Understanding the concept of resilience in relation to looked after children: A Delphi survey of perceptions from education,social care and foster care

Abstract

There has been a surge of interest regarding the application of resilience theory in childcare practice and how resilience can be promoted among vulnerable children, in particular, looked after children. However, little is known about how people working with looked after children understand the concept of resilience. This study aimed to explore how social workers, teachers and foster carers, working with looked after children, understand resilience and whether there is consensus as to what constitutes resilience. The study also sought to explore whether there are differences in how resilience is constructed across these groups. In total, 106 participants took part in a Delphi survey (34 teachers, 36 foster carers, 36 social workers). There was moderate consensus that resilience related to survival, coping and a sense of self-worth. Resilience was not considered a panacea but a concept that also had limitations. Participants understood resilience in ways that were both similar and different to existing conceptualisations within the literature. However, there were many aspects of resilience for which there was no consensus or significant difference of opinion between the participant groups. The need for further training and research is discussed, in order to support attempts to promote resilience among looked after children.

Keywords

Attachment Delphi looked after children resilience

‘Resilience’ is a broad concept pertaining to ‘relative resistance to psychosocial risk experiences’ (Rutter, 1999, p. 120). The concept of attachment is central to resilience theory and practice (Schofield, 2001). Attachment theory (Bowlby, 1969) provides a framework for understanding how early relationships with the primary caregiver impact a child’s social, emotional and cognitive development. Infants seek out attachment figures from whom they can experience feelings of safety and comfort, and they develop ‘internal working models’ (IWMs) of relationships based on these early attachment experiences. IWMs are said to influence subsequent mental representations of the self, others and relationships.

The protective nature of positive early relationships has been consistently highlighted in the literature (Werner, 2000; Wyman et al., 1999). Attachment relationships are considered important in the promotion of resilience (Yates & Masten, 2004). Indeed, it has been suggested that attachment and resilience theory are complementary, not separate bodies of knowledge (Atwool, 2006). However, although resilience is influenced by attachment theory, resilience theory moves beyond early attachment experiences (Gilligan, 2001) to consider the importance of relationships, such as peers and other significant adults (Dunn, 1993), and to other domains in which resilience may be fostered, such as education, talents and interests (Daniel & Wassell, 2002).

A group of children for whom the application of resilience theory is particularly pertinent is looked after children (Bostock, 2004; Gilligan, 2004). Although some looked after children do remarkably well despite their early adversity (Rees, 2013), research has consistently highlighted the poorer outcomes for children in substitute care, including the prevalence of mental health difficulties (Meltzer, Corbin, Gatwood, Goodman, & Ford, 2002), under-achievement and increased likelihood of exclusion from education (Department for Education and Skills (DfES), 2006a, 2006b) and subsequent unemployment (DfES, 2007). The relevance of resilience theory for working with looked after children is not only fuelled by the observation of poor outcomes for these children but also by the early adversity or experiences that brought children into care. The most common adversity is abuse or neglect by a primary caregiver which can have pervasive detrimental effects on a child’s social, emotional and cognitive development (Cicchetti, 2002; Stein, 2006) and the development of attachment relationships (Toth, Cicchetti, Macfid, Maughon, & Van Meenen, 2000).

The value of a resilience-led perspective for working with children in care has been recognised. For example, Daniel and Wassell (2002) designed a workbook outlining six domains of a child’s life in which resilience can be assessed and promoted. Gilligan (2001) developed a resource guide aimed at promoting resilience among children in care, emphasising the social aspects of resilience, the role adults can serve and the importance of positive school experiences and leisure activity. Despite increasing interest in applying resilience theory in practice, little is known as to whether there is a shared understanding of what resilience means among people working with looked after children. To date, only three studies have considered how resilience is understood or used in practice when working with vulnerable children. Daniel (2006) explored the utility of the concept of resilience as a guide to assessment and planning for eight social workers working with looked after children in Scotland. Social workers in the study were already familiar with the concept and considered it to be useful. In contrast, McMurray et al. (2008) reported that social workers found it difficult to conceptualise resilience and considered it an academic issue with no relevance to their practice. Moreover, all social workers deemed the child in their professional care to be resilient, which was reflected in a low frequency of reported mental health difficulties and of referrals to mental health services. However, their positive outlook was often incongruent with the child’s emotional well-being. Daniel, Vincent, Farrall and Arney (2009) considered practitioners’ understanding of resilience and how it was being used within organisations that aim to nurture resilience in vulnerable children in the United Kingdom and Australia. Practitioners were familiar with the concept and used definitions that were congruent with the literature. However, practitioners in the United Kingdom placed more emphasis on child and family domains, whereas in Australia greater priority was given to the family and the wider community. Despite recognising the utility of the concept, practitioners suggested that resilience meant different things to different individuals.

In summary, although a resilience-based approach may be useful in working with vulnerable children, there may be differences in the understanding and application of this concept. Furthermore, current research is limited to the views of social workers, and often those who have received training on resilience (Daniel, 2006) or work in organisations that explicitly seek to promote resilience (McMurray et al., 2008), limiting the generalisability of the findings. In addition to social workers, there are other people in the network around the child (Conway, 2009) whose work might promote resilience, such as teachers and foster carers. Given this, it is important to consider how the range of people working with looked after children understand resilience.

This study aimed to extend existing research by exploring how social workers, teachers and foster carers understand the concept of resilience with regard to looked after children and the extent to which there is consensus among the three professional¹ groups. Differences in professional understanding of resilience may impact their practice and, ultimately, decisions for the child. In view of the drive for inter-agency working for looked after children (Golding, Dent, Nissim & Stott, 2006; Iwaniec, 2006), it is important that research considers the perceptions of multiple professional groups. Although the role of teachers and foster carers in enhancing the resilience of looked after children has been recognised (Clarke & Clarke, 2001; Jackson & Martin, 1998), their voices have remained absent from the literature. Typically, social workers, teachers and foster carers are all highly involved in the care and development of looked after children. This study had two main aims:

To explore how social workers, teachers and foster carers understand the concept of resilience with regard to looked after children, and the degree of consensus as to what constitutes resilience.

To explore whether there are significant differences in how resilience is constructed and understood by social workers, teachers and foster carers.

Method

Design

Delphi methodology is a technique for surveying perceptions (Stone-Fish & Osborn, 1992) and structuring group opinion (Goodman, 1987) and is of particular value where little knowledge currently exists (Skulmoski, Hartman, & Krahn, 2007). A Delphi survey is an iterative, multi-stage process to collect and distil the opinion of a group of ‘experts’, with a view to establishing a consensus (Norcross, Hedges, & Prochaska, 2002). It involves designing and administering a series of questionnaires in two or more rounds, whereby feedback is given to participants in order to support consensus building (Linstone & Turoff, 1975).

The approach used in this study followed Powell (2003). In the first round, participants are asked unstructured open-ended questions about the topic. Qualitative analysis of the data generates statements that are used to construct the second-round questionnaire (R2Q). R2Q is sent to all participants or sometimes to a wider sample of participants. Each participant rates his or her level of agreement with each statement. In the final round, a third questionnaire (R3Q) is created for each participant, comprising the same statements, but also including both the individual’s R2Q response and the average rating across the participants for each statement from R2Q. In R3Q, participants are invited to consider their rating in light of the group’s response and revise their rating, if they so wish.

Participants

Participants should be purposively selected to meet four ‘expertise’ requirements: knowledge and experience of the issues under investigation, capacity and willingness to participate, time to participate and effective communication skills (Adler & Ziglio, 1996). The inclusion criterion was that participants were currently working with looked after children between the ages of 5 and 11 years. A specific age range was chosen because how people conceptualise resilience is likely to vary according to the child’s age (Newman, 2002).

In the first round, a small number of participants in each group were recruited. Additional participants were then recruited to complete R2Q and R3Q. Participants who met the inclusion criteria were recruited from local authorities in London and other unitary local authorities in England.

The number of participants in each round is displayed in Table 1. For round 1, 24 people were approached and 22 took part. In R2Q, 129 were approached and 106 took part. In R3Q, the 106 R2Q-completers were approached and 28 took part. Following other Delphi studies (e.g. Pipon-Young, Cupitt, & Callanan, 2010), if participants chose not to complete R3Q, then their responses from R2Q were used as final responses and included in R3Q. This gave a total sample size of 106, 90 (85%) of whom were female. Most (82%) were of White ethnic origin. The mean age was 47 years 2 months (standard deviation (SD) = 9 years 10 months). Mean length of time worked with looked after children was 14 years 1 month (SD = 10 years 3 months; range: 6 months to 41 years). For the foster carers, the mean number of years working with looked after children was 7.5 (SD 74.41).

Table 1.

Number of participants in each round by group.

Group	Round 1	Round 2	Round 3
Foster carers	9	36	6 (36)
Social workers	9	36	14 (36)
Teachers	4	34	8 (34)
Overall	22	106	28 (106)

Values in parentheses indicate the total number of participant responses included in the final analysis.

Measures

Round 1: interviews

An open-ended question asked how participants understood the concept of resilience in relation to looked after children.² Three prompt questions were asked if necessary.

Round 2: questionnaire (R2Q)

Round 1 data were transcribed and then analysed using Thematic Analysis (Braun & Clarke, 2006). Across all transcripts, aspects of the data that pertained to the first research aim were underlined and identified as ‘data extracts’. Conceptually similar data extracts were subsequently grouped and labelled as codes. The codes were then collated into overarching themes, which were named. Each data extract was thus placed within a code and then a theme. Some themes comprised numerous codes, while others had just one code and a single data extract. A data extract could also appear in more than one code.

The codes and data extracts were then used to form R2Q questionnaire statements. Where possible, the participants’ own words were used to phrase the statement. The number of statements was considered sufficient when each data extract and code could be related to at least one statement. The statements were subsequently refined to avoid overlap or duplication and presented in the themes in which they had originally been organised. In order to ensure reliability, the themes and statements were checked independently by the second and third author.

R2Q consisted of 58 statements, which were presented in the themes in which they had been categorised in the analysis. Participants were asked to rate the strength of their agreement with each statement on a 6-point Likert scale from 1 (strongly disagree) to 6 (strongly agree) and were invited to include comments. R2Q was piloted to ensure the wording was clear and to estimate completion time and tolerability. No changes were considered necessary.

Round 3: questionnaire (R3Q)

R3Q comprised the same items as R2Q. Each participant received an individualised version in which their R2Q response on each item was marked and the percentage of people selecting each response was shown. Furthermore, the most frequently selected rating (i.e. highest percentage) from R2Q was shown in bold for each item. A selection of participant comments from R2Q was also shown alongside each item. Participants were invited to review their rating for each item in light of the whole group’s scores and comments, and change their rating if they wished.

Procedure

Ethical approval was obtained from Canterbury Christ Church University. The Delphi survey process took 11 months to complete. Social workers and foster carers were recruited for the initial interview via a presentation given to children’s social workers and supervising social workers for foster carers in an outer London local authority. Supervising social workers invited foster carers to participate. Teachers were recruited by written invitation to all designated teachers for looked after children in the area. The interviews took place over 4 months and consisted of 10 interviews in person, 10 telephone interviews and two email responses. Additional participants for R2Q and R3Q were recruited through written invitations, presentations at foster carer support groups and meetings of professionals, and by circulating a message on the Clinical Psychology Looked after Children UK web forum.

Due to the large number of participants completing R2Q (n = 106) and the individualised nature of the third questionnaire (R3Q), R3Q was only created and sent to participants who explicitly agreed to continue to take part. Participants were asked to return R3Q with amendments or indicate whether there were no alterations to be made.

Results

Research aim one: conceptualising resilience with looked after children

The round 1 interview analysis yielded 58 statements pertaining to 13 resilience themes. The final data set comprised participants’ R3Q responses for each statement (or R2Q responses for participants who did not wish to complete R3Q). Only seven participants altered their responses between R2Q and R3Q.

This study used the definition of consensus employed by Graham and Milne (2003) who considered consensus as the amount and consistency of agreement between participants. Ratings were therefore collapsed from the 6-point scale into three categories as shown in Figure 1. The mid-range ratings were excluded because the research sought to identify statements that participants thought either were or were not aspects of resilience. Each statement was then classified according to the consensus categories in Table 2. Less than 50% endorsement was taken to indicate a lack of consensus at either end of the scale, and 50–100% endorsement was divided into tertiles such that 50–66.7% represented weak consensus, 66.8–83.3% moderate consensus and more than 83.3% high consensus to either include or exclude the statement from a definition of resilience. Four statements (6.9% of the data) would have reached consensus within the mildly agree–mildly disagree ratings if the mid-range data had not been excluded.

Figure 1.

Amalgamation of response categories.

Table 2.

Derivation of consensus categories based on Likert scale ratings and how statements were included or excluded from a definition of resilience.

	Sum of percentage of people who selected 1 and 2 (x)	Sum of percentage of people who selected 5 and 6 (y)
High consensus to exclude statement	83.3% < x
Moderate consensus to exclude statement	66.6% < x < 83.3%
Weak consensus to exclude statement	50% < x < 66.7%
No consensus	x ⩽ 50%	y ⩽ 50%
Weak consensus to include statement		50% < y < 66.7%
Moderate consensus to include statement		66.6% < y < 83.3%
High consensus to include statement		83.3% < y

The analysis revealed variation in the level of agreement of statements within themes. The statements are therefore displayed according to consensus category in Tables 3 and 4.

Table 3.

Statements with a moderate consensus to include in a definition of resilience.

Theme code	Theme	Statement	% strongly/moderately disagree (x)	% strongly/moderately agree (y)
(S2)	Survival	Resilience is keeping going and being able to carry on normal aspects of daily life despite difficult experiences	8.5	79.2
(S3)	Survival	Resilience is the child’s ability to ‘bounce back’ from a difficulty in a healthy way	10.4	70.8
(S4)	Survival	Resilience is what prevents a child from falling apart in the face of poor parenting or life experiences	4.7	69.8
(A3)	Ability	Resilience is having helpful coping strategies that children can use to cope with their difficulties	0	83
(A5)	Ability	Resilience is not an ability or characteristic that everyone naturally has; many children need help to become resilient	1.9	77.4
(A11)	Ability	Resilience is the child’s ability to adapt and manage big changes in his or her life, such as being separated from their parents and living with a new family	1.9	71.7
(Se3)	Self	Resilience is having a sense of self-worth that you matter	10.4	67
(N5)	Negative	Resilience has its limitations; a child should not simply be expected to survive everything that life presents himself or herself with	2.8	83

Table 4.

Statements with a weak consensus to include in a definition of resilience.

Theme code	Theme	Statement	% strongly/moderately disagree (x)	% strongly/moderately agree (y)
(S1)	Survival	Resilience is a survival instinct	17	60.4
(A4)	Ability	Resilience is the child’s ability to put a situation into context, in order to be able to move on from it	6.6	60.4
(A6)	Ability	Resilience can involve unhealthy or negative ways of coping	10.4	59.4
(A7)	Ability	Resilience is the child’s ability to cope with things that are difficult and learn from that	3.8	66
(A8)	Ability	Resilience is having something to draw on when things are difficult, such as your faith	5.7	56.6
(R3)	Responsibility	Resilience is a child being able to take responsibility for himself or herself and protect himself or herself	12.3	53.8
(Se1)	Self	Resilience is having good self-esteem/self-belief	11.3	61.3
(Se2)	Self	Resilience is the child’s ability to know himself or herself and their own mind	11.3	51.9
(Se4)	Self	Resilience is a child having confidence in himself or herself	9.4	64.2
(At1)	Attachment	In order to become resilient children need to feel safe and secure	8.5	60.4
(At2)	Attachment	Resilience involves the child’s ability to trust key adults in their life	9.4	65.1
(At3)	Attachment	Resilience involves the child’s ability to form a good relationship/attachment with an adult	6.6	64.2
(At4)	Attachment	In order to become resilient children need to feel loved and nurtured	10.4	60.4
(N3)	Negative	Labelling a child as being resilient is reassuring for professionals	8.5	50.9
(N6)	Negative	Children can be too resilient which indicates that things are not all right	7.5	62.3
(N7)	Negative	Resilient children often develop adult or parenting behaviours in order to cope but at the cost of not being able to develop normally as a child	7.5	61.3
(N8)	Negative	Resilience is a label which can mask the long-term implications of neglect and being in the care system	7.5	60.4
(P3)	Self-protection	Resilience is a way of protecting yourself in order to survive	10.4	55.7
(So2)	Social construction	Resilience is a term that says as much about the adult using the word, and their perception of a child, than the actual child, because children would not describe themselves as being resilient	10.4	53.8

No statements yielded a high (>83.3%) consensus to be included in or excluded from a definition of resilience. With regard to the statements for which there was a moderate consensus to include (see Table 3), participants agreed that resilience related to a broad notion of survival, carrying on and not falling apart in the face of adversity (S2–4). There was moderate consensus that resilience related to children having coping strategies (A3), maintaining a sense of self-worth (Se3) and being able to manage and adapt to big changes in their life, such as being separated from their parents (A11). There was also moderate consensus that, as a concept, resilience had its limitations and children should not be expected to survive everything they are faced with (N5). Moreover, there was moderate consensus that not everyone has the natural ability to be resilient and many children need help to become resilient (A5).

As shown in Table 4, there was a weak consensus that resilience related to factors pertaining to the individual child such as having a positive sense of self (Se1, 2, 4), taking responsibility for oneself (R3) and an ability to put events in a context (A4) and cope (A7), which may involve unhealthy coping mechanisms (A6). There was weak consensus that positive attachment relationships were part of resilience (At1–4), alongside having something to draw upon outside of the family, such as faith (A8). Interestingly, there was weak consensus that the term resilience has negative connotations both in the way in which it is used to meet professionals’ needs (N3) and the propensity for the term to disguise the needs of child (N6–8). Furthermore, there was a weak consensus that resilience is a term that has been socially constructed and only describes an adult’s perception of the child rather than the child itself (So2).

Only two statements were actively excluded from a definition of resilience. There was weak consensus that ‘Resilience results from a child having to fight for what they get’ (52.8% disagreement vs 12.3% agreement) and a moderate consensus that ‘Resilience is not a word I have used or considered before this research’ (69.8% disagreement vs 13.2% agreement).

For half of the 58 statements (i.e. n = 29), there was no consensus. These statements are listed in Table 5 and relate to the role of culture and religion, resilience as an aspect of the individual’s character or a form of self-protection. It could be suggested that the no consensus category may mask an underlying consensus of mildly agree or disagree. However, if this were the case, the percentage of responses in the strongly/moderately agree and strongly/moderately disagree would total less than 50%, which was only true for four statements, and for those statements, the total was nearly 50%. It therefore appears that there is a genuine lack of consensus on all other statements.

Table 5.

Statements with no consensus.

Theme code	Theme	Statement	% strongly/moderately disagree (x)	% strongly/moderately agree (y)
(A1)	Ability	A resilient child is able to quickly fall back into ordinary patterns of emotional and behavioural development	10.4	49.1
(A2)	Ability	Resilience is the child’s ability to switch off from his or her current situation, even if only for short periods of time	8.5	49.1
(A9)	Ability	Resilience is the child’s ability to close off or separate parts of his or her life in their mind	16	45.3
(A10)	Ability	Resilience is learning to deal with things in a different way	6.6	50
(E1)	Enforced	Children in care have to be resilient; they have no choice	42.5	20.8
(R1)	Responsibility	Resilience is a child managing not to take responsibility or blame for things that are beyond his or her control	18.9	38.7
(R2)	Responsibility	Resilience is a child being able to distance himself or herself from the problem	27.4	34.9
(C1)	Character	Resilience is having a strong character	29.2	27.4
(C2)	Character	Resilience is a personality trait	28.3	28.3
(C3)*	Character	Intelligence and understanding play a role in resilience	13.2	33
(C4)*	Character	Stubbornness is part of resilience	33	15.1
(C5)	Character	Resilience is an emotional stamina	5.7	50
(N1)	Negative	Resilience is not automatically a good thing	25.5	32.1
(N2)	Negative	Resilience comes at a cost to the child	29.2	28.3
(N4)	Negative	Resilience is a label that may prevent a child from accessing services like CAMHS	18.9	44.3
(N9)	Negative	Resilience is a label that when given to a child allows professionals or family members to step back and take less responsibility	23.6	38.7
(P1)	Self-protection	Resilience is a child putting up a barrier around himself or herself	23.6	29.2
(P2)	Self-protection	Resilience is a defence mechanism	19.8	43.4
(P4)	Self-protection	Resilience is a child coming across tougher than they really are	21.7	38.7
(P5)	Self-protection	Resilience is a negative defence mechanism as the child is covering up what needs to come out	36.8	21.7
(P6)	Self-protection	Resilience is children denying aspects of what is happening to them	34.9	22.6
(P7)	Self-protection	Resilience is a defence that is not really there and can be easily knocked down	42.6	17.0
(Cu1)	Culture	Resilience is something that is just expected of children in some cultures, irrespective of what has happened	20.8	31.3
(Cu2)*	Culture	The meaning of resilience is the same across cultures, but how children become resilient varies across culture	15.1	34
(Cu3)*	Culture	Resilience is often assumed as being naturally present in children from certain cultures	20.8	27.4
(Cu4)	Culture	Religious beliefs play a role in resilience	15.1	35.8
(Cu5)	Culture	Resilience is understood differently across cultures	8.5	43.4
(L1)	Living in the present	Resilience is children living in the here and now and not dwelling on the past or worrying about the future	42.5	20.8
(So1)	Social construction	Resilience is a concept society has constructed to describe amazing children who cope with shocking situations	21.7	42.5

CAMHS: Child and Adolescent Mental Health Service; *indicates statements for which x + y < 50%.

In order to explore the impact of consensus building and check the appropriateness of including R2 responses in the analysis (when R3 data were missing), we divided the participants into two groups – those who completed R3 and those who did not – and so had R2 data included in the analysis. A statistical comparison of these groups, using Mann–Whitney U-tests and an alpha level of .01 (because of the need to control for multiple comparisons), revealed that there were no significant differences in the groups’ responses to any of the statements. This suggests that the R3 stage of consensus building did not substantially alter the results and also that it was reasonable to include R2 data in the final analysis for participants who did not complete R3.

Research aim two: comparison by profession

A Kruskal–Wallis one-way analysis of variance was carried out to compare the ratings of social workers, teachers and foster carers on each statement, using the original 6-point Likert scale ratings. Due to the number of comparison tests conducted, alpha was set as .01. In all, 11 comparisons reached statistical significance (see Table 6). Mann–Whitney pair-wise comparisons were computed to ascertain the nature of the difference across the groups. Bonferroni correction was applied (alpha set at p < .016).

Table 6.

Statements for which the ratings differed significantly by a professional group.

Theme	Statement	Foster carer (N = 36) Median (IQR)	Social worker (N = 36) Median (IQR)	Teacher (N = 34) Median (IQR)	Kruskal–Wallis Chi-squared values and significance level	Mann–Whitney Foster carer vs social worker	Mann–Whitney Foster carer vs teachers	Mann–Whitney Teachers vs social workers
Ability	Resilience is the child’s ability to switch off from their current situation, even if only for short periods of time	5 (2)	4 (1)	5 (1)	χ² = 17.81****	U = 325.5****	ns	U = 341***
Enforced	Children in care have to be resilient; they have no choice	4 (4)	2 (2)	3 (2)	χ² = 18.75****	U = 299****	ns	U = 398.5*
Responsibility	Resilience is a child being able to distance himself or herself from the problem	4.5 (4)	3 (2)	4 (1)	χ² = 10.07**	Ns	ns	U = 343.5***
Self-protection	Resilience is a child putting up a barrier around himself or herself	4 (3)	3 (2)	3.5 (2)	χ² = 12.07***	U = 362***,	ns	ns
Self-protection	Resilience is a defence mechanism	5 (2)	4 (2)	4 (2)	χ² = 20.05****	U = 282.5****	U = 363.5***	ns
Self-protection	Resilience is a way of protecting yourself in order to survive	5 (2)	4 (1)	5 (1)	χ² = 14.66**	U = 340.5****	ns	ns
Self-protection	Resilience is a child coming across tougher than they really are	5 (3)	3 (2)	4 (2)	χ² = 22.00****	U = 281.5****	ns	U = 328***
Self-protection	Resilience is a negative defence mechanism as the child is covering up what needs to come out	4 (3)	2 (2)	3 (2)	χ² = 18.71****	U = 304.5****	U = 398.5*	ns
Self-protection	Resilience is children denying aspects of what is happening to them	4 (4)	2 (2)	3 (1)	χ² = 12.05***	U = 375***	ns	ns
Self-protection	Resilience is a defence that is not really there and can be easily knocked down	4 (3)	2 (2)	3 (1)	χ² = 15.71****	U = 375***	ns	U = 303.5****
Unfamiliar term	Resilience is not a word I have used or considered before this research	2.5 (5)	1 (1)	1.5 (2)	χ² = 15.05***	U = 348****	ns	ns

p < .05; **p < .01; ***p < .005; ****p < .0005.

Significant differences were found mostly between foster carers and social workers. Moreover, teachers tended to show greater alignment with the views of foster carers, differing significantly on only two items. Foster carers reported being less familiar with the concept of resilience and there was a trend for greater variability in their responses (as indicated by the interquartile range (IQR)) compared to other groups.

There were significant differences on all statements pertaining to self-protection. Foster carers expressed more agreement with statements that suggested that resilience is a way of protecting oneself in order to survive, which may involve negative defence mechanisms.

Discussion

This study represents the first attempt to explore how social workers, teachers and foster carers understand the concept of resilience. These professionals agreed that resilience relates to the notion of survival, coping and having a sense of self-worth, and that, often, children need help to become resilient. The findings suggest that people working directly with looked after children emphasise aspects of resilience that are both similar and different to those implicated in the literature. Moreover, there were many aspects of resilience for which social workers, teachers and foster carers did not reach a consensus and significantly varied in opinion.

A total of 13 themes reflecting views of resilience were extracted from analysis of open-ended data provided by social workers, teachers and foster carers. The majority of themes located resilience within the attributes of the child, rather than in the family or wider community, a pattern that has been highlighted in other research (Daniel et al., 2009). Statements with the strongest consensus considered resilience as a form of survival and adaptation in the face of difficult experiences and emphasised the importance of self-worth. However, in contrast with the literature, participants placed greater emphasis on the overlap between resilience and coping and the need for children to have helpful coping strategies. It is possible that this may reflect a practical translation of what resilience means and that coping is a more familiar terminology. However, this is interesting because it has recently been suggested that resilience ‘should be considered an important part of the conceptual bridge between coping and development’ (Leipold & Greve, 2009, p. 40).

Participants recognised and agreed that resilience was not universal and that many children need help to become resilient. This suggests a view that the development of resilience is a dynamic process, which can be influenced and enhanced. The data also highlighted views regarding the limitations of the concept of resilience. Specifically, one of the items achieving moderate consensus was children should not be expected to survive everything life presents them with. In the context of looked after children, this may reflect an awareness of the adversity and suffering that these children have experienced. Consistent with this, the data showed some agreement that consideration needed to be given to how the concept of resilience was used with this population, both in terms of the propensity for it to reassure professionals and to mask the underlying needs of the child. McMurray et al. (2008) found that social workers identified all children in their care as being resilient, which was often incongruent with the child’s emotional well-being. This study suggests that people working with looked after children might need a sense of reassurance themselves and might use the idea of resilience to provide such reassurance. An overly positive perception may serve to defend against the distressing and emotional nature of working with vulnerable children (Conway, 2009).

There was only weak consensus that attachment related to resilience. This is perhaps surprising given the centrality of attachment in the theoretical and empirical literature and the hypothesis that the need for a secure base may underpin all other domains of resilience (Daniel, 2006). Weak consensus suggests that while respondents considered attachment to be implicated in the development of resilience, it was less dominant in their view than in the literature. It is noteworthy that this emerged in the context of looked after children, for whom insecure or disorganised attachments and separation from key attachment figures are commonplace. It is possible that people working with looked after children may have a less definitive view of attachment because of the very nature of working with children with disrupted attachments. Rather than focus exclusively on attachment, professionals may be recognising other aspects in the child’s life that may foster resilience. This is in accordance with suggestions that providing a secure base alone is not sufficient to rework IWMs, but that children in care need the opportunity to engage in relationships with other significant adults (Atwool, 2006), siblings and peers (Dunn, 1993, 2004). Moreover, the emphasis on coping strategies, which may have been considered more amenable to change than attachment, is interesting in light of research which suggests that coping style and emotional regulation are differentially linked to attachment style and may mediate the relationship between adversity and resilience (Leontopoulou, 2009).

There was no consensus for aspects of resilience that are strongly emphasised in the literature such as individual character and intelligence (Luthar, 2005). There were also varying opinions regarding the importance of culture. Uncertainty regarding the role of culture may reflect the literature to date, in which the cultural context in which resilience develops is an emerging development (Ungar et al., 2007).

This study found some significant differences in how resilience was conceptualised between professionals. Most were between social workers and foster carers. Although foster carers and social workers are often considered the two key parental figures in the life of a looked after child (Conway, 2009), such differences may reflect that, in practice, their role and relationship with the child are quite different. Foster carers were least familiar with the concept of resilience. This is perhaps unsurprising given that the application and teaching of resilience theory have largely neglected foster carers (Daniel, 2006; McMurray et al., 2008).

With regard to the nature of the differences between professionals, foster carers and, to some extent, teachers tended to view resilience as children being able to cut off, deny or distance themselves from what is happening to them. It is not possible to ascertain whether they considered this to be positive or negative, and it might reflect their experience of being with looked after children rather than their views about the construct of resilience per se. Social workers were less inclined to view resilience in this way, perhaps because the very nature of their work involves managing the practical reality of what is happening to the child; they may have less opportunity to observe and experience the child cutting off or distancing themselves from what is happening, particularly given the limited nature of their contact.

Perhaps the most important implication of this study lies in the fact that many items did not reach consensus. This suggests, in this study at least, the lack of a clear ‘common language of resilience’ (Daniel et al., 2009, p. 9) or a shared multiagency perspective among those working with looked after children. Further research is needed, involving more social workers, teachers and foster carers to support and extend the current findings. Research, in particular, would benefit from exploring further the variation in foster carers’ understanding of resilience and its impact on practice. Moreover, inclusion of other professional groups, such as independent reviewing officers and looked after children nurses, might enhance understanding of the different ways in which resilience is understood. Thereafter, research is needed to consider what people in the network around the child are doing or believe they can do in practice to promote resilience among looked after children.

This study employed a modified Delphi technique in which statements were generated by a subsection of the participants. Teachers were underrepresented in this group (n = 4). However, this is perhaps not too great a concern as, in the final analysis, teachers did not significantly differ as much as foster carers and social workers. Although some participants’ R2Q responses needed to be included in the final analysis (due to missing R3Q data), this approach is acceptable within the Delphi method and has the advantage of providing a larger sample size and, consequently, greater statistical power. Furthermore, no significant differences were found when the R2Q responses included in the final analysis were compared with the R3Q responses alone, suggesting that their inclusion did not bias the findings.

This study has not considered the perceptions or attitudes of children themselves. Future studies would be enhanced by considering the perspective of young people and the level of accordance between their understanding of resilience and professionals around them. Finally, while it would have been interesting to compare consensus building between the different professional groups, this was not possible because of the relatively small samples that would have been obtained if R3Q responders had been divided into the separate professional groups. Studies that specifically assess understanding and consensus within professional groups might further elucidate the concept of resilience and its application.

Footnotes

Acknowledgements

This research was carried out as part of the first author’s clinical psychology doctorate at Canterbury Christ Church University.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author biographies

Rebecca South is a highly specialist clinical psychologist who works within Greenwich CAMHS in Oxleas NHS Foundation Trust.

Fergal W Jones is a clinical psychologist and senior lecturer. He works part time on a clinical psychology training programme and part time in an Improving Access to Psychological Therapies service.

Elaine Creith is a consultant clinical psychologist and is now the professional lead of psychology and psychological therapies in CAMHS/children and young people services in Sussex Partnership NHS Foundation Trust.

Laura M Simonds is a lecturer in psychology at the University of Surrey.

References

Adler

Ziglio

(1996). Gazing into the oracle: The Delphi method and its application to social policy and public health. London, England: Jessica Kingsley.

Atwool

(2006). Attachment and resilience: Implications for children in care. Child Care in Practice, 12, 315–330.

Bostock

(2004). Promoting resilience in fostered children and young people (Resource guide no. 4). London, England: Social Care Institute of Excellence.

Bowlby

(1969). Attachment and loss: Vol. 1. Attachment. New York, NY: Basic Books.

Braun

Clarke

(2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77–101.

Cicchetti

(2002). The impact of social experience on neurobiological systems: Illustration from a constructivist view of child maltreatment. Cognitive Development, 17, 1407–1428.

Clarke

A. D. B.

Clarke

A. M.

(2001). Early adversity and adoptive solutions. Adoption & Fostering, 25(1), 24–32.

Conway

(2009). Falling between minds: The effects of unbearable experiences on multi agency communication in the care system. Adoption & Fostering, 33(1), 18–28.

Daniel

(2006). Operationalizing the concept of resilience in child neglect: Case study research. Child: Care, Health & Development, 32, 303–309.

10.

Daniel

Vincent

Farrall

Arney

(2009). How is the concept of resilience operationalised in practice with vulnerable children? International Journal of Child and Family Welfare, 12, 2–21.

11.

Daniel

Wassell

(2002). The school years: Assessing and promoting resilience in vulnerable children II. London, England: Jessica Kingsley.

12.

Department for Education and Skills. (2006a). Care matters: Transforming the lives of children and young people in care. London, England: The Stationery Office.

13.

Department for Education and Skills. (2006b). Outcome indicators for looked after children. London, England: The Stationery Office.

14.

Department for Education and Skills. (2007). Care matters: Time for change. London, England: The Stationery Office.

15.

Dunn

(1993). Young children’s close relationships: Beyond attachment. Newbury Park, CA: Sage.

16.

Dunn

(2004). Children’s friendships: The beginnings of intimacy. Oxford, UK: Blackwell.

17.

Gilligan

(2001). Promoting resilience: A resource guide on working with children in the care system. London, England: British Association for Adoption & Fostering.

18.

Gilligan

(2004). Promoting resilience in child and family social work: Issues for social work practice, education and policy. Social Work Education, 23, 93–104.

19.

Golding

Dent

Nissim

Stott

(2006). Thinking psychologically about children who are looked after and adopted: Space for reflection. Chichester, UK: John Wiley.

20.

Goodman

(1987). The Delphi technique: A critique. Journal of Advanced Nursing, 12, 729–734.

21.

Graham

L. F.

Milne

D. L.

(2003). Developing basic training programmes: A case study illustration using the Delphi method in clinical psychology. Clinical Psychology & Psychotherapy, 10, 55–63.

22.

Iwaniec

(2006). The child’s journey through care: Placement stability, care planning and achieving permanency. Chichester, UK: John Wiley.

23.

Jackson

Martin

(1998). Surviving the care system: Education and resilience. Journal of Adolescence, 21, 569–583.

24.

Leipold

Greve

(2009). Resilience: A conceptual bridge between coping and development. European Psychologist, 14, 40–50.

25.

Leontopoulou

(2009). Approaching diversity. The Psychologist, 22, 487.

26.

Linstone

H. A.

Turoff

(1975). The Delphi method: Techniques and applications. Reading, MA: Addison-Wesley.

27.

Luthar

(2005). Resilience in development: A synthesis of research across five decades. In Cicchetti

Cohen

D. J.

(Eds.), Development psychopathology: Risk, disorder and adaptation (2nd ed., pp. 739–795). New York, NY: John Wiley.

28.

McMurray

Connolly

Preston-Shoot

Wigley

(2008). Constructing resilience: Social workers’ understandings and practice. Health and Social Care in the Community, 16, 299–309.

29.

Meltzer

Corbin

Gatwood

Goodman

Ford

(2002). The mental health of young people looked after by local authorities in England: Summary report. London, England: Department of Health.

30.

Newman

(2002). Promoting resilience: A review of effective strategies for childcare services. Ilford, UK: Barnardo’s.

31.

Norcross

J. C.

Hedges

Prochaska

J. O.

(2002). The face of 2010: A Delphi poll on the future of psychotherapy. Professional Psychology, Research and Practice, 33, 316–322.

32.

Pipon-Young

Cupitt

Callanan

(2010). Experiences of clinical psychologists in assertive outreach teams: A Delphi survey. Clinical Psychology Forum, 212, 17–21.

33.

Powell

(2003). The Delphi technique: Myths and realities. Journal of Advanced Nursing, 41, 376–382.

34.

Rees

(2013). The mental health, emotional literacy, cognitive ability, literacy attainment and ‘resilience’ of ‘looked after children’: A multidimensional, multiple-rater population based study. British Journal of Clinical Psychology, 52, 183–198.

35.

Rutter

(1999). Resilience concepts and findings: Implications for family therapy. Journal of Family Therapy, 21, 119–144.

36.

Schofield

(2001). Resilience and family placement: A lifespan perspective. Adoption & Fostering, 25(3), 6–19.

37.

Skulmoski

G. J.

Hartman

F. T.

Krahn

(2007). The Delphi method for graduate research. Journal of Information Technology Education, 6, 1–21.

38.

Stein

(2006). Resilience and young people leaving care: Over-coming the odds. York, UK: Joseph Rowntree Foundation.

39.

Stone-Fish

Osborn

(1992). Therapists’ views of family life: A Delphi study. Family Relations, 41, 409–416.

40.

Toth

Cicchetti

Macfid

Maughon

Van Meenen

(2000). Narrative representations of caregivers and self in maltreated pre-schoolers. Attachment and Human Development, 2, 271–305.

41.

Ungar

Brown

Liebenberg

Othman

Kwong

W. M.

Armstrong

Gilgun

(2007). Unique pathways to resilience across cultures. Adolescence, 42, 287–310.

42.

Werner

E. E.

(2000). Protective factors and individual resilience. In Schonkoff

Meisels

(Eds.), The handbook of childhood intervention (pp. 97–116). Cambridge, UK: Cambridge University Press.

43.

Wyman

P. A.

Cowen

E. L.

Work

W. C.

Hoyt-Meyers

Magnus

K. B.

Fagen

D. B.

(1999). Care giving and developmental factors differentiating young at-risk urban children showing resilient versus stress-affected outcomes: A replication and extension. Child Development, 70, 645–659.

44.

Yates

T. M.

Masten

(2004). Prologue: The promise of resilience research for practice and policy. In Newman

(Ed.), What works in building resilience? (pp. 6–15). Ilford, UK: Barnardo’s.