Item 3 is an evidence-taking session with the researchers and Scottish Government officials involved in the growing up in Scotland study. I welcome to the meeting Paul Bradshaw from ScotCen Social Research; Donna Bell and Wendy van Rijswijk from the Scottish Government; and Marco Biagi MSP, who is attending in his role as co-convener of the cross-party group on children and young people.
Indeed, convener.
Thank you very much.
The participants were chosen at random. We took what we call a stratified sampling approach and applied it to child benefit records. Essentially, we selected a period of 12 months; children who were born in that period were eligible for selection. We took data zones—small geographic areas at postcode level—and listed all the data zones for Scotland according to local authority and area deprivation. We selected about 130 areas at random, and all the children in those areas were invited to participate in the study.
A few years ago, a lot of us were quite shocked by the main findings of the UNICEF report “Child poverty in perspective: An overview of child well-being in rich countries”, which showed that the UK was near the bottom and, in some cases, at the bottom of the rankings on five of the six dimensions of children’s quality of life that were reviewed. Has your study picked up good reasons for that? Why was that the case? How can we ensure that things change for the better in the years to come?
We are just at the stage of looking at that aspect of the data. We are about to launch the next round of fieldwork with our existing birth cohort. The children will be eight years old. We will ask them personally what they think about their lives. As part of that, we will attempt to measure their wellbeing. Once we have that data, we will be in a much stronger position to look at the factors that are present in children’s lives up to that point and how, for better or worse, those factors affect their perspective of their wellbeing. To date, that is not something that we have looked at specifically.
Section 3 of your briefing paper deals with social, emotional and behavioural development. It points out that, by the time they start school,
Certainly. The data that you referred to was measured at the point at which children were entering school, but we have earlier data on the same children, which was measured when they were around the age of three. We found that a large number of those children who exhibited difficulties on entering school also exhibited the same difficulties when they were three, which suggested that those children could be identified at an earlier stage. In addition, we found that certain aspects of their lives up to the age of three—in particular, aspects of their home environment, such as how they were parented and their material circumstances—were associated with poorer outcomes.
How are the GUS findings feeding into Scottish Government policy?
We want to be clear that all the work that Paul Bradshaw and his colleagues have been doing on GUS is having a great impact on informing our thinking.
I open up the discussion to committee members. We will have a question from Michael McMahon, followed by one from Mark McDonald.
Thank you, convener. I know how concerned you are about our diplomatic skills this morning, so I will preface my question by saying that, as a sociology graduate, I am aware that sociology has sometimes been described as complex explanations of the patently obvious. That does not mean that longitudinal studies such as GUS are not hugely important and do not have a huge impact in guiding policy, rather that they often reflect what one suspected might be the case in the first place.
A lot of our findings tend to confirm what people thought already; we produce very robust Scottish evidence that those differences and associations exist.
Has the study thrown up any of the structural reasons behind those differences? Are you concerned about the structural reasons for them, or do you just want to identify where the problem lies?
The design of the study and the data that we produce are better for identifying the factors that are associated with difficulties arising and poorer outcomes than for identifying structural issues that are preventing the achievement of better outcomes. We rely on those who are more involved in using our data to say what might be preventing something from happening or whether something could be changed to improve it.
Have you examined the impact that post-natal depression in the mother has on the child? It has an obvious impact on the mother, but is there evidence of its impact on the child when it is not detected early, or not treated, compared with its impact on the child when it is treated?
The answer is yes and no. We have looked at the influence of maternal mental health on child outcomes. A report was published at sweep 4—which would have been about May 2010—that used the first four waves of data to consider the influence of maternal mental health on social, emotional and behavioural outcomes for children. However, we did not specifically measure post-natal depression because the first contact that we had with families in the birth cohort was at 10 months, by which point it is too late to measure post-natal depression in a lot of mothers who suffer from it. Nevertheless, we measure maternal mental health on an on-going basis, and we have shown that poorer maternal mental health is associated with poorer early social, emotional and behavioural outcomes for children aged three and four.
Were the findings the same irrespective of background—whether the mother suffered disadvantage—or were they relative? For example, would a child from a less-advantaged background whose mother had maternal mental health issues suffer a greater disadvantage because of their background and that factor than a child from a more advantaged background whose mother had poor maternal mental health?
All our analysis in the reports that were published in sweep 3 used complex statistical procedures to allow us to control for the effect of background. Therefore, we are able to say that, irrespective of household income and the level of parental education, maternal mental health has an association with behavioural outcomes. Irrespective of whether someone lives in a high-income household or a low-income household, if their mother experiences poor mental health they are more likely to have poor behavioural outcomes.
Is the effect not accentuated? Are children in a less-advantaged situation not likely to suffer more as a result of poor maternal mental health than children in a higher-income household?
I am not sure what the specific evidence will tell you, but it has emerged from the study and other work that we have done that the issue is not specifically about the impact of mental health but about parents’ ability or willingness to engage with services. Those who have a lower standard of education and fewer advantages are much less likely to engage with services, so they are less likely to get treatment. If the problem is not being treated, the impact on the children will be greater.
I was about to come on to that. Your report says that those from a less-advantaged background, or who have a lower standard of education, are less likely to access services. Birmingham City Council touched on that point when it came before the committee. How do we reach those people who are not necessarily unwilling but who do not access the services? Do the services have to make more of a proactive effort to find those people and get them involved? Are you looking at that?
Absolutely. We are very keen to see the light touch of initiatives such as the play talk read bus going to areas in which few standard services are available, through to health visitors taking a more proactive approach. The third sector also plays an important role. Organisations such as Family Circle Care in Edinburgh and the Jeely Piece Club in Glasgow and a range of others throughout Scotland are engaging with people in their family life rather than waiting for them to come to a place. That is an important point.
Like the convener, I was a bit surprised to read that
We use an instrument called the strengths and difficulties questionnaire to measure social, emotional and behavioural development across a number of domains. At that age, a conduct problem simply means exhibiting bad behaviour, if you like. That is probably the easiest way to explain it.
What would you call bad behaviour in a child of four?
I cannot recall the specific items off the top of my head, but I could certainly point you in the right direction. We are talking about unruly behaviour beyond that which would be expected of a child aged four or five, and which would cause the parent to report it frequently. The instrument measures difficulties with peer relationships, such as children not finding it easy to get on with friends, or arguing with friends. It reports on emotional symptoms, which is the extent to which children have low self-esteem or no confidence. The instrument also measures hyperactivity and inattention, or the child’s ability to sit still or pay attention or follow through an action or task that they have been set.
Could it not be argued that some of the expectations that we have of young children going to school are unrealistic? Other countries take a less rigid approach to the very early years so that slightly more exuberant behaviour is not regarded as disruptive.
Curriculum for excellence looks to do things differently. The early phase up to age eight will be much more—I am reluctant to say play based, because the approach is much more structured than those words might infer—engaging for children and will place realistic expectations on them. We have moved on from the approach whereby children need to sit down and listen for long periods of time.
The study suggests that another negative factor is harsh punishment by parents. There is a view in society that children are badly behaved because they are not disciplined enough. That is perpetuated by the television shows that show Supernanny telling parents that they are too soft, sticking the kid on the naughty step and all the rest of it.
That is very difficult. We were very cautious with that finding, because all that we have determined is that there is a relationship between harsh punishment and a greater likelihood of, for example, conduct difficulties or hyperactivity. We do not know in which direction that relationship flows. We do not know whether the harsh punishment is a parental reaction in response to a child who is more difficult to control, or whether the difficult behaviour occurs as a result of a particular style of punishment. That is clear in the report.
Parental consistency is a key issue. Children need to know exactly where they are. You are right that they need firm boundaries, but they need to be able to breathe within those boundaries.
The issue has come up a wee bit because the data that we are looking at now, ahead of the report that is due to be published in May, considers some issues about transition and the reasons that parents gave for deferring their kids’ entry to school. We have been slightly surprised by the findings, because most deferral decisions have been based on parental concerns about development. We expected that deferrals would be more common among more educated parents who would decide, “My child is too young to start school and they would benefit more from a later start.” In fact, most of the decisions are fuelled by real concerns about speech, behaviour or some sort of physical health issue that is pushing them back.
I welcome Gavin Brown back from his 35-minute natural break.
Apologies, convener.
Paul Wheelhouse is next.
I will touch on a couple of issues. First, I will reel back a bit to the methodology of collecting the data and selecting the children who were involved in the study. You talked earlier about that and mentioned that 130 data zones throughout Scotland had been selected.
The question about local-level data is another one that we are often asked. The study was not designed to produce data at a local level, but we have argued about its value with those who are involved in collecting data for community planning partnerships. We have had a number of conversations with them about it, because we can demonstrate from our quite extensive data set the expected outcomes for children who live in an area with high deprivation and high unemployment and, perhaps, in a lone-parent household. We can also demonstrate the different behaviours, experiences and circumstances that would benefit such children.
I totally accept that and understand that that is what you set out to achieve. Is there any way that we can be a bit smarter about using the information that you collect to inform local decision makers about whether they are on track if, for example, there have been improvements in the samples that you have taken locally?
That would be difficult, because we do not have the numbers at the local level to demonstrate it. Obviously, we have children who live in all local authority areas but, in some of those areas, we may have only 60 or 70 families. We could pull out their data and say that, because families in Fife, for example, have been permitted to take advantage of a particular parenting programme since 2008, we will examine how their outcomes have changed. However, the statistical caveats that we would have to put on any such analysis would render it dangerous and mostly meaningless, so we have been reluctant to use the data in that manner.
So that alternative route could be used to inform local decisions.
Yes.
That is helpful.
Yes. We have a small amount of data that hints at that, but with the new birth cohort we are collecting better data, which will give us more of an inkling. There definitely is an indication that more-educated parents are more able to identify developmental delays and will seek help for their children earlier. Less-educated parents might be less aware of the speech or physical development stage at which their children should be at certain points.
Elaine Murray made a point about behavioural differences being spotted in children from less-advantaged backgrounds. Could some of that behaviour be due to undiagnosed ADHD or autism, for example? It might not necessarily be that the parents’ behaviour is having an impact, but that they are not putting their children forward for assessment and there are, therefore, undiagnosed disabilities.
Absolutely.
Both you and Donna Bell have talked about messages regarding behaviour and the symptoms of problems. How big an impact could the issue have on the delivery of teaching, particularly at the early years stage? Will it help to inform us in changing the approach? I am on the parent council of my local school, and have often seen how a change of teacher can have a dramatic impact on the outcomes for some of those who are seen as problem children. A change of approach, with perhaps a fresher view and teaching style and more buy-in to modern teaching methods can have a dramatic impact on individual children’s performance. Have you picked up any key messages there?
We have not measured that. In our data collection, we have not asked whether there has been a change of teacher, but I agree that that could have an impact on children’s experiences at school.
In pre-school environments and early learning settings in primary school, the workforce is clearly the key measure of quality. We have invested quite a bit in the workforce in those environments to ensure that it is absolutely up to date with the latest thinking in child development, modern teaching methods and engagement with children—exactly the matters that Paul Wheelhouse raised. We have done a lot of work on that in the past three to four years, and it is starting to pay dividends in the early learning environment.
As part of our dissemination programme, we regularly share our findings with all student teachers attending Moray house. Over the past two years, we have spoken to all bachelor of education and professional graduate diploma in education students. There has been a practical presentation on the study and on how to understand the findings, followed by workshop exercises on how the students might use the findings on a day-to-day basis when they become teachers.
That is excellent. Thank you.
Following on from what Paul Wheelhouse said, my question is about paragraph 3.1.1 in your briefing paper, which is on cognitive ability. I am interested in whether there were changes in the three cohorts. Some of the figures that are referred to are quite stark, such as the 18-month gap in vocabulary ability by the age of three, which persists during the pre-school year. Further on, the paper talks about the importance of early communication skills “at age 22 months”, which is very young. Going on what has been said previously, if we produce more leaflets asking parents to come forward, that would tend to help the more educated ones; if anything, it might widen the gap. Might the gap widen or narrow in future?
We do not yet have data that allows us to look at differences in cognitive ability between the cohorts. That was not measured with the child cohort. We will look to measure it with the new birth cohort, but our first contact with them at age three will be next time round.
The committee has heard a lot of evidence about preventative spend. We cannot afford to see that gap widen any further.
Are the initiatives that you are talking about largely targeted at the families that are most at risk?
No. The bookbug programme is universal—it is for everybody in Scotland—although the planned assertive outreach programme will be targeted at specific areas or specific groups of parents who could do with that extra helping hand.
You mentioned policy interventions. The underlying motivation for the study is to inform how we try to improve outcomes. Do you measure at an individual level whether children access these sorts of services? You mentioned looking at an area in which a service had been offered. Can you make comparisons between those who take up services and those who do not?
We can do that to an extent, on the basis that parents are able to tell us that they are accessing a particular service. For the past couple of sweeps with the existing birth cohort, we have asked specific questions about involvement in parenting programmes, and particularly whether families have been involved in the triple P or incredible years programmes. We have not analysed that data yet, but the intention is to gather enough information on families who access those programmes to try to draw comparisons between them.
Is access to services more broadly, such as health visitors or specialist speech support, included in the data set?
Yes.
That is interesting.
That is difficult to say. Given the strength of different circumstances and characteristics, any robust measure of advantage or disadvantage, such as the level of parental education or income, will have a much more dominant effect than more minor characteristics such as parenting, so it is difficult to define the extent to which that will make a difference. What we know for certain, however, is that such characteristics can make a difference, particularly for children in disadvantaged circumstances. There are programmes that are capable of providing support for that and improving children’s experiences.
A lot of the international evidence and some of the evidence from Scotland suggests that some people in financially straitened circumstances are very good parents. We would not make a link between such circumstances and poor parenting. It is very clear to us that it is about what parents do, not what they are—we just need to remember that.
I am interested in how far it closes the gap when a parent from a family that lacks resources and faces lots of challenges is able to deliver better parenting. However, you are saying that you cannot really measure that at the moment.
We have not looked at that specifically in the context of the data that we have gathered thus far.
The feedback that we are getting in our discussions on the parenting strategy is that, although circumstances might be difficult, parents love their children and want the best for them, but circumstances can sometimes be so overwhelming that the parents do not have the capacity to provide that. In circumstances in which a lost shoe is the end of the world, it is the straw that breaks the camel’s back. An awful lot of work remains to be done to give parents the capacity to get themselves to a point at which they are able to look after their children as they want to.
I thank our witnesses for coming along and giving evidence. I also thank members for their questions.