What Makes Parenting Programmes Work In Disadvantaged Areas?
By by Stephen Scott, Thomas O’Connor and Annabel Futh
Ensuring that a child is brought up experiencing warmth, love and encouragement within safe boundaries is far harder for parents who live in the stressful conditions found in poor neighbourhoods. Children raised in poverty do less well than children raised in more favourable circumstances on a range of measures of attainment and quality of life. Yet if children are brought up with warm, firm encouraging parenting, the evidence is clear that they can succeed even in these more adverse circumstances. The Government is pursuing this agenda energetically, with policy initiatives such as Every Child matters (DfES 2004), and the Respect agenda (Cabinet Office 2006) which proposes parenting classes to help families improve child behaviour and achievement. This project describes an evaluation of what factors influenced the effectiveness of a parenting intervention in one of the poorest parts of Britain. It was called the Primary Age Learning Study (PALS).
Parents from minority ethnic groups may have extra burdens. Those who are well established may experience discrimination across several contexts, while those who have arrived more recently may additionally struggle with language difficulties, lack of information about how to access services and benefits, and isolation. Parenting styles that fit familiar circumstances in the country of origin may be challenged by and found unacceptable in the new settings in Britain. For example, some disciplinary practices may be frowned upon, and might even instigate referral to social services. Equally, families from minorities may bring parenting styles that are advantageous in the new settings, for example more social cohesiveness and closer supervision of children that helps the well-being of the parents and protects children in higher risk urban conditions. Currently many public services are under-used by minority parents – the services are failing them by not being acceptable and accessible. There is little information whether programmes based on Western ideas are acceptable to, and work for, minority families. The PALS study aimed to address these issues.
The capacity of parents to respond sensitively to their child’s overtures and needs is crucial to the development of secure attachment patterns and for optimal social and emotional adjustment, and was a key outcome measure. This, along with positive parental attention, is associated with the growing child’s ability to make other relationships successfully, such as those with friends, and later, intimate love relationships. In contrast, children raised by parents who are insensitive and use harsh, unpredictable discipline display insecure attachment patterns and control their emotions poorly, often displaying aggressive behaviour and grow up to become seriously anti-social.
The Webster-Stratton Incredible Years programme was used in combination with the SPOKES (Supporting Parents on Kids Education) programme to help reading. In common with the most effective programmes, it offers both emotional support and coaching in skills to improve the relationship with the child. With troubled (clinically referred) children, several studies using these interventions have shown large improvements in child outcomes, parent satisfaction has been high, and dropout rates low.
However, prevention trials have had more mixed success. Difficulties include not getting many parents to enrol (typically only about a third of families), and individual children with no problems, so there is little room for measurable improvement. Against this, advantages of approaches which target the whole population include widening access (the potential to enrol families who would not use a formal service) and, through screening the needs of all, identifying who is experiencing difficulty but is not receiving help.
A previous prevention trial by the King’s College research group, called SPOKES, used the same programme with troubled children and has proven effective (Scott, Sylva et al. 2006). The PALS trial was (1) shorter, with 16, not 26 sessions; (2) offered to parents of any child, rather than only those with children in difficulty, though some places were reserved for them; (3) took place in an area with a predominantly minority ethnic population around the Old Kent Road in Peckham, London, one of the most multiply disadvantaged wards England.
Characteristics of participants
All reception and Year 1 pupils were assessed by teachers and parents using a short questionnaire; a subsample was intensively studied. Families were randomly chosen to receive the intervention or be controls, according to which class they were in. In the intervention classes, over half the parenting group places were open access, but some places were reserved for parents revealed by the questionnaire to have children with difficulties. Parents of 8-10 children were invited to attend a group for two-and-a-half hours in the morning, after dropping their children off at school. The programme comprised the basic 12-week Incredible Years parenting programme combined with a 6-week version of a previously proven reading readiness programme devised by Professor Kathy Sylva of Oxford University. Group leaders, who were already trained in the intervention, were supervised weekly using videotapes of sessions; this was crucial to ensure treatment fidelity and provided group leaders with support and advice for managing difficult cases.
Two-thirds of the families were from minority ethnic groups, mainly West African and African-Caribbean. Half were lone parents; a third were on incomes of £175 a week or less, the level of the poorest 5 per cent of Britain.
The white British families were using the drop-in service at school provided by the local Child and Adolescent Mental Health Service three times as much as minority ethnic families.
According to a widely validated questionnaire measure (SDQ), 24 per cent of this sample showed elevated levels of anti-social behaviour. That is greater than the national rate of 18 per cent, but only slightly so, and implies that the use of demographic information may not be adequate for identifying those families most in need of intervention.
Two-thirds of parents who were offered the programme attended at least one session; all ethnic groups attended equally. Mean attendance was 7 sessions out of 18 offered; half of starters attended at least 5 sessions, the minimum likely to lead to change. A separate qualitative study of 32 non-engagers (who declined the invitation to attend) and non-completers (who started but then dropped out) found that the overwhelming reason given was other commitments, in over 90 per cent of cases new jobs or training courses being attended. That implies that lack of attendance at the intervention should not be interpreted as resistance or non-compliance, but instead reflects the complex and frequently changing demands on the families.
Changes in parenting
93 per cent of parents said they were well or extremely satisfied with the programme, with equally high satisfaction rates across ethnic groups. All schools welcomed the programme, none refused it. The local authority (Southwark) has now taken over mainstream funding of the programme and expanded the number of participating schools.
Sensitive responding was assessed using the ‘gold standard’ of direct observation during a play task with the child. For those offered the intervention, responding improved substantially compared with controls (this includes those who did not attend, so is a strict test at a population level). Those who attended at least five sessions improved more.
Similar changes were seen in child-centred parenting (attending to, praising the child), but there was no change in the number of directions given. Reported use of calm discipline (withdrawal of privileges, sending the child for a short time to a boring place) increased, and demeaning criticism reduced; again, better attendance led to bigger effects. Reported use of praise and rewards did not change (though on observation it had increased). Changes in all parenting measures were equally good across ethnic groups.
Changes in children
Child attention and on-task behaviour were also assessed using direct observation. They improved substantially; anti-social behaviour was not measurable using direct observation as almost none was seen during the play task with the parent. Parent report measures (interview and questionnaire) of child anti-social and ADHD (attention deficit) problems showed low levels to begin with, and showed no improvement compared with the control group. This may be because some of the measures could not detect change – for example, children were not lying or stealing at the start, so could not improve on this measure. Tests of reading ability also did not show improvement, which may be because on average, parents only attended two sessions on how to read with children. These findings underscore the importance of using a range of outcome measures, and choosing ones likely to be sensitive to change in children recruited from the general community as opposed to from clinics.
The researchers conclude that:
- A high proportion of parents from all ethnic backgrounds are prepared to enrol in parenting programmes despite living in disadvantaged, stressed circumstances, provided the intervention is attractive, well planned and well supported; adding a reading component helps universal appeal.
- Behaviourally based programmes change not just individual parenting acts, but can change the whole pattern of relating so it becomes more warm and reciprocal – a style known to be related to child attachment security and emotional well-being.
- Despite the considerable cultural differences in beliefs about how children should be disciplined and brought up, parents from all ethnic groups improved equally after attending the programme. This is a finding of major importance for those planning to offer parenting programmes in areas with minority ethnic populations.
- To get substantial changes across a wide range of outcomes, strategies need to be in place to help parents attend a reasonable number of sessions, say around ten. To achieve this may require individual home visits in some cases.
- Intervention projects should routinely gather simple outcomes data such as attendance, satisfaction and, crucially, child outcomes. Only this way will less effective practice be uncovered, which then allows for improvement.
- Targeting by geographical area is inefficient – in this study the rate of child problems likely to lead to social exclusion was not particularly high. In the country as a whole, by far the majority of those at risk of exclusion do not live in small pockets of high deprivation; even in such pockets, most do reasonably well. It is more effective, and more cost-effective, to use a simple questionnaire assessment and then only select those in need.
About the project
Initially, the Strengths and Difficulties Questionnaire (SDQ) was filled in for the whole population of reception and Year 1 pupils in four schools, with a return rate of 99 per cent by teachers (n= 665) and 80 per cent by parents (n=532). Then a subsample of 174 children were intensively studied using direct observation, interviews, and questionnaires. Classrooms were randomly allocated to intervention or control groups; to avoid potential bias, researchers were blind to family group allocation. The intervention programme cost around £1,300 per child.
For more information
The full report, What makes parenting programmes work in disadvantaged areas? The PALS trial by Stephen Scott, Thomas O’Connor and Annabel Futh, is published by the Joseph Rowntree Foundation