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

This report investigates the extent and nature of social, emotional and behavioural difficulties among Scottish school children around the age they enter primary one, and shows which children are most likely to have these difficulties.


1 Further information on the design, development and future of the project is available from the study website:

2 Throughout this report, to avoid over-repetition, the four difficulty sub-scales of the SDQ - conduct problems, emotional symptoms, hyperactivity/inattention and peer problems - will be referred to variously as sub-scales or behavioural domains.

3 Note that within the report, scores in the 'borderline' range are also referred to as 'moderate' and scores in the 'abnormal' range are also referred to as 'severe'.

4 Normative data from British samples is available at

5 The statistical analysis and approach used in this report represents one of many available techniques capable of exploring this data. Other analytical approaches may produce different results from those reported here.

6 The odds ratios from the regression models are included in Table A1.1 in Appendix 1. The interpretation of odds ratios is explained in Appendix 1.

7 Note: for all SDQ scales, except pro-social, a higher score indicates greater difficulties and should be considered more negative. For the pro-social scale, a higher score is more positive. Thus, on the difficulty scales, where the mean score decreases between pre-school and primary school this is considered a positive improvement in behaviour whereas an increase in scores between sweeps is considered negative.

8 The results of the regression analysis are shown in Table A1.2 in Appendix 1.

9 We also attempted to use hierarchical clustering (where the number of clusters is not specified beforehand but is dictated by the data), however, this method did not produce satisfactory clusters. Those which were produced tended to consist of one very large cluster that made up 95% of the data and a number of tiny clusters that contained individuals who were in fact very different. K-means clustering gave more relevant results.

Back to top