Scottish Study of Early Learning and Childcare: final report
This report brings together data from across the 6 phases of the Scottish Study of Early Learning and Childcare to consider some key questions about the impact of the expansion of funded ELC in Scotland from 600 to 1140 hours
Footnotes
1 Cognitive and language development was measured in the SSELC using data from the Ages and Stages Questionnaire (ASQ), which formed part of the keyworker observations (see Appendix A).
2 Social, behavioural and emotional development was measured using the Strengths and Difficulties Questionnaire (SDQ), which formed part of the keyworker observations (see Appendix A).
3 Mental wellbeing was measured in the parent survey using the short form Warwick-Edinburgh Mental Wellbeing Scale (SWEMWBS).
4 Broadly, family wellbeing in the context of ELC is considered to be a combination of children and parents’ health and wellbeing, and the ability of parents to undertake suitable parenting and activities that may contribute to the long-term prosperity of the family unit. Responses to the questions around the home environment in the parent survey have been used with reference to the high-level outcome on ‘family wellbeing’. See the Evaluation Strategy for more detail.
5 Throughout this report the term ‘parents’ should be understood to refer to carers also.
6 More information on the eligibility criteria for funding is available at: Funded early learning and childcare - mygov.scot
7 See Early learning and childcare expansion evaluation - gov.scot
8 See Funded Early Learning and Childcare Statistics 2025 - gov.scot
9 Broadly, family wellbeing in the context of ELC is considered to be a combination of children and parents’ health and wellbeing, and the ability of parents to undertake suitable parenting and activities that may contribute to the long-term prosperity of the family unit.
10 Early learning and childcare expansion evaluation - gov.scot
11 See, for example: Coronavirus (COVID-19): impact on children, young people and families - evidence summary October 2020 - gov.scot (www.gov.scot); or COVID-19: impact on child poverty and on young people’s education, health and wellbeing - House of Lords Library (parliament.uk)
12 Although there was an increase in average adult mental wellbeing between 2022 and 2023. See The Scottish Health Survey 2023 - volume 1: main report - gov.scot (www.gov.scot)
13 See Understanding the Cost of Living Crisis in Scotland
14 See Marmot Review 10 Years On - IHE
15 Newlove-Delgado, T, et al. (2023). Annual Research Review: The impact of Covid‐19 on psychopathology in children and young people worldwide: systematic review of studies with pre‐ and within‐pandemic data - Newlove‐Delgado - 2023 - Journal of Child Psychology and Psychiatry - Wiley Online Library; and Fitzsimons, E., & Blundell, R. (2021). The impact of the COVID-19 pandemic on children’s development and well-being: Evidence from the UK Household Longitudinal Study. Understanding Society Working Paper Series, No. 2021-09. Institute for Social and Economic Research, University of Essex. Understanding Society Innovation Panel wave 13: results from methodological experiments – Understanding Society – Understanding Society
16 COVID-19 Early Years Resilience and Impact Survey (CEYRIS) – Round 4
17 UoY-mitigating-impacts-of-covid19-in-early-years-rapid-evidence-review.pdf and Sustained benefits of early childhood education and care (ECEC) for young children’s development during COVID-19
18 Early child development statistics - Scotland 2023 to 2024 - Early child development - Publications - Public Health Scotland published April 2025
19 Parents' views and use of early learning and childcare: report 2017 - gov.scot (www.gov.scot)
20 Both the ELC Leavers and the Comparator 3s were randomly drawn samples, representative of children receiving funded ELC at the relevant age. As such, they may include a small number of children who were eligible for funded ELC at age two. Hence not all Comparator 3s were at the start of their funded ELC.
21 See ‘Non-random approach for a discussion of the sampling approach at Phase 1’ for further information on sampling approach taken at Phase 1. Information on local authority areas participating in individual phases is including in the standalone reports for each phase accessible via the Scottish Government website.
22 Some differences were noted in terms of the ethnic make-up of local authorities, with a smaller minority ethnic population in the participating authorities. There was also a higher proportion of first-time mothers under the age of 25 in participating authorities and a greater rate of unemployment.
23 Ages and Stages Questionnaire (agesandstages.com)
24 Strengths and Difficulties Questionnaire (sdqinfo.org)
25 A statistically significant finding is one that would be very unlikely to be observed in a sample of data simply by chance if there is no real underlying change or difference in the wider population. The magnitude of the difference is only an estimate but is very likely to be greater than zero.
26 See Appendix A for details on how long-term health conditions were defined and a possible explanation for these anomalous findings.
27 A model is basically a mathematical equation, in this case, that predicts the likelihood of a child being on schedule on the ASQ communication domain, given certain information that we may know about them. Table 4.1 summarises the information that we need to know in order to make this prediction.
28 This is a slightly different definition of long-term health condition than that used previously, as it is taken from the parent questionnaire. The question asks whether the child has any physical or mental health conditions or illnesses lasting, or expected to last, for 12 months or more; with a follow-up of whether the condition(s) or illness(es) have an effect on the child at play or on any other normal activity for a child her/his age. See Appendix A for more details.
29 For any factor identified as protective, the inverse could be said to be a risk factor – for example, looking at books at least three times a week could be described as protective against delayed cognitive and language development, while the opposite, looking at books no more than twice a week could be described as increasing the risk of delayed cognitive and language development.
30 Figures are based only on those with completed keyworker questionnaires at both age two and age three. Hence, they may differ from figures published previously.
31 In such models, the earlier measure is expected to dominate, accounting for some or all of the association between other variables and the outcome. More details of how the models were constructed, and how to interpret them are included in Appendix A.
32 Table 4.2 states whether the factors included from the model were taken from the age two or the age three data. Most variables were available at both time points. For longitudinal models it is common practice to take the variable from the earlier time point. Home learning activities at age two, for example, could reasonably be expected to affect the outcome a year later. Whereas home learning activities at age three would not affect a child’s cognitive development at exactly the same time point, unless they are interpreted as something which has been going on for a while. Some variables, however, such as those related to health, have been taken from the age three data, as the condition may not have been identified by age two.
33 Melhuish, E., Gardiner, J., and Morris, S. (2020). Study of Early Education and Development (SEED): Impact Study on Early Education Use and Child Outcomes up to Age Five Years. London: Department for Education. Available at: Early education and outcomes to age 5 - GOV.UK; Sylva K, Melhuish E, Sammons P, Siraj-Blatchford I and Taggart B (2004) The Effective Provision of Pre‑school Education (EPPE) Project: Final Report. Findings from Pre‑school to the End of Key Stage 1. London: Department for Education and Skills; Goodman A and Gregg P (eds) (2010) Poorer children’s educational attainment: How important are attitudes and behaviour?; and Melhuish, E. (2010). Impact of the Home Learning Environment on Child Cognitive Development: Secondary Analysis of Data from Growing Up in Scotland. Edinburgh: Scottish Government Social Research.
34 Figures for the Eligible 2s at age three differ slightly from those in Figure 4.2 because they include a small number of additional cases who did not have a keyworker questionnaire at age two.
35 Help for early years providers: Fine motor skills
37 See Appendix A and the supplementary tables.
38 This is expected of autoregressive models, as much of the variation is explained by the outcome variable at the earlier time point.
39 Behaviour in Scottish schools: research report 2023 - gov.scot
40 The ASQ is designed to pick up developmental milestones relevant to the age of the child. Two different questionnaires were used for the ASQ, one for those less than four years and nine months, and one for those of this age and above. Differences in outcomes for which questionnaire was used were not significant, but being older here refers to being older within each of the two age groups.
41 The significance tests included within the regression models are considered better than those conducted for the straightforward cross-tabulations as they account for more of the variation within the data. More details are provided in Appendix A.
42 An additional consideration is that paying for additional hours is not entirely independent of the timing of the data collection. This affects the accuracy of the model. More details are provided in Appendix A.
43 See, for example, Speech, language and communication development among children in Scotland during the COVID-19 pandemic, 7. Impact of COVID-19 on the closing of the poverty-related attainment gap - Closing the poverty-related attainment gap: progress report 2016 to 2021 - gov.scot and COVID-19 and literacy: The attainment gap and learning loss | National Literacy Trust
44 On the other ASQ domains, the proportion on schedule at age three was similar to or greater than the proportion on schedule at age two, both pre- and post-expansion. The ASQ is designed to be age-specific, so with no external influences, the proportion on schedule at age two would be expected to be the same at age three as at age two. However, on specific measures it appears that developmental milestones are being reached earlier or later than when the ASQ was first calibrated. Hence, we should not place too much emphasis on the decline in gross motor skills between ages two and three, nor should we put too much emphasis on the increase in problem-solving skills. Instead, it is the comparison in the size of the change pre- and post-expansion that is more relevant.
46 Young Children Develop in an Environment of Relationships: Working Paper No. 1; Sylva K, Melhuish E, Sammons P, Siraj-Blatchford I and Taggart B (2004) The Effective Provision of Pre‑school Education (EPPE) Project: Final Report. Findings from Pre‑school to the End of Key Stage 1. London: Department for Education and Skills; Desforges C and Abouchaar A (2003) The Impact of Parental Involvement, Parental Support and Family Education on Pupil Achievements and Adjustment: A Literature Review. Department for Education and Skills Research Report 433.; and Social Mobility Commission (2020). The long shadow of deprivation: Differences in opportunities across England. London: HM Government.
47 While it may appear contradictory that there has been an increase in full-time working but no increase in mean hours worked, this is because mothers who were not in employment were not included in the calculation of mean hours worked. The increase in hours for some mothers from part-time to full-time is balanced by an increase in the number of mothers moving from not working to part-time work. As this latter group are only included in the calculation when they are in work, the overall effect is of no increase in mean hours worked.
48 Odds are the probability of something occurring (in this case, the mother being in work, training or full-time education) divided by the probability of it not occurring. It is common to compare these for two separate groups, for example, pre- and post-expansion, giving what is known as an ‘odds ratio’. This odds ratio can be interpreted as the magnitude of the impact of everything that has changed between 2019 and 2024, including the expansion of funded ELC, as well as changes to work patterns following the protective public health measures following the COVID pandemic. See Appendix A for a full explanation of odds and odds ratios in relation to regression analysis.
49 Scotland's Labour Market Insights: April 2025 - gov.scot
50 HI11 Regional labour market: headline indicators for Scotland - Office for National Statistics
51 Who are the hybrid workers? - Office for National Statistics
52 While all figures in this section apply to the parent respondents, in most cases this was the child’s mother. Figures are provided in the supplementary tables for mothers, and for all three measures, figures for the mother are almost identical to those for the parent respondent.
53 The Scottish Health Survey 2023 - volume 1: main report - gov.scot
54 Oates J, Gervai J: Mothers Object Relations Scale: Assessing mothers’ models of their infants. Open Univ. 1984. See also MORS Tools | Mothers Object Relations Scales (MORS) (morscales.org)
55 Who are the hybrid workers? - Office for National Statistics
56 Scottish Surveys: Core and Harmonised Questions - gov.scot
57 Scottish Government (2012) The Scottish Child Health Programme: Guidance on the 27-30 month child health review, Edinburgh: Scottish Government.
58 Bedford, H., Walton, S., Ahn, J. (2013). Measures of Child Development: A review, London: Centre for Paediatric Epidemiology and Biostatistics, UCL Institute of Child Health.
59 See Ages and Stages Questionnaire (agesandstages.com)
60 See Strengths and Difficulties Questionnaire (sdqinfo.org)
61 Goodman R (1997) The Strengths and Difficulties Questionnaire: A Research Note. Journal of Child Psychology and Psychiatry, 38, 581-586.
62 Matheny AP, Wachs TD, Ludwig JL and Phillips K (1995) Bringing order out of chaos: Psychometric characteristics of the confusion, hubbub, and order scale, Journal of Applied Developmental Psychology, 16(3), 1995, 429-444, https://doi.org/10.1016/0193-3973(95)90028-4.
64 Data collection and publication guidance: Disability
65 Care Inspectorate, A quality framework for daycare of children, childminding and school-aged childcare (2022).
66 The interaction term allows for the examination of whether the relationship between the outcome variable and the phase variable changes depending on the value of the sex variable. It is basically an additional variable in the model which is the product of the values for the two independent variables (phase and sex).
Contact
Email: socialresearch@gov.scot