a notion is now generally discredited. There are wide variations in resistance to stress, but rather than being a fixed quality, resilience is the product of the interaction between risk and protective factors or processes.

Garmezy has identified three different types of resilience (Masten, Best and Garmezy, 1990). The first, "overcoming the odds," is defined by the attainment of positive outcomes despite high risk status. An example of this kind of resilience would be a child from a family with a record of drug misuse and criminality who does well at school and is involved in non-deviant community activities. The second type is "sustained competence under stress:" the child who shows an ability to cope with, for example, family violence or parental illness, while continuing to function well in other aspects of life. The third type of resilience is "recovery from trauma" which includes children who continue to behave and develop normally despite having experienced an intensely stressful event, such as the death of a parent, suffering a serious injury or being exposed to armed conflict. In practice, vulnerable children are likely to experience many variable stresses either simultaneously or at different times in their lives. All three types of resilience may be necessary if they are to achieve a satisfactory quality of adult life.



Risk factors are any influences that increase the probability of onset, exacerbation or maintenance of a problem condition, ranging from prenatal and genetic to broad environmental conditions. The effect of risk factors on development is cumulative and also sequential; a risk factor may or may not set off a risk process. Rutter has pointed out that psychosocial risk is multifactorial. Single negative experiences do not carry a major risk of later psychopathology. For example, Rutter and Quinton (1977) found that six family variables were strongly associated with child psychiatric disorder, but children with only one of these risk factors were no more likely to have mental health problems than those with no risk factors at all. But when any two of these stresses occurred together, the risk increased fourfold, and similarly as each additional factor was added. These findings were confirmed by Fergusson and Lynskey (1996) in New Zealand. They developed an adversity index based on 39 measures of family life covering variables such as economic disadvantage, maladaptive parent-child interaction, martial conflict and parental separation. They used various measures of psychosocial problems to assess their subjects and found a strikingly steep risk gradient, with the severe problem behavior's over 100 times more likely in adolescence for those with high family adversity scores as compared with the low adversity group.

On the other hand there is also evidence from longitudinal studies that things can change in unexpected directions. Experiences in adolescence or early adult life can lead to good outcomes for individuals who seemed "born to fail." Antisocial behaviors in childhood normally carries a high risk of later psychosocial problems and deviancy (Rutter et al, 1995). Such behavior may elicit hostile responses in other people, setting off a chain of negative effects. However, negative chain effects can be interrupted by experiences that break the cycle (Rutter, 1996), and such "turning points" are often prominent in the life stories of resilient people. For example, a stable and harmonious marriage to a non-deviant partner much reduces the risk of antisocial behavior continuing into adulthood.



The remainder of this paper will focus on a particular group of especially vulnerable children: those separated from their parents and living in foster or residential care. In England and Wales, the Children Act of 1989 introduced the term "looked after by local authorities" to describe the situation of these children. The change in terminology had the laudable aim of reducing the stigma attached to care and emphasizing the voluntary nature of most care arrangements. "In care" technically refers only to those who are placed away from home under a court order. However, in this paper it should be taken to mean children in out-of-home placement, either in group homes or families.

There is no doubt that separation from one's family of origin, for whatever reason, is a high risk factor for psychopathology and poor developmental outcomes. However it is also clearly a product of previous risk factors. We know that children who come into care are overwhelmingly drawn from families with multiple disadvantages. They are much more likely than other children to come from a single-parent household, to have a minority ethnic background, to live in poor housing with a low family income and to have a larger than average number of siblings. But even when all these factors contributing to a likelihood of material disadvantage are combined, the chances are only one in ten that a child will enter the care system (Bebbington and Miles, 1986). For this to happen, as Osborn and St. Claire (1987) pointed out, there have to be other factors, most commonly the mental illness of a parent, domestic violence and physical and/or sexual abuse of the child. The family has probably become isolated or alienated from friends and relatives or someone would have stepped in to care for the children. Reception into care is normally a last resort; families will fight fiercely to keep their children whatever the difficulties and, at least in Britain, social workers are reluctant to place children away from home unless it is absolutely unavoidable.

Coming into care may be either a continuation of a risk process that has probably started for this child at or possibly before birth, or it can and should be the start of a protective process. There is ample evidence that this can be the case when the child is placed for adoption with a stable, loving and competent family. The adoption of children from Romanian orphanages is a striking illustration of the remarkable powers of resilience that children can display with adequate nurturing. These children had suffered at least two years of severe emotional and physical privation and at the time of adoption most scored in the moderately to severely retarded range on intelligence tests. By the age of six, the majority had caught up by an amazing 40 points, with the mean developmental quotient now in the upper end of the normal range.
Adoption generally has a very good record of success, especially when it occurs at an early age (Treseliotis, 1997). This is not difficult to understand, since children are normally removed from a situation with multiple risk factors to one where most of these factors are absent, thus initiating a protective process which may in time compensate for the risk factors deriving from the child's family of origin or pre-care experiences.

Short stays in care can also be protective, helping families to regroup and recover from crises and may have positive benefits for the children (Bullock, Little and Millham, 1993). For those who stay in care for longer periods, the picture is much less hopeful. There are clear indications that the out-of-home care system as it currently operates in Britain is failing large numbers of children and young people. The strongest evidence comes from overrepresentation of ex-care people among the homeless and in custodial institutions and treatment centers. For example, a recent inquiry by the Howard League for Penal Reform into the use of prison custody for teenage girls found that 40 percent of 15- to 17-year olds had been in care (Russell, 1998). Leaving care studies consistently show that careleavers are many times more likely than their peers to be unemployed, to become pregnant in adolescence, to be homeless, to engage in health-threatening behavior such as smoking and misusing alcohol or drugs and to suffer from eating disorders and mental health problems (Biehal, Clayden, Stein and Wade, 1992; Garnett, 1992; Garnett, 1995; Stein, 1994; Action on Aftercare Consortium, 1996; Broad, 1997).
Because, as already described, children "looked after" by local authorities almost always come from severely disadvantaged families, it is sometimes argued that poor outcomes are only to be expected and could have been predicted from the preponderance of pre-care risk factors. Such a pessimistic view ignores the impact of the care experience itself and its potential, largely unrealized at present, for initiating protective processes.



A working party set up by the UK Department of Health to explore the measurement of outcomes for children in care came to the conclusion that one of the highest risk factors for poor outcomes was the absence, for most children, of a consistent parental figure with an holistic overview of the child's developmental needs (Parker, Ward, Jackson, Aldgate and Wedge, 1991). As a result, many of the ordinary actions that parents take to protect their offspring from adversity, to anticipate risk processes and promote advantageous ones simply did not occur. Building on this work, the research team developed a system known as Looking After Children designed to improve the quality of parenting received by separated children and introduce greater continuity in their care (Ward, 1995). The core of the system consists of six age-related "Assessment and Action Records" covering seven domains of child development. These are Health, Education, Family and Social Relationships, Identity, Emotional and Behavioral Development, Social Presentation and Self-care Skills. For each domain, developmentally appropriate outcomes are specified and questions asked about whether actions have been taken that are likely to lead to their achievement. This scheme has been very widely adopted by British local authorities and adapted for use in over ten other countries. It represents perhaps the most structured attempt yet to reduce risk and enhance protective factors for children in out-of-home placement, but as yet there are no reliable research findings to show that it actually does so (Jackson, 1998a).



The Looking After Children system assumes that the answer to this question is "yes" - at least there is no explicit ranking of the different dimensions except what is implied in the order in which they occur in the Assessment and Action Records. I want now to suggest, on the basis of a recently completed empirical research study, that if we truly want to promote resilience in separated children, education should be given the highest priority. That does not mean that children's other needs should be neglected, it simply means that there is clear evidence that educational success is the most potent protective factor and conversely, educational failure, which is the experience of the majority, carries the greatest risks.

The school achievement of children in care has always been poor in Britain, as in other English-speaking countries, but in recent years there has been a drive to improve the educational level of the population. The result of this is that a wider and wider gap has opened between children and their own families, even poor children, and those in the care system (Jackson, 1998b). Leaving care studies generally estimate that over 75 percent leave school with no qualification at all. Only five percent have five or more good passes in the General Certificate of Secondary Education (GCSE), which is the public examination taken at 16, and the entry level for most forms of further education or training, compared with 68 percent of all 16- to 19-year-olds. Very few continue their education after school and only one in 300 goes to university (Fletcher-Campbell, 1997), compared with 41 percent of the age group.

There was a sharp deterioration in the educational opportunities for looked after children following the passing by the Conservative Government of so-called Education Reform Acts, which set schools in competition with each other and initiated the publication of league tables of examination results. This gave schools a strong incentive to exclude, or refuse to accept in the first place, children in care whom they perceived as likely to be disruptive and unlikely to contribute positively to exam scores. The result is that a high proportion of children in the care of local authorities are denied their right to education, especially those in residential care (Blyth and Milner, 1996). It has been estimated that the chances of a child from a residential home being excluded from school is 80 times that of a child living in his or her own home.



The proportion of looked after children succeeding in education is so small that it is not surprising that very little attention has been paid to the resilient minority. This paper describes three linked research studies of people who have grown up in care and have been relatively successful educationally. The idea for this research grew from previous work on the education of children in care, which showed how far their educational achievement fell below that of their peers (Jackson, 1987, 1994), with disastrous implications for their future employability and life chances. A contributing factor to their poor performance was the low priority given to education and schooling by the child welfare system. The original purpose of the research was to identify individuals who, unlike the vast majority, had been successful in education in order to find out how their care experience had been different. It was hoped that the study would highlight aspects of good social work practice which would point the way to better opportunities for all children growing up away from home. However, the findings of the study served rather to underline the weaknesses in the care system with regard to education already revealed by earlier research and official reports (Jackson, 1989, 1992, 1994); Fletcher-Campbell and Hall, 1990; Social Services Inspectorate/OFSTED, 1997; Fletcher-Campbell, 1997).



Most local authorities know very little about the subsequent careers of children for whom they are or were responsible. In seeking to trace resilient ex-care people, therefore, there was no way of obtaining a representative sample. Instead, subjects were contacted by means of letters and articles in newspapers and, most successfully, by an insert in a magazine titled "Who Cares?" which is produced by an advocacy organization for young people in care and distributed through social service departments to about half of those over ten in the care system.
The leaflet asked for people who had spent more than a year in care and obtained five or more O levels or GSCEs at grades C or above, or were in further or higher education, to contact the researcher if they would like to take part in the study. This standard was set because it was identified by Doria Pilling in her rather neglected follow-up of the most disadvantaged children in the National Child Development Study (Pilling, 1990) as drawing a very clear line between those whose adult lifestyle was similar to that of their families of origin and those who had "escaped from disadvantage."

A total of 256 responses were received. Some did not meet the criteria, usually because the respondents had not yet taken their GSCE exams. Other wrote long letters with tales of life in care, usually with voluntary organizations, before the 1948 Children Act. Although fascinating in themselves, these obviously had less relevance to the modern care system. In all, 154 responses fully met the criteria and received questionnaires that were designed to elicit as full a picture as possible of the respondents' care and educational experiences and their family background.
The questionnaire was comprised of 90 items requiring mainly forced choice responses, though with space for additional information, and was divided into three sections. Section one asked respondents to give information about their school experience, section two examined their experience of higher education and employment and the third section asked for more sensitive personal information. The final question was open-ended and allowed respondents to add any further information. 105 questionnaires were returned fully completed, a response rate of 68 percent, very high for a postal questionnaire which must have taken considerable time to fill in. Several people used the opportunity to write at length about their lives.

This study, which we called "Successful in Care," produced much interesting and relevant information, some of which is discussed below. However, it left many questions unanswered. Because it inevitably imposed a structure on responses and did not ask for a narrative account, the connections between events, processes rather than factors, were often unclear and it was sometimes evident that respondents were reluctant to commit to paper intimate personal experiences that they might well have been willing to talk about face-to-face.

When funds became available (in the form of a grant from the Leverhulme Trust) it was decided to conduct a further study consisting of in-depth interviews with a sub-sample of those who had returned completed questionnaires. This time the criterion for inclusion was set by participation in further or higher education, and all those eligible who could be traced were interviewed: 38 people (12 men and 26 women, with a mean age of 26 years). With one exception, all were under 35. They had experienced the care system essentially in the form in which it exists today, though mainly before the full implementation of the Children Act of 1989. Thirty-seven of these people had reached an A level or diploma standard, 25 had obtained a first degree at the time of interview, several had masters degrees and one individual, a PhD.

The in-depth, semi-structured interviews usually took between two and three hours and employed open-ended questions to further explore the participant's family background, school experience, care experience, higher education, career and personal aspirations and achievements. All interviews were taped and later transcribed. Immediately following the interview the participant was asked to complete the 12-item version of the General Health Questionnaire (Goldberg and Williams, 1988), Internal-External Locus of Control Scale (Rotter, 1966), the Life Satisfaction Index Z (Wood, Wyle and Sheafer, 1969) and a Self-esteem Scale (Rosenberg, 1965). These questionnaires were completed in 15 to 20 minutes. The 38 people who took part in the second study are referred to in this study as high achievers.
After all the interviews had been completed, two focus group meetings were held, one in London and one in Leeds, each attended by two of the researchers and all the available participants in the study. The focus group meetings lasted for approximately three hours and were tape-recorded and transcribed.

In the third phase of the study a further group of subjects were contacted and matched as closely as possible to the high achievers study group, except that they did not meet the educational threshold criteria. The outcomes for these people were strikingly different although their pre-care and care experiences were similar.



How typical were the high achievers of children who enter and stay in the care system? It might be that because of the method of recruitment they represent a skewed selection. There may be many successful ex-care people in the population who have put their care experience behind them, perhaps wanting nothing to do with social services. Indeed we know that many see their care background as stigmatizing and avoid revealing it to friends or colleagues.

However, the background of those interviewed for the high achievers study conforms closely to what we know of the British care population generally. 55.3 percent categorized themselves as white British, 7.9 percent black (African-Caribbean), and 36 percent were of Asian, Chinese or mixed parentage. Of those who knew the reason why they had come into care, 28.9 percent gave physical or sexual abuse as the principal cause. This is almost certainly an underestimate.
Most of their parents had no educational or occupational qualifications (where this was known). One provisional hypothesis, that more successful people might come from a "sunken middle class" background, like children from working class homes who, in the days before comprehensive education, were selected to attend grammar schools (Jackson and Marsden, 1962), was clearly disproved. Only a handful of cases could have been seen as falling into that category, for example, a boy whose father was killed in a car crash shortly after his mother had died of cancer and a young woman whose father was a professional engineer working in the Middle East. Whatever protective processes might have contributed to these people's resilience, the social class of their birth parents was not a significant factor.



Comparing outcomes
For the third phase of the study a total of 22 subjects were contacted, ten males and 12 females, with a mean age of 25 years. They were initially categorized on the basis of obtaining a maximum of three GSCEs at grade C or above, i.e. below the criterion for the Successful in Care study. However, only two subjects had obtained as many as three GSCEs and one had two GSCE passes. The remaining 19 subjects had no qualifications of any type. Of those who knew the reasons for being admitted to care, 29.4 percent gave physical or sexual abuse as the main cause, compared with 28.9 percent among the successful group. Twenty percent of the men and 66.6 percent of the women entered the care system at age 12 or older. Their care experience was similar to that of the study group subjects, except that the genders were reversed, with males having more placements on average than females (5.7 and 4.4, respectively). 63.6 percent of this sample characterized themselves as white British, nine percent as black (African-Caribbean) and 27.2 percent as Asian, Chinese or of mixed parentage.


  Profile of Sample Characteristics High Achievers
Comparison Group
Sex 12 males 26 females 10 males 12 females
Age M = 26 years M = 25 years

White British

55.3% 63.6%

Black (African-Caribbean)

7.9% 9%

Asian, Chinese or Mixed Parentage

36% 27.2%
Qualifications Obtained    

A level or diploma


no qualifications


first degree


3 GSCEs or less


higher degree

Admitted to Care Due to
Sexual or Physical Abuse
28.9% 29.4%
Entered Care at Age 12 20.7% males,
Over 50% females
20% males,
66.6% females
Number of Placements M = 3.0 males,
M = 5.5 females
M = 5.7 males,
M = 4.4 females


In contrast to the high achievers, among whom a third had learned to read at the age of four, very few read early, most not until well into school and one aged 11 years. This finding suggests that learning to read early and fluently is one of the protective factors associated with later educational success. Early reading cannot be achieved without access to books, and this was a factor that clearly distinguished between the study and comparison groups. In the high achieving group, 34 people (89.5 percent) had regular access to books compared with only nine (40.9 percent) in the comparison group (chi-square = 13.88, p<0.0002).

However, there was no significant difference in the ownership of books, which implies that high achievers made more use of libraries. Nearly a third joined the local library on their own initiative, compared with only two in the comparison group, suggesting that self-motivation was an important factor.

Two-way between subjects analysis of variation revealed a significant difference between the high achievers and the comparison group in the level of encouragement by a parent, caregiver or significant adult to go into further education. Indeed the high achievers were given more encouragement of all kinds than the comparison group. Parents or caregiver of high achievers were significantly more likely to attend school events and show an interest in the child's educational progress than those in the less fortunate comparison group. Thirty-one (81.6 percent) of the high achievers were also given career advice (although they generally rated it unfavorably) compared with only seven (31.8 percent) of the comparison group (Chi-square = 12.94, p<0.0003).

The comparison group was significantly more likely to be out of school for a term or more (Chi-square = 8.37, p<0.04), whether by their own choice or because they had been excluded. They were also far more likely to have been suspended or excluded (63.6 percent) than the high achievers (23.7 percent). It is not surprising therefore that all but one of the comparison group left school at the age of 16 or earlier, whereas 21 (55.3 percent) of the high achievers stayed on after the age of 15 (Chi-square = 16.25, p<0.006).

Two-way between subjects analyses of variance was employed to examine the differences between the high achievers and the comparison group on measures of locus of control, life satisfaction, self-esteem and psychological well-being. The high achievers were significantly more internal in their locus of control than the comparison group. In other words, they were more likely to feel that they had control over their environment and their life course rather than being blown about by chance. In harmony with other research in this area (Strickland and Haley, 1980; Parkes, 1988), a significant main effect showed that the males were more internal in their locus of control (M = 11.46) than the females (M = 13.97). TABLE 2
Summary of the statistical analyses of the four questionnaire
measures of psychological well-being
Measure High Achievers Comparison Group Main Findings
Locus of Control M = 12.42 M = 14.14 High achievers are
[The lower the score more internal
the more internal] (p<.018).
General Health M = 12.16 M = 15.82 High achievers are in
(GHQ) better mental health
[high scores equal (p<.034).
poorer mental health]
Life Satisfaction M = 17.53 M = 12.09 Greater life satisfaction
in the high achievers
Self-esteem M = 31.62 M = 29.09 No significant difference. In contrast to the comparison group, the high achievers showed lower scores on the General Health Questionnaire, which suggests better mental health in this group. They also gave significantly higher scores on the life satisfaction scale: F(1.56)=17.74, p<0.001. However, although the high achievers had slightly higher self-esteem scores than the comparison group, this difference failed to reach significance: F(1.54)=3.94, N.S. It is interesting to observe that our findings suggest that relatively low self-esteem does not necessarily preclude internal locus of control or a sense of self-efficacy, a point which has not yet emerged from other studies of resilience. Taken together, these findings indicate that regardless of self-esteem the high achievers were strongly internally motivated to do well and were more mentally resilient than the comparison group. We cannot tell if they brought these characteristics with them when they entered the care system or if, at least in some cases, they resulted from contact with an influential caregiver or other adult who took an interest in them.

What is quite clear is that the post-care experience of the comparison group was very different from that of our study group, conforming much more closely in terms of outcomes to what we know of care-leavers in general. Most were unemployed and about a quarter of them were currently homeless or serving custodial sentences. Just under half of the females were single mothers. The vast majority of the sample did not own their own home and were unable to rent privately, probably because most renting agencies make it a policy not to accept either unemployed persons, single mothers or people with a criminal record. They were therefore entirely dependent on social housing, which usually consisted of a flat in a high-rise block or other unpopular accommodation.



Differences in the Current Post-Care Lifestyles between
the High Achievers and the Comparison Group

High Achievers Comparison Group
N % N %
Unemployed 1 2.6 16 72.7
Single mothers 1 3.8 5 41.7
Serving custodial sentences 0 - 4 18.2
Rented private or own home 28 73.7 3 13.6
Rented council or lodgings 9 23.7 14 63.6
Homeless 1 2.6 5 22.7

This small comparative study provides strong evidence that for this extremely vulnerable group of children, reaching at least average levels of educational achievement is a vital protective factor. How then was this educational resilience achieved, against the odds, by the participants in our main study?



All the participants in the study were keenly aware that their lives might have taken another direction, and in their accounts obstacles and difficulties that we might characterize as risk factors were at least prominent as protective ones. It is useful to look at these as the other side of the coin; reduction of risk is as important as promotion of protection.

From an ecological risk and resilience perspective (Fraser, 1997), risk factors for educational failure among this group of people are very obvious. As already noted, nearly a third had suffered physical, sexual or emotional abuse and/or neglect before coming into care. Many commented on the negative effects of this on their self-esteem, which has been specifically identified as a potential protective factor (Werner, 1992). It is interesting to note that, despite their relative success and their scores falling nearer the internal end of the locus of control scale, indicating resiliency, scores on the self-esteem scale were still generally low. It seems that locus of control, or belief in one's own effectiveness can exist independently of self-esteem and may protect children by promoting adaptive behavior in the face of risk.

Other factors were also prominent in the pre-care lives of almost all subjects, for instance, conflict between parents often culminating in their being left in the care of a single mother; disability and ill health in parents, especially mental problems; poverty; inadequate housing; living in a poor neighborhood; isolation and racism. Once in care, risk factors for low educational achievement were a matter of common experience. For example, in residential care many subjects reported that no one took any interest in what went on at school, other than making sure you were properly dressed in the morning. "No one ever asked what you'd done in the day or said 'well done' if you got a good mark. It was 'get you uniform off and it's time for tea.'"
No facilities were provided for doing homework; there were seldom any books on the premises, few respondents could remember ever being read to by a member of the staff. It was difficult to find a quiet space to read or work. "The bedrooms were locked in the day and you weren't allowed upstairs. You could work in the dining room after tea but it smelled of food and made me feel ill." Sarah remembered putting a board on her bed and trying to do homework in the dim light of a 40 watt bulb. Tracey found a plank to stretch between lockers in the cloakroom. It takes strong motivation to overcome practical obstacles like these.

Placement moves carried a high risk of disrupting school progress, especially when they occurred at critical times. Moves were often arranged with marked insensitivity to the rhythms of school life, within a week or two of the end of the term, or involving a change of school three or four weeks into the start of a new term, when they could equally well have occurred in the holidays - a point noted by Fletcher-Campbell (1990) and Bullock (1993). These seemingly small matters could make all the difference between a child's continuing to do well and initiating a rapid deterioration (an example of a risk factor becoming a risk process). In short, the accounts of these subjects fully confirmed the risk factors for school failure identified in earlier research (Colton and Heath, 1994).

The only factor, which by definition could not apply in the case of high achieving subjects, was low ability or lack of intelligence. Some said that they had always had a clear sense of their own ability, even if it was not recognized by their carers, but others had accepted the attribution of caregivers, teachers or peers, even if it was negative, and had carried the low self-esteem which this induced into adulthood, despite their achievements.

Bella came into care at 14 after many years of sexual abuse by her father: "I was useless, absolutely useless. I had no confidence at all. I couldn't do anything right. I was always messing about, attention-seeking and misbehaving. I just believed I was stupid. Kids do, don't they, when they are told they are."

The self-esteem of children in care is, at best, fragile, easily undermined by a single disparaging remark from a teacher or carer. This is an area that is explicitly addressed in the Looking After Children Assessment and Action Records.



The Influence of Birth Parents
On the whole, early and pre-care experience seemed to offer little protection against later adverse experiences in care. Some subjects had no memories at all of their early life. Most pre-care memories were unhappy, rarely including celebratory occasions, family outings or simply a sense of being loved or cared for, but there were exceptions, and some people showed an amazing capacity to forgive if not forget even serious and prolonged abuse.

A few had a warm relationship with one or both of their parents. It is perhaps an indication of this that nearly a third of the "successful" group remembered being read to at bedtime by a parent. By contrast, only three people reported being read to by a foster parent, even if they had come into care very young, and only two individuals had memories of bedtime reading by a residential worker.

Although no association was found between the occupational or educational level of the parents and the level of education achieved by the children, there was clear anecdotal evidence of a connection between parental interest in education and the importance given to it by the subject's original family, and the later motivation of the child. It seemed that this could operate as a sleeping factor, of which the child was less aware in middle adulthood, but which would then resurface in adolescence. Some subjects quoted verbatim remarks on the subject made to them by their father or mother ten or more years earlier. Others commented that a powerful factor in encouraging them to struggle against all the obstacles they faced was the thought of how proud their parents would be if they succeeded. It is interesting to note that this group included a disproportionate number with ethnic or minority backgrounds.

Wanda was in foster care from the age of two when her mother left. Her father, a railwayman, "couldn't cope," but she continued to see him regularly until she was 15.

"He thought education was incredibly important. It's the thing that sticks in my mind. Every game that we played had an educational element to it. On Sunday morning we'd get the paper and look at the weather forecasts for different countries and I'd get out my felt pen and put down the temperature in Germany and Denmark. We used to have math quizzes. He taught me that learning was fun and it still is for me."

Sarah, now a social worker, gave a moving account of the great happiness she felt when she was able to invite her father to attend her degree ceremony, after a separation of ten years: "I cried, he cried, we all cried."

The importance of parental interest and belief in the value of education is a commonplace of educational research (Taylor, 1991, Osborn, 1990), but what is striking from the high achievers study is that it remains so potent despite distance in time and space. Janet told us that the motivating factor in her case was the thought that her mother, who died when she was 14 and whom she had not seen since she was five years old, would have been proud of her for going to university.

This finding has important implications for practice and gives strong support to the policy of maintaining contact with both parents even when there is no realistic prospect of reunification. There is also a job for social workers in convincing parents, who do not already have that perception, of the contribution they can make by valuing education and recognizing their children's achievements.



In earlier publications I have suggested that the low priority given to education by social workers, and their ignorance of the education system, its imperatives, rhythms and time scales, is a major factor in looked after children's low achievement (Jackson, 1989, 1994). As mentioned above, it was hoped that a study of more successful ex-care people would bring to light examples of the positive role that social workers can play and produce evidence of good practice. This was only true in a limited number of cases. All respondents were asked, "What part did your social worker play in your progress in education and planning for higher education?" To this, the reply of 92 percent of respondents to the "successful in care" questionnaire was "none." Even allowing for faulty memories (how many children are aware of a fraction of what their parents do to promote their well-being?), this was discouraging.

Among the high achievers there was some evidence that social workers had made a significant contribution, most often by organizing financial help for a young person to continue their education after school, or occasionally by helping a child to get back into school after a period of exclusion or a change of placement. Direct help to overcome educational problems or provide enhanced opportunities was seldom mentioned, and most respondents claimed that social workers played no part at all in encouraging or supporting their education.

This is not to suggest that arranging financial help is a negligible contribution, and it must often have involved considerable time and effort for the social worker. Many local authorities still do not recognize their responsibility to provide at least as much support for young people in further or higher education as a concerned parent would do, and some of our subjects had to fight for every penny. Some authorities made no allowance at all for vacations, when most students can return to their family homes, and one young woman was reduced to sleeping in a railway station over the Christmas holiday. The experience of the people in our study appears to be typical. A comparison of provision by local authorities for financial support of careleavers in further and higher education actually showed a decline between 1994 and 1996 (Broad, 1998). Many local authorities in Britain pursue a policy of moving young people at 16 to "semi-independent" living, which could hardly be more disruptive at the time when they are taking vitally important public examinations and need maximum support to continue their education, either at school or in college (Action on Aftercare Consortium, 1996).

Paradoxically, a few young people in more generous authorities found themselves better off than their home-based peers. This was a positive advantage, considering that they usually had many other problems to contend with. It much reduced the risk of failure and dropout at the tertiary education level if, for example, they did not have to take evening jobs or work during every vacation. If poverty is a powerful risk factor, an adequate income is an equally powerful protective factor.

The contribution of the few social workers who did make energetic attempts to promote and support the child's educational opportunities was immensely valuable and clearly there is enormous scope for development in this area.

Among the protective factors that were identified as most strongly associated with later educational success were:

1.) stability and continuity
2.) early reading
3.) having a parent or carer who valued education and saw it as the route to a good life
4.) having friends outside care who did well at school
5.) developing out of school interests and hobbies (which also helped to increase social skills and bring them into contact with a wider range of non-care people
6.) meeting a significant adult who offered consistent support and encouragement and acted as a mentor and possibly a role model
7.) attending school regularly

The last point might seem too obvious to mention were it not for the overwhelming evidence that large numbers of "looked after" adolescents attend school very intermittently, if at all, and run a much higher risk than other children of being formally or informally excluded (Blyth and Milner, 1996, Fletcher-Campbell, 1997). In this respect our subjects were not at all typical, only one had been out of school for any length of time. Going to school is an almost essential factor in educational success. Nevertheless it is an indication of the extraordinary resilience displayed by some of our subjects that one young woman, who had almost no opportunity to attend a mainstream school, and an appalling range of other adverse factors in her life, still finally achieved a good Honours degree from a leading English university.

The identifiable factors in her case were a passionate love of reading and an English teacher in a college who took a special interest in her. Literacy is, of course, vitally important for all children, as the government's current campaign recognizes, but for children who enter the care system early it is particularly crucial because of the absence of consistent and attentive adults in their lives and the poor educational quality of their care environments (Jackson, 1989, 1994). Being able to read well makes them less dependent on adult attention, therefore they are less demanding and less likely to be seen as disruptive in school or at home. It also gives them access to other worlds and ways of life as well as a means of escape from the stresses of family conflict, group living or emotional demands which they cannot meet. In this way, a protective factor, early reading, becomes a protective process. Few of our subjects had clear memories of learning to read, probably an indication that they were early readers, but the majority had learned to read before they were six, and 13 at the early age of four. Black women were the exception, learning to read much later, some not until seven or eight years old.

Enjoyment of reading was frequently mentioned in interviews as was the struggle to obtain books. Over a third never owned a book during their time in care other than school texts, but 89.5 percent nevertheless obtained access to reading material either through foster carers or their own efforts, a much higher proportion than reported by Bald, Bean and Meegan (1995).
This is a significant finding since reading difficulties in older children can easily be mistaken for cognitive deficits or may manifest themselves as disruptive behavior in the classroom. Literacy is the essential foundation for educational success but is seldom discussed in child care literature. There is enormous scope for development in this area, and a number of small-scale projects give grounds for optimism. For example, the PRAISE reading project in Salford successfully trained residential social workers and foster parents to use reading strategies with children they looked after, leading to marked improvements in their progress (Menmuir, 1994). A project organized by the Who Cares? Trust enabled looked after children to choose and own books for themselves, with adult support and encouragement (Bald et al, 1995). Although this project was very successful, it did also reveal how few books were available to looked after children and how little opportunity is given for reading within the care system. "If you read, " commented a young black woman, who complained that her success in athletics evoked much more interest than her ten GSCEs, "you're considered an oddity."



Few of our subjects had had the benefit of a truly stable care experience. Only one man, now a university lecturer, had remained in the same residential care setting throughout his childhood. An African-Caribbean girl, placed in infancy with a foster mother of the same ethnic background (though from a different island) was one of very few people who had had an untroubled educational progress through primary and secondary school, GSCEs, A levels and university. Throughout her education, she remained in intermittent contact with her father, and her foster parents stayed in England to support her through higher education before carrying out their plan to retire to Barbados.

Such continuity was unusual among our subjects, the majority of whom had experienced multiple placements. The average number of moves was 3.0 for men and 5.5 for women. One woman had obtained access to her case file and recorder 36 moves. These were not all different placements but neither did they include, as might have been expected, periods of returning home to parents or relatives. All the subjects commented on how difficult this chronic instability made it for them to function well in school. However, it seems that while greater stability, or at least continuity of schooling, is a valuable protective factor, some movement between placements, if well managed, need not preclude educational success.



A third of the high achievers had first been accommodated at the age of 12 or over: 20.7 percent of the men and 50 percent of the women. The gender discrepancy is almost entirely accounted for by Asian girls in conflict with their families. In retrospect most respondents felt they already had clear educational aims by that stage, and often saw coming into care as a positive experience, at least compared with their previous life.

Helen was sexually and physically abused by her father over a period of five years after coming from Hong Kong to join her family in Wales. She has no doubt that care provided her with a means of escape from a grim future:

"If I hadn't gone into care I'd still be in my parents' take away, working for about £30 a week, having no life at all, no friends, staying in cleaning, doing the dishes...Care meant I got my freedom, and time to study, and people to talk to as well. It was more of a family than I ever had with my parents. If I was at home I know for a fact that I wouldn't have made it to college. Succeeding educationally was something I knew I had to do, not something I wanted to do, but something I had to do. And I did it."

Helen "didn't need anyone to tell me to go and study," but she was helped by the stability of her placement of three and a half years in one residential home and by her close relationship with one of the care workers, who offered her a home after she completed her law degree.
Most of our respondents reported a special relationship with at least one person who made time to talk with and listen to them. The presence of a positive adult role model in the child's life setting as well as the amount of time spent with that person has been found in other studies to be important in fostering resilience (Maluccio, Abamczyk and Thomlison, 1996).



Even though the fact of coming into care might provide relief from a stressful family situation, this did not mean that there were no difficulties. In fact, most of our subjects remarked on how difficult it was to get their educational aspirations recognized. Career advice was either absent or pitched at a very low level. Women who now hold higher degrees were advised to go for nurse or secretarial training. Catering was the career most often recommended to boys.
Residential workers often saw it as a more important part of their job to encourage good social relations and participation in the life of the home rather than facilitating the long hours of private study necessary for success in examinations. Individual care workers might be kind and supportive, but because they normally lacked education themselves, they often had no conception of what was involved in serious academic work. Lack of reference books and difficulty of access to libraries was another frequent problem for those in residential care.
Leaving care grant guidelines were geared to people setting up home, not, for example, going to university to study architecture, "so they could give you money for a bed but not for drawing equipment." A young Asian woman was two-thirds through A-level course before the social services department released the money for her to buy essential textbooks.

Contrary to the rather common social work belief that teenagers prefer residential to foster care because it is less likely to make emotional demands on them, many of our subjects commented appreciatively on the emotional support they received from foster carers. A few had been placed with highly educated foster carers who gave them intensive educational help and were able to provide informed advice on further and higher education. Interestingly, the beneficial educational effects did not seem to depend on the quality of the relationship. At least in retrospect, people were able to appreciate the value of the stimulation and discipline they received, even if at the time they had not been particularly happy or even liked the family where they were placed. The majority, however, developed an attachment to their foster carers, and had a long-term post-care relationship with them, ringing them up for advice and visiting on a regular basis. Foster parents often acted as advocates with social workers and managers, especially when it came to pressing for grants to enable young people to continue their education.



At the end of the twentieth century, in western societies, educational competence is a crucial factor in ensuring social inclusion. There is ample evidence of widespread educational failure and underachievement among children in the British care system. The study reported in this paper underlines the high risk that these children will grow up socially as well as educationally excluded and drop into the underclass. The first point to make then is that in both systemic and case levels, every possible effort needs to be made to compensate for preexisting risk factors and to initiate protective factors and processes. The priority necessary for education needs to be reflected at every point in the care system, for example, in the selection and training of residential workers and foster carers, in the design and regime of children's homes, in additional support and teaching resources for children who have fallen behind or missed school. Children's right to education, which should be guaranteed under the United Nation's Convention, should be far more vigorously asserted by social workers. All decisions about children's placements and living circumstances should give full weight to the impact on their education.

Education in school is only one aspect. Children's whole living environment should be designed to contribute to their educational and cognitive development. There should be generous support for post-school of all kinds, both financial and emotional, and this should not be age-limited. Return to education can be just one of those turning points that transform the life trajectory of vulnerable people who were overwhelmed by risk factors in childhood.


Action on Aftercare Consortium. (1996). Too Much Too Young: the failure of social policy in meeting the needs of careleavers. Ilford, Essex: Barnardo's.

Anthony, E.J., and Cohler, B.J. (1987). The Invulnerable Child. New York: Guilford.

Bald, A., Bean, J., & Meegan, F. (1995). A Book of My Own. London: Who Cares? Trust.

Bebbington, A. and Miles, J. (1989). The background of children who enter local authority care. British Journal of Social Work, 19(5), 349-68.

Biehal, N., Clayden, J., Stein, M., & Wade, J. (1992). Prepared for Living? A Survey of Young People Leaving the Care of Three Local Authorities. London: National Children's Bureau.

Blyth, E. and Milner, J. (1996). Exclusion form School: Interprofessional Issues for Policy and Practice. London: Routledge.

Broad, B. (1998). Young People Laving Care: Life after the Children Act 1989. London: Jessica Kingsley.

Bullock, R., Little, M., & Milham, S. (1993). Going Home. Aldershot: Dartmouth.

Colton, M. and Heath, A. (1994). Attainment and Beheviour of Children in Care and at Home. Oxford Review of Education, 20(3), 317-28.

Fergusson, D. and Lynskey. (1996). Adolescent Resiliency to Family Adversity. Journal of Child Psychology and Psychiatry, 37(3), 281-292.

Fletcher-Campbell and Hall. (1990). Changing Schools, Changing People. Slough: NFER.

Fraser, M. (Ed.) (1997). Risk and Resilience in Childhood. Washington: National Association of Social Workers.

Garnett, L. (1992). Leaving Care. London: National Children's Bureau.

Garnett, L. (1995). The Educational Attainments and Destinations of Young People Looked After by Humberside County Council. Hull: Departments of Education and Social Services, Humberside County Council.

Goldberg, D. P., and Williams, P. (1988). A user's guide to the General Health Questionnaire. Windsor: NFER-Nelson.

Jackson, B. and Marsden, D. (1962). Education and the Working Class. London: Routledge and Kegal Paul.

Jackson, S. (1987). The Education of Children in Care. Bristol: University of Bristol Papers in Applied Social Studies No. 1.

Jackson, S. (1989). Residential Care and Education. Children and Society, 2(4).

Jackson, S. (1998). Lookinf After Children: a new approach, or just an exercise in formfilling? British Journal of Social Work, 28(1).

Menmuir, R. (1994). Involving Residential Social Workers and Foster Carers in Reading with Young People in their Care; the PRAISE reading project. Oxford Review of Education, 20(3), 329-338.

Osborn, A. and St. Claire, L. (1987). The ability and behaviour of children who have been in care or separated from their parents. Early Childhood Development and care, 28(3), 187-354.

Osborn, A. (1990). Resilient children: a longitudinal study of high-achieving socially disadvantaged children. Early Child Development and care, 62, 23-47.

Parker, R., Ward, H., Jackson, S., Aldgate, J. & Wedge, P. (1991). Assessing Outcomes in Child Care: the report of an independent working party established by the Department of Health. (Vol. 1). London: HMSO.

Parkes, K. R. (1985). Dimensionality of Rotter's locus of control scale: An application of the "very simple structure" technique. Personality & Individual Differences, 85(6), 115-9.

Pilling, D. (1990). Escape from Disadvantage. London: National Children's Bureau.

Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80, 1-28.

Russell, F. (February 1998). Lost Inside - the imprisonment of teenage girls. Criminal Justice, 16, 4-5.

Rutter, M. and Quinton, D. (1977). Psychiatric disorder - ecological factors and concepts of causation. In McGurk (Ed.). Ecological Factors in Human Development. Amsterdam: North-Holland.

Rutter, M. (1985). Resilience in the face of adversity: Protective factors and resistance to psychiatric disorder. British Journal of Psychiatry, 147, 598-611.

Rutter, M. (1996). Transitions and turning points in developmental psychopathology as applied to the age span between childhood and mid-adulthood. International Journal of Behavioral Development, 19, 603-626.

Rosenberg, M. (1965). Society and the Adolescent Self-Image. Princeton, NJ: Princeton University Press. Social Services Inspectorate and Office for Standards in Education. (1997).

he Education of Children who are Looked After by Local Authorities. London: Department of Health.

Stein, M. (1994). Leaving Care: Education and Career Trajectories. Oxford Review of Education, 20(3), 349-60.

Strickland, B. R. and Haley, W. E. (1980). Sex differences on the Rotter I-E Scale. Journal of Personaity & Social Psychology, 39, 930-39.

Taylor, A. (1991). Social competence and the early school transition. Education and Urban Society, 24, 15-26s.

Triseliotis, J., Shireman, J. & Hundleby, M. (1997). Adoption Theory, Policy and Practice. London: Cassell.

Ward, H. (Ed.). (1995). Looking After Children: Research into Practice. London.

Werner, E. (1992). The Children of Kauai: Resiliency and Recovery in Adolescence and Adulthood. Journal of Adolescent Health, 13, 262-268.

Wood, V., Wylie, M. L. and Sheafor, B. (Eds.). (1969). An Analysis of a short self-report measure of life satisfaction: Correlation with Rater Judgements. Journal of Gerontology, 24, 465-469.

Back to Top

Copyright for the I.U.C. Journal of Social Work Theory and Practice is owned by the Social Work Program, Department of Social Relations and Services, Bemidji State University, Bemidji, Minnesota, USA. One copy may be made (printed) for personal use; teachers may make multiple copies for student use if the copies are made available to students without charge. Permission must be secured from the editors for sale of any copies of articles or for any commercial use of the material published in the Journal.

2001 Copyright BSU/IUC Journal of Social Work Theory & Practice