Growing up in Scotland: children's social, emotional and behavioural characteristics at entry to primary school

Chapter 2 THE EXTENT AND NATURE OF SOCIAL, EMOTIONAL AND BEHAVIOURAL DIFFICULTIES

2.1 Key findings

• At the point of entry to primary school the vast majority of children do not present with any social, emotional or behavioural difficulties as measured via the SDQ. Amongst the difficulty scales, on all but conduct problems, over 80% of children return scores within the accepted normal range, that is the range in which most children would be expected to score. 73% do so in relation to conduct problems.
• Between 10% and 27% of children are reported by their parents to have behaviour which places them above the normal range on any subscale indicating moderate or severe difficulties. These proportions broadly match that which is expected of SDQ scores taken from any community population.
• Conduct problems were the difficulties most prevalent, whereas emotional problems were those least so.
• Using cluster analysis, it was possible to group children into five groups based on their characteristics across the 5 SDQ sub-scales. The largest groups were those in which the children had no, or only singular difficulties. Two groups of children with more problematic behaviour emerged. One had considerably higher than average hyperactivity/inattention scores along with conduct, peer and pro-social difficulties. The other had higher than average emotional difficulties along with issues around conduct, peers and hyperactivity.

2.2 Prevalence of individual difficulties

Scoring information for the SDQ provides a useful, although acknowledgedly rough and ready, method through which individual scores on each of the scales can be classified as normal, borderline or abnormal. ³ The parameters for each classification are slightly different on each of the scales and are detailed in Appendix 2.

As shown in Table 2.1, at the point of entry to primary school the vast majority of children do not present with any social, emotional or behavioural difficulties as measured via the SDQ. Amongst the difficulty scales, on all but conduct problems, over 80% of children return scores within the 'normal' classification. The remainder on each scale are split fairly evenly between borderline and abnormal classifications. Between 5% and 15% of children have scores which fall within the borderline range in any one scale, and

between 5% and 12% of children are reported by their parents to have behaviour which places them in the abnormal classification. These proportions broadly match that which is expected of SDQ scores taken from any community population ⁴ and are comparable to those found in earlier Scottish research with a similarly aged, though smaller, sample drawn from Edinburgh and North Lanarkshire (Dunlop et al., 2008). Children were most likely reported to have difficulties in relation to conduct, where a little over one quarter had scores classified as borderline or abnormal. The least prevalent difficulties were related to emotional symptoms, where just 1 in 10 children had a score outwith the normal range. The data also indicates that most children have good pro-social behaviour with just 7% scoring in the borderline or abnormal range for that scale.

Table 2.1 Classifications and mean scores on all SDQ scales at entry to primary school

Taken together, it is unsurprising therefore, that scores on the overall difficulties scale are also concentrated towards the lower end. Whilst the full scale ranges from 0 to 40, the mean score amongst Primary 1 children is just 7.8, well within the 'normal' range of 0 to 13. Indeed, almost 9 in 10 children have total scores within the normal range, 6% have

scores classed as borderline (a score between 14 and 16), and 5% have scores classed as abnormal (a score of 17 or above). Again, these results broadly match the normative SDQ data in which 8% of 5-10 year olds score within the abnormal range for the total difficulties scale, and where the mean difficulties score is 8.6.

2.3 Patterns of shared difficulties

Consideration of the various sub-scales and overall difficulties scores provides a useful indication of the spread of various developmental difficulties amongst children at school-entry age. However, whilst the total difficulties score is indicative of the existence of multiple difficulties, it does not illustrate in what combination and to what extent children exhibit difficulties concurrently in each of the individual domains.

The co-occurrence of problems associated with attention-deficit hyperactivity disorder ( ADHD) and conduct disorder ( CD) or oppositional defiant disorder ( ODD) in childhood have been demonstrated in a range of research studies (see Campbell et al, 2000 for a discussion). A number of these studies indicate further that children with co-occurring conduct problems have more severe and persistent difficulties later in life (Barkley et al., 1990; Weiss and Hechtmann, 1993). Research by Sonunga-Barke and colleagues (1997), for example, showed that pre-school children could be categorised into six groupings based on their scores on the Behavioural Checklist, an instrument similar to the SDQ. The first and largest grouping contained children with no problems and a further three had children with specific but low-level issues. The remaining two groups of children displayed greater conduct problems; for one group this was combined with significant emotional difficulties, and for the other with extreme hyperactivity. Following up the children at age 8, the research found that children in the hyperactive/conduct group were likely to have prolonged problems.

It is pertinent, therefore, to examine patterns of shared difficulties in the GUS data. To do so, cluster analysis was undertaken in an attempt to create identifiable groups of children with similar characteristics across each of the SDQ sub-scales. A description of the cluster analysis is provided in Appendix 3.

2.3.1 The five clusters

The cluster analysis produced five 'clusters' of children on the basis in which they shared difficulties, or not, on each of the SDQ sub-scales. Table 2.2 shows the mean individual scores for each cluster on each of the SDQ scales, plus scores for the the overall sample. A summary of the behavioural characteristics of the children in each cluster is provided in Figure 2 A.

Figure 2 A Behavioural characteristics of children in each of the five clusters at entry to primary school

The majority of children fell into clusters where behaviour was generally non-problematic; 62% are in either cluster 2 or 5 where scores on the problem scales were all or mostly low and pro-social scores high. Only two clusters contain children who exhibit a particularly concerning pattern of shared difficulties. Children in cluster 3 have high scores on all of the problem scales with particularly high hyperactivity scores. Children in cluster 4 also have higher scores on the problem scales, but with particularly high emotional symptoms scores. Around one-fifth of children fell into one of these groups.

Table 2.2 Mean scores on SDQ scales at entry to primary school by cluster

2.4 Summary

The vast majority of children do not present with any social, emotional or behavioural difficulties when they start primary school. Between 70% and 80% of children return scores within the accepted normal range on all of the difficulty scales, that is the range in which most children would be expected to score. Only between 5% and 12% of children are reported by their parents to have behaviour which places them within the abnormal range on any subscale indicating severe difficulties. These proportions broadly match that which is expected of SDQ scores taken from any community population. The most prevalent difficulties were related to conduct problems, whereas emotional problems were least evident.

The cluster analysis produced five groups of children based on their characteristics across the five SDQ sub-scales. The largest groups were those in which the children had no, or only singular difficulties. Two groups of children with more problematic behaviour emerged. In cluster 3, children had considerably higher than average hyperactivity/inattention scores along with conduct, peer and pro-social difficulties. In cluster 4, children had higher than average emotional difficulties along with issues around conduct, peers and hyperactivtity.