Can we plan services for children in foster care? Or do we just have to cope with what comes through the door?

Can we plan services for children in foster care? Or do we just have to cope with what comes through the door?

Roger Bullock, Dartington Social Research Unit, Totnes, UK

To what extent is it possible to plan ahead for children coming into care? How can we ensure that there are enough foster homes available for new entrants and that the range of facilities is sufficiently wide to meet all of their different needs? There are several possible approaches. We can wait and see what comes through the door and try to respond as best we can. Or, we can construct a range of services and try to fit the children into them. These strategies might be enough to contain the problem but will almost certainly lead to frustration, dissatisfaction and instability. So what else can be done? The Dartington Social Research Unit in England has developed a methodology to achieve a better match between the foster care that is provided and the needs of the children. It is called Matching Needs and Services (MNS).

1           Stage one – Selecting the study sample

The MNS method requires the selection of a large and clearly defined study sample. This should be appropriate to the issues under consideration – creating a better match between the needs of children and the services they receive. It can, however, be applied to any population, not just children coming into care (or for that matter children).

It comprises six stages which, if followed in sequence, should produce a more balanced and effective service. For example, in children’s services it can be used to plan provision for those entering care, those in care at any one time or specific groups such as adolescents in residential homes. For the purposes of this illustrative paper the focus will be children coming into care.

Initially, two facts highlighted by research into children in care have to be acknowledged. The first is that the term ‘in care’ is an administrative category, not a clinical one; so anything that is said about the children involved will be true for some but not for others. Their needs vary widely – some are infants, others are adolescents; some have been abused and neglected, others are casualties of family breakdown; some present challenging behaviour, others do not. Clearly, no ‘one size fits all’ in terms of the service responses that have to be made.

Secondly, if the interest is in foster care, it is important to appreciate that this can take many forms: respite, short-term, long-stay, permanent, pre-adoption, therapeutic, single parent, gay and lesbian or specialist with regard to children’s ages, ethnicity, behaviour or disabilities. In addition, some foster homes will take groups of siblings, some already have foster children living there and there may be birth family children resident.

The aim of the MNS exercise is to match these two situations so that the right sorts of placement are available to meet the range of needs presented by the population of children under scrutiny.

In this paper, the method will be illustrated using data on a sequential sample of the first 99 children coming into care from a selected date. This has the strength of being fully representative as there are no missing cases. The study was conducted in 2010 in an English industrial town with a population of 300,000 and about 200 children entering and leaving care each year.

Stage two Charting the needs of the children

Having selected the sample, the second part of the exercise is to chart the needs of the 99 children. This requires collecting information from files or from questionnaires completed by professionals who know the child and family. The emphasis is on the needs of the child, i.e. what does this child need?, and on the services necessary to meet them. However, some relevant background characteristics, such as children’s ages and gender, also have to be recorded, along with the needs of birth parents, in order to understand the wider context of the admission and the likelihood of the birth parents being able to resume care of their child or be part of his or her life. For this purpose, it is necessary to structure the data collected. MNS does this by assessing the needs in five different areas of the child’s and birth family’s life: living situation, family and social relationships, social and anti-social behaviour, physical and mental health, and education and employment.

Naturally, in a short paper it is not possible to present all of the data. However, the following selection of results should be sufficient to illustrate some of the needs and characteristics of the 99 children and their families.

Most of the details that emerged will be familiar to those working in welfare services but there can be some surprises. For example, in this study the figures of 51 for the children admitted in emergencies and 34 for entry with a sibling were higher than people thought.

Table one – The situation on entry to care

(Figures are the numerical numbers out of the 99 admissions)

CHARACTERISTICS AND HISTORY

 

 

SITUATION ON ENTRY

 

Gender (male)

54

 

Single parent family

42

Ethnicity (white British)

77

 

Both parents at home

18

Legal status (compulsory)

45

 

Step parent

19

 

 

 

Only child

34

On protection register now

32

 

Four or more siblings

21

On protection register in past

21

 

 

 

Previously in care

19

 

Overcrowded home

11

Past service involvement

88

 

 

 

 

 

 

Frequent movement

14

Placed with sibling

34

 

 

 

 

 

 

Family new to area

11

Emergency admission

51

 

Low income

57

 

 

 

 

 

Age on admission

 

 

Socially isolated

28

0-1

27

 

 

 

2-3

12

 

Poor relations child/mother

46

4-5

9

 

Poor relations child/father

50

6-11

23

 

Poor relations child/siblings

18

12-12+

28

 

 

 

 

 

 

Child recently harmed

58

Who referred case?

 

 

-by mother

48

Child

2

 

-by father

27

Parent/relative

21

 

-by others

8

Social worker (in-house)

13

 

 

 

Education

5

 

Nature of harm – sexual

8

Health

27

 

-emotional

18

Police

23

 

-physical

20

Other local authority

4

 

Poor parenting

50

Voluntary NGO organisation

1

 

Neglect

39

Other

3

 

 

 

 

 

 

 

 

ADULTS

 

 

FAMILY HISTORY

 

Violent at home

37

 

Family discord

72

Harmful sexual behaviour

16

 

Breakdown/divorce

47

Convicted of serious offence

21

 

Domestic violence

41

Learning disability

12

 

No significant adult for child

10

Alcohol abuse

24

 

 

 

Drug abuse

12

 

Poor relations with services

12

Depressed

33

 

 

 

Stress/inability to cope

69

 

 

 

Unhappy

39

 

 

 

Isolated

40

 

 

 

Pregnant

20

 

 

 

Unemployed

61

 

 

 

 

 

 

 

 

CHILD’S EDUCATION

 

 

CHILD

 

In mainstream school

41

 

Aggressive at home

26

Full time special school

4

 

Aggressive at school

19

 

 

 

Sexual behaviour problems

9

Permanently excluded

6

 

Lack of social network

13

Temporarily excluded

7

 

 

 

Underachiever

15

 

Poor behaviour at home

33

Poor relations with teachers

13

 

Poor behaviour at school

26

Bullies other children

11

 

Poor behaviour - community

11

 

 

 

Poor peer relationships

17

 

 

 

 

 

Isolated

15

 

Mental health problem

4

Poor attendance

15

 

Physical health problem

1

 

 

 

Learning disability

7

Above average ability

2

 

Alcohol misuse

5

Hard working

12

 

Drug misuse

6

Likes school

20

 

Unhappy

28

Liked by pupils

20

 

Developmental delay

19

Liked by teachers

25

 

Isolated

19

 

 

 

Pregnant

3

 

 

 

 

 

CHILD’S SOCIAL SKILLS

 

 

FIRST PLAN FOR CHILD

 

Pleasant to be with

42

 

Return home quickly

23

Wants to change behaviour

15

 

Live with relatives

6

Confidence to solve problems

8

 

Long-stay substitute care

29

Social skills with people

17

 

Permanent placement

13

 

 

 

Independent Living

2

 

 

 

Other/unclear

26

 

Stage three Identifying groups of children with similar needs

The next stage is to disaggregate the mass of information. One method is to identify groups of children with similar needs which, in the MNS exercise, are labelled ‘Needs Groups’. Statistical cluster analysis can be used for this purpose as this shows which factors do and do not go together and indicates the number of children with combinations of needs. It is similar to some other widely used taxonomies, such as the ICD-10 WHO Classification of Diseases used in medicine where causes, symptoms and aetiology are outlined for each item. But MNS is more flexible in that the number and distinguishing features of the needs group emerging can vary, reflecting the nature of the sample and so are not the same in every case. Moreover, in child welfare, grouping is more complicated because of the plethora of relevant factors, their multiple effects and the fact that many carry a ‘meaning’ for the individual that can affect their impact.

Any number of groups could emerge form the analysis so a balance has to be found between establishing general patterns of need without being overwhelmed by individual differences.

When this was done for the 99 children in the study, four discrete needs groups emerged.

Group One

Young children in families where the need is to alleviate stress and poor living situations in order to improve parenting

Group Two

Children for whom the need is to reduce the risks and consequences of chronic family discord and violence

Group Three

Older children whose needs require help to improve their behaviour in a range of areas and contexts

Group Four

Children who need protection from risks of harm and help to recover from the effects of serious abuse and neglect

Stage four – Looking at the needs of children in each group

The next stage is to look at the distribution of background characteristics and needs across the four groups. In Table 1, the figures were those for the whole study population of 99 admissions. But in order to allow comparisons across the four groups, the figures in Table 2 are percentages out of the total for each group for the same factors as before.

Table 2 – The distribution of children’s needs and characteristics across the four needs groups

 

 

 

 

 

 

 

NEED GROUPS

1

2

3

4

 

ALL

TOTAL CASES IN EACH GROUP

31

33

14

21

 

99

 

Figures below are all percentages

 

Gender (Male)

45

58

64

52

 

54

Ethnicity (White UK)

74

76

71

86

 

77

Legal status (care order)

29

55

21

71

 

45

 

 

 

 

 

 

 

On Protection Register now

39

36

7

33

 

32

On Protection Register in past

10

15

36

40

 

21

Previously in care

19

18

21

19

 

19

Previous service involvement

90

79

86

100

 

88

Placed with sibling

26

45

7

48

 

34

Emergency admission

52

39

57

62

 

51

 

 

 

 

 

 

 

AGE ON ADMISSION

 

 

 

 

 

 

0-1 years

39

30

0

24

 

27

2-3 years

17

15

0

10

 

12

4-5 years

3

15

0

15

 

9

6-11 years

12

30

21

30

 

23

12-12+ years

29

9

78

24

 

28

 

 

 

 

 

 

 

WHO MADE REFERRAL

 

 

 

 

 

 

Child

3

0

7

0

 

2

Parent/relative

28

9

50

14

 

21

Social services (in-house)

24

12

7

5

 

13

Education

7

6

7

0

 

5

Health

31

27

0

43

 

27

Police

3

33

29

33

 

23

Other local authority

0

12

0

0

 

4

Voluntary NGO organisation

3

0

0

0

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SITUATION ON ENTRY

 

 

 

 

 

 

Single parent family

36

55

50

29

 

42

Both birth parents at home

16

18

7

29

 

18

Step parent

23

21

29

5

 

19

Only child

48

30

21

29

 

34

3+ siblings

32

27

7

5

 

21

 

 

 

 

 

 

 

Overcrowded home

26

0

14

5

 

11

Frequent movement

29

0

7

19

 

14

Family new to area

16

12

0

10

 

11

Low income

68

67

21

48

 

57

Socially isolated

39

3

0

71

 

28

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Poor relations with mother

13

35

92

81

 

46

Poor relations with father

13

52

89

71

 

50

Poor relations with siblings

8

15

60

7

 

18

 

 

 

 

 

 

 

Child recently ill-treated

42

79

29

67

 

58

- by mother

32

61

29

67

 

48

- by father

26

30

0

38

 

27

- by others

6

6

7

14

 

8

 

 

 

 

 

 

 

Nature of harm: sexual

0

6

0

29

 

8

Emotional

10

12

14

43

 

18

Physical

29

18

14

14

 

20

Poor parenting

39

64

21

62

 

50

Neglect

32

49

7

57

 

39

 

 

 

 

 

 

 

Family discord

58

88

100

48

 

72

Breakdown/divorce

39

58

64

29

 

47

Domestic violence

26

76

29

19

 

41

No significant adult

7

0

29

19

 

10

 

 

 

 

 

 

 

Poor relationship with services

0

3

57

14

 

12

 

 

 

 

 

 

 

CHILD

 

 

 

 

 

 

Aggressive at home

10

21

93

14

 

26

Aggressive at school

6

9

86

10

 

19

Sexual behaviour problems

3

3

36

10

 

9

Lack of social network

3

3

43

14

 

13

Behaviour problems at home

16

27

93

29

 

33

- at school

6

21

79

29

 

26

- in community

0

3

64

5

 

11

 

 

 

 

 

 

 

Learning difficulty

6

3

7

14

 

7

Alcohol misuse

0

0

36

0

 

5

Drug misuse

0

0

36

5

 

6

Unhappy

13

21

71

33

 

28

Developmental delay

23

21

7

19

 

19

Isolated

3

18

43

29

 

19

Pregnant

0

9

0

0

 

3

 

 

 

 

 

 

 

Pleasant to be with

55

18

14

81

 

42

Wants to change behaviour

16

3

29

24

 

15

Confidence to solve problems

16

0

7

10

 

8

Social skills with people

16

6

7

43

 

17

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CHILD'S EDUCATION

 

 

 

 

 

 

Full-time mainstream school

32

39

71

38

 

41

Full-time special school

3

0

14

5

 

4

 

 

 

 

 

 

 

Permanently excluded

0

0

43

0

 

6

Temporarily excluded

3

3

29

5

 

7

Underachiever in education

0

0

64

29

 

15

Poor relations with teachers

3

3

79

0

 

13

Bullies other children

0

3

64

5

 

11

Isolated

6

18

36

10

 

15

Poor attendance

3

3

79

10

 

15

 

 

 

 

 

 

 

Above average ability

3

0

0

5

 

2

Hard working

16

0

0

33

 

12

Likes school

32

9

0

33

 

20

Liked by pupils

23

12

0

43

 

20

Liked by teachers

26

15

7

52

 

25

 

 

 

 

 

 

 

ADULTS

 

 

 

 

 

 

Violent at home

26

55

29

33

 

37

Harmful sexual behaviour

13

18

0

29

 

16

Convicted of serious offence

26

27

0

19

 

21

Learning disability

6

18

0

19

 

12

Alcohol abuse

10

33

14

38

 

24

Drug abuse

16

9

0

19

 

12

Depressed

9

36

29

67

 

33

Stress/inability to cope

68

76

64

62

 

69

Unhappy

16

58

50

38

 

39

Isolated

45

30

29

57

 

40

Pregnant

26

30

14

0

 

20

 

 

 

 

 

 

 

FIRST PLAN FOR CHILD

 

 

 

 

 

 

Return home quickly

39

21

29

0

 

23

Live with relatives

6

0

0

19

 

6

Long-stay substitute care

13

45

43

19

 

29

Permanent placement

10

3

0

43

 

13

Independent living

3

0

0

5

 

2

Other/unclear

29

31

28

14

 

26

 

2           Stage five – Compiling a profile of the needs of children in each group

Stage Five uses the data obtained in Stage Four to compile a profile of the needs of children in each of the four needs groups identified earlier. When this is done, the following patterns emerge.

2.1     Group One

Young children in families where the need is to alleviate stress and poor living situations in order to improve parenting

Nearly one third (31%) of the 99 children scrutinised fell into this group where their families were under stress and living in poor circumstances that made it difficult for them to cope. There were problems associated with poverty (68%[1] had low incomes), 26% lived in crowded accommodation, 29% had moved frequently and for 26% pregnancy exacerbated an already difficult situation.

Just over half (55%) of the children were girls and most (71%) were accommodated under voluntary arrangements. Thirty nine per cent were under the age of two on admission to care and the referrals came equally from social services, health and parents. Nearly half of the children (48%) were the only children in the family.

Difficult family relationships and serious abuse were not salient issues compared with some other children, although more (29%) had been physically abused than for any other group.  Domestic violence and family breakdown were also relatively low compared with the rest of the study population.

Only one in ten of the children needed help with behaviour and, apart from convictions for a serious offence (26%), the parents displayed the lowest levels of any group for depression, alcohol and drug abuse and unhappiness. Given the children’s young ages, schooling problems were rarely a major issue.

2.2     Group Two

Children for whom the need is to reduce the risks and consequences of chronic family discord and violence

This was the largest group of children, comprising a third of the sample. The children have many needs in common with Group One but the important differences are the high levels of domestic violence and the need to address its effects on children’s development and behaviour. In addition, the children’s ages are more widely spread, siblings are more likely to be involved and care admission undertaken by out-of-hours duty teams following police referrals. The commitment by welfare agencies to the care of these children is, therefore, longer.

Over half of the children in this group (55%) were living in single parent families at the time of admission and there were frequently issues of low income (67%) rather than the quality of housing. Levels of harm were high mostly due to chronic neglect and poor parenting compounded by domestic violence (76% of cases). The children were more aggressive than those in Group One, echoing their violent carers by presenting more difficulties at home and school. Compared with the adults in Group One, parents were equally likely to have been convicted for a serious offence but displayed much higher levels of depression, alcohol abuse, learning difficulties, unhappiness and other mental health problems. More of the children in this group are placed with relatives and there is expectation of a longer-term commitment by the care authorities to their welfare.

2.3     Group Three

Older children whose needs require help to improve their behaviour in a range of areas and contexts

The third group, comprising 14 children, two-thirds of them boys, is easy to identify. They are older children with a variety of complex needs and presenting a range of serious problems.

Harm by parents is less of an issue for this group. It might have been in the past (36% had previously had their names placed on the child protection register) but only 29% of the children have recently been obviously harmed in any way. Levels of neglect are also lower than for the other children in the sample. Moreover, in the 50% of cases, it was the parents who first contacted welfare services for help.

Compared with the sample as a whole, the children are older on admission to care and accommodation (78% over 11). Siblings are rarely accommodated at the same time. As many as 50% come from single parent families and 29% from step-families, all coping with a range of difficult behaviours and disabilities. The young people score the highest of all the groups on every measure of aggressive behaviour, whether at home, school or in the community, and for a plethora of other difficulties, such as poor relations with peers and adults, alcohol and drug misuse and general unhappiness. Schooling is especially fraught with high levels of exclusion, academic underachievement, irregular attendance and poor relationships with teachers and pupils; indeed, 14% attend full-time special schools.

The children’s parents, in contrast, have relatively few problems other than the stress and unhappiness brought about by the poor relationships between them and their offspring and their children’s seeming intractability. They seem relatively well off and integrated into and supported by their local communities. Family discord, breakdown and violence are relatively uncommon.

These difficult adolescents are familiar to child care workers throughout the world. They are the least likely group to accept services and are judged by professionals to display the worst outcomes. The proportion in the sample, however, is only 14%, of all admissions. This may reflect alternatives to care, policies on the use of special schools and the contribution of youth justice teams. Nevertheless, in terms of resource, these young people remain a major consumer of services.

2.4     Group Four

Children who need protection from risks of harm and help to recover from the effects of serious abuse and neglect

As with Groups One and Two, the children in this group have needs for protection and better parenting but have relatively few other needs compared with their parents. These families have been a continuing concern to children’s services. Most (76%) referrals come from health and police and emergencies are common, with the admission of siblings occurring in nearly half of the cases. The age distribution of the children is relatively wide.

The families are socially isolated and parent-child relationships are almost as poor as for Group Three. Recent harm comprises not only neglect (57%) but also sexual (29%) and emotional abuse (43%). Parents show higher figures than the other groups for poor mental health, alcohol abuse, depression and isolation, as well as inappropriate sexual behaviour although levels of family breakdown and domestic violence are relatively low.

The outstanding feature of the children, in contrast, is the high score for many of the protective factors, especially those concerned with education. Most are hard working and much liked.

Permanency plans had been agreed for 43% of them. A quick return home is not perceived as an option in most cases as the parents are seen as unlikely to meet the children’s needs in the foreseeable future.

Having established patterns of need among the admissions to care, the focus now shifts to the services that will best meet them.

3           Stage six – Designing new services

To appreciate the strengths and weakness of existing provision and areas for improvement, it is helpful to ask the following question for each of the four needs groups, again emphasising what the children need, not just what they are like.

·         What are the needs of this group?

·         What do we wish to achieve for these children in five areas of their lives (living situation; family and social relationships; social and anti-social behaviour; physical and mental health; education and employment)?

·         What services does research show best achieve these aims?

·         What thresholds need to be put in place to help professionals identify suitable children, to make sure that the right children get the right services and to ensure consistent responses to children and families presenting similar needs?

·         What support in terms of administrative structures, management and training need to be put in place to make the services work?

·         How does the envisaged service compare with what the children and families in the needs group already receive?

Having answered these questions, it is next necessary to ask ‘What does a MNS exercise offer that is new or different from other audits and research investigations?’

I would suggest the following:

·         Better information based on a structured analysis of the needs of children coming into care over a specified period (e.g. one year)

·         Useful for planning as patterns are unlikely to change very much in future years

·         A more accurate estimate of the numbers of children with particular needs

·         A better indication of the links between different needs

·         A method of designing services that puts children’s and families’ needs first and perceives administrative structures, training and management as facilitators rather than as ends in themselves. It produces services that are needs-led and evidence–based and so are likely to meet the needs of different groups of children more effectively

In practical terms, this can lead to many changes and improvement. Examples are:

·         A better balance between prevention-early intervention-treatment-social intervention

·         A history of previous service involvements (or lack of them) that highlights points at which earlier or different interventions might have prevented admissions to care 

·         Effective responses to acute and emergency situations

·         Greater understanding of the co-morbidity of needs

·         How children causing concern in one area, e.g. for abuse or neglect, have needs in other areas, e.g. health and education.

·         Groups that might be diverted from care

·         Stronger links between research evidence and service design

4           Examples of using MNS to change services

Three specific examples of applications of the MNS method in other local authorities in England are now described to illustrate these benefits and show how the information can be used to review different policy and practice issues. They are as follows.

4.1     Local Authority 1

In a sample of care admissions, 20 out of the 100 children studied were not recorded as having a child protection plan prior to the point at which they came into care, despite the fact that many of them were babies or very young children where there were allegations of non-accidental-injury or sexual abuse. This finding led service managers to inquire how many of these children had been in contact with other services in the previous 12 months and why they had suddenly come in to care.

In a needs group defined by family violence and child physical abuse, there were a significant number of cases where team managers wanted more information or thought that the case might have been dealt with in other ways.  As 20% of these children had been in care before, they reviewed the quality of the assessment and care planning and made changes to make the process more responsive to the children’s needs earlier on.

Another needs group defined by poor family relationships, child behaviour problems and parental mental health difficulties included many teenagers. Again, the MNS exercise led managers to check how many of these young people were known to the adolescent prevention service and how many had received its help in the 12 months prior to becoming looked after, and what type of help had been given. Half of this group was identified as suitable for possible diversion, but more information was required on quality of previous assessments and care planning.

4.2     Local Authority 2

In this local authority, there was concern about the rising number of children being referred for a child protection plan and that the length of time for which these were in force was increasing. A MNS exercise of all children entering care over the period of one year found that over half of them fell into this category at the time of admission. The fear was that a continued increase in the number of such children would lead to a sustained rise in the number of children entering care, putting a strain on foster care services. For reasons of quality of outcomes and cost, it was necessary to understand better what was happening and what could be done to prevent a flood of inappropriate care admissions.

4.3     Local Authority 3

In this local authority, senior managers and members of a steering group took part in a ‘diversion exercise’ that involved reading through all of the case summaries of 100 children entering care in sequence. They then reflected on whether: i) the child had to be admitted to care; ii) the child could have been diverted away from care; or iii) it was not possible to choose either way on the information available. The purpose of the exercise was to assess the proportions of children in each category

Of the 100 cases in the sample, those that were not found to have a significant impairment to their health and development were further considered by an independent consultant by way of validating the method. In instances where a clear view did not emerge, the cases were further discussed during later panel meetings.

Of the 100 cases, it was agreed that:

·         57 should have been in care

·         31 might have been diverted

·         5  a consensus could not be reached about the appropriate response

·         7  not enough information available to make a decision

Twenty eight of the 31 cases identified as possibly 'diversion' were in a needs group defined by adolescents displaying emotional and behavioural difficulties that had strained relationships with their parents.

This raised significant concerns about the effectiveness of current services for this group and the often difficult individual practice decision of whether admitting troubled and troublesome adolescents to care has any real long-term benefit.

5           Conclusion

It would be foolish to claim too much for this methodology; it will not resolve all the problems of providing substitute care. There are criticisms that it forces children into categories that are insensitive to individual differences and that it does not tell people what to do.

Neither should the exercise be seen as a one-off event. It should be part of a continuous monitoring and development process in an agency. Hence, it can be repeated regularly to monitor progress but it can also be used to facilitate more radical reviews of services. For example, if managers and practitioners are involved in the data collection, there is an immediate rapport between them and the researchers providing an auspicious context for subsequent discussions and implementation with regard to recommendations. Even more radical is to ask service users, i.e. foster parents, birth families and children, to take part or comment on the service changes suggested.

But whatever its weaknesses, using the MNS methodology is better than just waiting to see which children turn up in care and trying to fit them into a rigid service structure. It also indicates the numbers likely to be involved which facilitates planning as numbers are unlikely to vary much from year to year. With this information, the right sorts of foster parents can be recruited, trained and supported. Thus, the contribution of MNS to creating an effective service for separated children is likely to be significant.

References:

Matching Needs and Services: The Audit and Planning of Provision for Children Looked After by Local Authorities. Totnes: Dartington Social Research Unit

Author´s Address:
Roger Bullock
Dartington Research Unit
Totnes, England
roger.bullock@dartington.org.uk



[1] For the sake of simplicity, comparisons are made between the figure for the group and the figure for the whole sample. A more accurate comparison would be between the group and the rest of the 99 children in the sample. This would have the effects of increasing the contrasts. In addition, when comparisons are made, it is usually relative to other looked after children and not the national population, most of whom are much less disadvantaged.