Afterschool care for school-aged children in Ireland

NUMBER 222, November 2025 Access to Autonomous Housing for Beneficiaries of International Protection in Ireland KEIRE MURPHY AND AMY STAPLETON ESRI RESEARCH SERIES NUMBER 184 June 2024 E V I D E N C E F O R P O L I C Y Access to Autonomous Housing for Beneficiaries of International Protection in Ireland KEIRE MURPHY AND AMY STAPLETON ESRI RESEARCH SERIES NUMBER 184 June 2024 E V I D E N C E F O R P O L I C Y Access to Autonomous Housing for Beneficiaries of International Protection in Ireland KEIRE MURPHY AND AMY STAPLETON ESRI RESEARCH SERIES NUMBER 184 June 2024 E V I D E N C E F O R P O L I C Y Afterschool care for school-aged children in Ireland HELEN RUSSELL, EMER SMYTH AND EVA SLEVIN ESRI RESEARCH SERIES AFTERSCHOOL CARE FOR SCHOOL-AGED CHILDREN IN IRELAND Helen Russell Emer Smyth Eva Slevin November 2025 RESEARCH SERIES NUMBER 222 Available to download from www.esri.ie https://doi.org/10.26504/rs222  2025 The Economic and Social Research Institute Whitaker Square, Sir John Rogerson’s Quay, Dublin 2. This Open Access work is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. ABOUT THE ESRI The Economic and Social Research Institute (ESRI) advances evidence-based policymaking that supports economic sustainability and social progress in Ireland. ESRI researchers apply the highest standards of academic excellence to challenges facing policymakers, focusing on ten areas of critical importance to 21st century Ireland. The Institute was founded in 1960 by a group of senior civil servants led by Dr T.K. Whitaker, who identified the need for independent and in-depth research analysis. Since then, the Institute has remained committed to independent research and its work is free of any expressed ideology or political position. The Institute publishes all research reaching the appropriate academic standard, irrespective of its findings or who funds the research. The ESRI is a company limited by guarantee, answerable to its members and governed by a Council, comprising up to 14 representatives drawn from a crosssection of ESRI members from academia, civil services, state agencies, businesses and civil society. Funding for the ESRI comes from research programmes supported by government departments and agencies, public bodies, competitive research programmes, membership fees, and an annual grant-in-aid from the Department of Public Expenditure, Infrastructure, Public Service Reform and Digitalisation. Further information is available at www.esri.ie THE AUTHORS Helen Russell and Emer Smyth are Research Professors at the Economic and Social Research Institute (ESRI) and Adjunct Professors at Trinity College Dublin (TCD). Eva Slevin is a Research Assistant at the ESRI. ACKNOWLEDGEMENTS This publication was funded by the Department of Children, Disability and Equality (DCDE) through a joint research partnership with the ESRI. The views, opinions, findings, conclusions and/or recommendations expressed here are strictly those of the authors. They do not necessarily reflect the views of the DCDE, which takes no responsibility for any errors or omissions in, or for the accuracy of, the information contained in this publication. It is presented to inform and stimulate wider debate among the policy community, and among academics and practitioners in the field. We are grateful to members of the programme steering group (Ciara Pidgeon, Nuala Connolly, Aisling Murray and Morag Henderson) for their support and constructive comments on the report. We are also grateful to the editor of the Research Series (Anne Nolan), an external reviewer and two internal reviewers for the comments provided. As always, this work would not be possible without the contribution of the former Growing Up in Ireland (GUI) team, the young people and their families. The report draws on Growing Up in Ireland data. GUI is funded by the Government of Ireland. GUI is managed as a partnership between the DCDE and the Central Statistics Office (CSO). The CSO is responsible for the survey element of GUI. Results in this report are based on analyses of data from Research Microdata Files provided by the CSO. Neither the CSO nor the DCDE take any responsibility for the views expressed in, or the outputs generated from, these analyses. This report has been accepted for publication by the Institute, which does not itself take institutional policy positions. All ESRI Research Series reports are peer reviewed prior to publication. The authors are solely responsible for the content and the views expressed. iv | Afterschool Care for School-Aged Children in Ireland TABLE OF CONTENTS List of tables ....................................................................................................................................................... v List of figures ..................................................................................................................................................... vi Abbreviations ................................................................................................................................................... vii Glossary ........................................................................................................................................................... viii EXECUTIVE SUMMARY .................................................................................................................................... ix CHAPTER 1: INTRODUCTION ............................................................................................................................ 1 1.1 Previous Research ................................................................................................................................. 1 1.2 Afterschool Care in Ireland and Europe ................................................................................................ 4 1.3 Outcomes for Children in Afterschool Care .......................................................................................... 8 1.4 Key Policies for Afterschool Care in Ireland ........................................................................................ 10 1.5 Data and Methods............................................................................................................................... 14 1.6 Outline of the Report .......................................................................................................................... 18 CHAPTER 2: PATTERNS OF AFTERSCHOOL CARE: A CROSS-COHORT COMPARISON ........................................ 19 2.1 Patterns of Afterschool Care at Age 9 and Age 13 .............................................................................. 19 2.2 What Factors Influence the Use of Group Afterschool Care at Ages 9 and 13? ................................. 23 2.3 Predictors of All Care Types at Age 9 .................................................................................................. 32 2.4 Conclusions ......................................................................................................................................... 35 CHAPTER 3: AFTERSCHOOL CARE AND ACTIVITIES AMONG 9AND 13-YEAR-OLDS ........................................ 37 3.1 Introduction ........................................................................................................................................ 37 3.2 Friendships and Peer Interaction ........................................................................................................ 37 3.3 Organised Sports and Cultural Activities ............................................................................................. 39 3.4 Screentime .......................................................................................................................................... 43 3.5 Conclusions ......................................................................................................................................... 43 CHAPTER 4: AFTERSCHOOL CARE AND COGNITIVE AND SOCIO-EMOTIONAL OUTCOMES AT AGE 13 ............. 48 4.1 Introduction ........................................................................................................................................ 48 4.2 Cognitive Outcomes ............................................................................................................................ 48 4.3 Socio-Emotional Outcomes ................................................................................................................. 52 4.4 Conclusions ......................................................................................................................................... 54 CHAPTER 5: CONCLUSIONS AND POLICY IMPLICATIONS ................................................................................ 56 5.1 Patterns of Afterschool Care at Age 9 and Age 13 .............................................................................. 56 5.2 Afterschool Care and Activities Among 9and 13-Year-Olds .............................................................. 57 5.3 Afterschool Care Type and Socio-Emotional and Cognitive Outcomes .............................................. 58 5.4 Policy Implications............................................................................................................................... 59 5.5 Future Research .................................................................................................................................. 61 REFERENCES ................................................................................................................................................... 62 APPENDIX I .................................................................................................................................................... 68 List of tables and figures | v LIST OF TABLES Table 1.1 Estimated capacity of childcare for school-age children 2024 (Pobal) ............................................ 6 Table 1.2 Policy developments for school-age childcare from 2000 ............................................................. 11 Table 1.3 Breakdown of Growing Up in Ireland waves used in this report ................................................... 16 Table 2.1 Main term-time care type at age 9 and age 13 by cohort ............................................................. 20 Table 2.2 Hours of any care per week at age 9 (childminder, afterschool club/centre or relative) .............. 20 Table 2.3 Mean hours of care per week at age 9 by care type and cohort ................................................... 21 Table 2.4 Mean hourly care costs for parents at age 9 by care type and cohort .......................................... 21 Table 2.5 Comparison of urban and rural afterschool care costs (at age 9) .................................................. 22 Table 2.6 Principal-rated adequacy of afterschool facilities .......................................................................... 22 Table 2.7 Principal-rated adequacy of afterschool facilities by school DEIS status ....................................... 23 Table 2.8 Care type at age 9 by principal-rated adequacy of afterschool facilities (pooled across cohorts) 23 Table 2.9 Logit model of participation in afterschool centre/club at age 9 and age 13 (average marginal effects) ........................................................................................................................................... 25 Table 2.10 Participation in afterschool care at age 9, controlling for care type at age 5: Cohort ’08 only (average marginal effects) ............................................................................................................. 31 Table 2.11 Model of care type at age 9: Multinomial relative risk ratios ........................................................ 34 Table 3.1 Ordered logit models of the relationship between care provision, number of friends and frequency of seeing friends among 9-year-olds (odds ratios) ....................................................... 38 Table 3.2 Ordered logit models of the relationship between care provision, number of friends and number of close friends among 13-year-olds (odds ratios) ......................................................................... 39 Table 3.3 Logit models of the relationship between care provision and involvement in a sports club among 9-year-olds (average marginal effects) .......................................................................................... 40 Table 3.4 Models of the relationship between care provision and frequency of reading and involvement in structured cultural activities among 9-year-olds ........................................................................... 41 Table 3.5 Logit models of the relationship between care provision and involvement in a sports club among 13-year-olds (average marginal effects) ........................................................................................ 42 Table 3.6 Ordered logit models of the relationship between care provision and frequency of reading among 13-year-olds (odds ratios) .................................................................................................. 42 Table 3.7 Models of the relationship between care provision and involvement in structured cultural activities among 13-year-olds (average marginal effects) ............................................................. 43 Table 3.8 Logit models of the relationship between care provision and the 9-year-old having their own mobile phone (average marginal effects) ...................................................................................... 44 Table 3.9 Ordered logit models of the relationship between care provision and amount of time spent watching TV among 9-year-olds (odds ratios) ............................................................................... 44 Table 3.10 Ordered logit models of the relationship between care provision and amount of time spent online among 9-year-olds (odds ratios) ......................................................................................... 45 Table 3.11 Ordered logit models of the relationship between care provision and time spent watching TV among 13-year-olds (odds ratios) .................................................................................................. 45 Table 3.12 Ordered logit models of the relationship between care provision and amount of time spent computer gaming among 13-year-olds (odds ratios) ..................................................................... 46 Table 3.13 Ordered logit models of the relationship between care provision and amount of time spent online among 13-year-olds (odds ratios) ....................................................................................... 46 Table 4.1 OLS regression model of Cohort ’98 reading scores at age 13....................................................... 50 vi | Afterschool Care for School-Aged Children in Ireland Table 4.2 OLS regression model of Cohort ’08 cognitive scores at age 13 .................................................... 51 Table 4.3 Cohort ’98 maths scores at age 13 ................................................................................................. 51 Table 4.4 Cross-cohort comparison of SDQ scores at age 13 (OLS regression coefficients) .......................... 52 Table 4.5 SDQ subscales – cross-cohort comparison ..................................................................................... 54 Table A2.1 Participation in afterschool care at age 9 (odds ratios) ................................................................. 68 Table A3.1 Frequencies of involvement in activities among 9-year-olds, Cohorts ’98 and ’08 ....................... 70 Table A3.2 Frequencies of involvement in activities among 13-year-olds, Cohorts ’98 and ’08 ..................... 71 Table A3.3 Ordered logit models of the relationship between care provision, number of friends and frequency of seeing friends among 9-year-olds, with interaction terms between cohort and provision type (odds ratios) ........................................................................................................... 72 Table A3.4 Ordered logit models of the relationship between care provision, number of friends and number of close friends among 13-year-olds, with interaction terms between cohort and provision type (odds ratios) ................................................................................................................................... 72 Table A3.5 Logit models of the relationship between care provision, involvement in sports clubs, reading and cultural activities among 9-year-olds, with interaction terms between cohort and provision type (odds ratios) ........................................................................................................................... 73 Table A3.6 Logit models of the relationship between care provision and involvement in sports clubs and cultural activities among 13-year-olds, with interaction terms between cohort and provision type (odds ratios) ................................................................................................................................... 73 Table A3.7 Logit models of the relationship between care provision, mobile phone ownership and time spent on TV and other online activities among 9-year-olds, with interaction terms between cohort and provision type (odds ratios) ......................................................................................... 74 Table A3.8 Ordered logit models of the relationship between care provision and time spent on TV, computer gaming and other online activities among 13-year-olds, with interaction terms between cohort and provision type (odds ratios) .......................................................................... 74 Table A4.1 Cohort ’98 reading scores at age 13............................................................................................... 75 Table A4.2 Cohort ’08 reading scores at age 13............................................................................................... 76 Table A4.3 Cohort ’98 maths outcomes at age 13 ........................................................................................... 77 Table A4.4 Factors influencing SDQ scores at age 13: Cross-cohort analysis .................................................. 78 LIST OF FIGURES Figure 2.1 Probability of afterschool centre/club participation at age 9: Cohort by social class interaction . 27 Figure 2.2 Probability of afterschool centre/club participation at age 9: Cohort by income quintile interaction ...................................................................................................................................... 28 Figure 2.3 Probability of afterschool centre/club participation: Cohort by mother’s employment interaction ...................................................................................................................................... 29 Figure 2.4 Probability of afterschool centre/club participation: Cohort by family structure interaction ....... 30 Abbreviations | vii ABBREVIATIONS AME Average Marginal Effects CCSP Community Childcare Subvention Plus CPI Consumer Price Index DCDE Department of Children, Disability and Equality (2025–present) DCEDIY Department of Children, Equality, Disability, Integration and Youth (2020–2025) DCYA Department of Children and Youth Affairs (2011–2020) DEIS Delivering Equality of Opportunity in Schools DoHC Department of Health and Children ECCE Early Childhood Care and Education ELC Early Learning and Care ESRI Economic and Social Research Institute EU European Union GUI Growing Up in Ireland LLC Long-Lasting Condition NCS National Childcare Scheme OECD Organisation for Economic Co-operation and Development OLS Ordinary Least Squares RMF Research Microdata File RRR Relative Risk Ratio SAC School-Age Childcare SCP School Completion Programme SDQ Strengths and Difficulties Questionnaire TCD Trinity College Dublin viii | Afterschool Care for School-Aged Children in Ireland GLOSSARY Afterschool care – refers to non-parental care arrangements for school-aged children during term time. It includes care by childminders, relative care, care in an afterschool centre/club and care by sibling/self. Afterschool centre/club – refers to formal afterschool care during term time that occurs in a group setting as reported by the primary caregiver. Childminder care – refers to paid care provided by a childminder in the child’s own home or in the childminder's home as reported by the primary caregiver. Relative care – refers to care provided by a relative or friend either in the child’s home or in the home of the relative/friend as reported by the primary caregiver. This includes paid and unpaid care. School-age childcare – is defined in Irish legislation (Childcare Support Act 2018) as any service that caters to children under the age of 15 years enrolled in a school providing primary or post-primary education, and provides a range of activities that are developmental, educational and recreational that take place outside of school hours. The definition includes term-time and holiday services, and excludes services that solely provide activities relating to the arts, youth work, competitive/recreational sport, tuition or religious teaching. 4 | Afterschool Care for School-Aged Children in Ireland central to children’s preferences for what they do after school, centring this principle in the provision of afterschool care becomes crucial for their wellbeing and enjoyment (Kane, 2016). 1.2 AFTERSCHOOL CARE IN IRELAND AND EUROPE 1.2.1 Afterschool care in Ireland Prior to the last decade the school-age childcare system in Ireland was characterised by a low level of provision and a lack of standardisation and regulation (Byrne, 2016; Indecon, 2021). Significant policy developments have taken place in recent years, though many of the changes occurred after the data collection point for the 9-year-olds in Cohort ’08 (2016–2017) and Cohort ’98 (2007), on which most of the analysis in this report is based. On foot of the Action Plan on School Age Childcare (Department of Children and Youth Affairs (DCYA), 2017) regulations on the registration of school-age childcare services were introduced in 2019,4 while guidelines on quality were published in 2020 (DCYA, 2020). A consultation is currently underway to inform the development of comprehensive school-age care regulations to replace the 2019 regulations (Government of Ireland, 2025). The policy landscape is discussed in detail in Section 1.4 below. In Census 2022, one-third (331,783) of all children (under 15 years old) in the State were in some form of childcare (Central Statistics Office, 2023). The figure for those aged 5–12 was 30 per cent, while 7 per cent of 13and 14-year-olds were in some form of childcare. Of those aged 5–12 in childcare,5 most were either cared for by an unpaid relative (32.9 per cent) or in a formal playgroup or afterschool group (32.6 per cent). A further 6.4 per cent were cared for by a paid relative, and 24.8 per cent used a childminder (in either the childminder’s home or the child’s home). The remainder stated no childcare or an ‘other’ option. Of the 5–12-year-olds, most were in care for between 1 and 10 hours each week (50.9 per cent). Within this, 32.2 per cent were cared for by an unpaid relative, while 34.9 per cent were in formal childcare (crèche, Montessori, playgroup or afterschool), with the remainder in childminder or paid relative care. Very few children who were in childcare were there for more than 21 hours per week, at 18 per cent of children aged 5–12 (Central Statistics Office, 2023). The Ipsos B&A telephone survey of 771 parents and guardians (Loscher and Mulcahy, 2025) found that most school-going children (in primary school) were cared for by a parent after school during the school term (56 per cent). A further 22 per cent of primary school children attended an afterschool or school-age childcare service. The average number of weekly childcare hours that parents 4 S.I. No. 575/2018 – Child Care Act 1991 (Early Years Services) (Registration of School Age Services) Regulations 2018. 5 This includes 5-year-olds who had not yet started school. Chapter 1 | 5 reported for school-going children after school was 10.3 for formal afterschool childcare, 10.4 for childminder, and 10.5 for grandparent. This aligns with the statistics reported in Census 2022. There is little available data that indicate the quality of afterschool programmes in Ireland (Byrne, 2016; Indecon, 2021), which would be useful in assessing the choice of afterschool provision. The National Action Plan for Childminding (Department of Children, Equality, Disability, Integration and Youth (DCEDIY), 2021a) notes that childminding in Ireland suffers from a lack of regulation, with very low registration with Tusla, the Child and Family Agency.6 This brings a risk of low-quality service, which Eurofound (2020) identifies as affecting take-up of afterschool care in a country. New childminding regulations came into force in 2024, alongside legislative amendments to the Child Care Act 1991, which provide for enhanced enforcement measures for childcare services. This should increase the quality of afterschool care in Ireland and is currently in a transitional phase. Pobal7 (2025) has created a set of dashboards which give some indication of service provision within the school-age childcare system. The figures refer to services that received support under the NCS or the Community Childcare Subvention Plus (CCSP). Of services offering school-age childcare in 2023–2024,8 over a quarter (27 per cent) offered only afterschool services, while the remainder also offered early learning and care (ELC). Of services only providing school-age care,9 64.1 per cent had their premises located in a school. School-based services are not directly provided by the schools but by private or community providers.10 While these private providers do not represent a direct integration with the school system, their presence on the school grounds creates a better link between in-school care and out-of-school care, meaning that parents do not have to find a way of getting their child from school to afterschool care. There is also scope for state-led facilities to be developed in school premises in future (Department of Children, Disability and Equality (DCDE), 2025c). Among providers that only offered school-age services, over three-quarters were established since 2016, with 51 per cent established since 2021. This is likely to reflect the impact of the National Childcare Scheme. 6 However, new regulation of childminders was introduced in late 2024 and is being rolled out at the time of writing. 7 Pobal is a state-sponsored agency that manages funding and provides support for programmes in the areas of social inclusion and employment, and early learning and care. The dashboards here refer to services receiving at least one DCDE funding contract, and ‘service characteristics’ data refer to those who completed the Annual Early Years Sector Profile survey. 8 Services offering both early learning care (ELC) and school-age childcare (SAC): 1,728. Services only offering SAC: 651. Services only offering ELC: 2,235. Figure calculated here refers to the number offering SAC only divided by the total offering SAC, including those offering ELC and SAC together. This figure refers to all services with a DCDE funding contract. 9 N = 557. Note that this figure is different from the total number offering SAC only; the former figure refers to all those with a DCDE funding contract, while ‘service characteristics’ refers to those responding to the Annual Early Years Sector Profile survey. 10 Of 557 providers responding to the survey, 127 were community providers (39 per cent located in schools) and 430 were private providers (72 per cent located in schools). 6 | Afterschool Care for School-Aged Children in Ireland Among services that provided both school-age and ELC places the picture was somewhat different, with 13.6 per cent located in schools. Most were established pre-2016, with 10 per cent established after 2021. The average hourly wage for school-age childcare (SAC) staff in 2023/2024 was €15.43, but was lower for those working directly with the children (€14.42). Of 2,320 staff members, 48.4 per cent listed a qualification, 30.1 per cent had a level 5 or 6 qualification, and 51.6 per cent had no relevant qualification. This does not indicate SAC-specific qualifications, as there is no formal qualification for school-age care. The cost of early years childcare in Ireland is high compared to the OECD average and other European Union (EU) countries (Doorley et al., 2021; OECD, 2025), though there have been significant improvements with the introduction of the free part-time pre-school entitlements (through the Early Childhood Care and Education (ECCE) scheme) and the National Childcare Scheme (NCS). The NCS covers children under the age of 15 attending registered services. The Department of Children, Disability and Equality notes that the total of number of children who received a subsidy for school-age care under the NCS in the calendar year 2024 was 110,738 (DCDE, personal communication). Pobal dashboard figures provide a breakdown of the age of school-going children enrolled in DCDE-funded places in May/June 2024 (Table 1.1).11 These figures also highlight substantial unmet demand for places, as well as a significant number of vacant places, suggesting a mismatch between supply and demand within local areas. TABLE 1.1 ESTIMATED CAPACITY OF CHILDCARE FOR SCHOOL-AGE CHILDREN 2024 (POBAL) Age Range Enrolled Vacant places Waiting list 4+ to 5 years school going 10,970 3,081 5,001 5+ to 6 years school going 19,307 3,234 5,040 6+ to 8 years 23,972 2,953 4,231 8 years+ 24,934 2,621 2,988 Total 79,183 11,889 17,260 Source: Pobal dashboards 2025 www.pobal.ie/childcare/capacity/. See footnote 11 for further details. Additional afterschool provision is supported through the Delivering Equality of Opportunity in Schools (DEIS)12 programme, including through the School Completion Programme (SCP). SCP projects cover 467 primary schools. Approximately 23 per cent of primary schools in the programme provide afterschool clubs (Smyth et al., 2025).13 These programmes are seen as important 11 The figures relate to services with a contract for at least one DCDE funding scheme. They are based on responses to the Annual Early Years Sector Profile survey, which Pobal have reweighted to correct for non-response. The survey has a response rate of 87 per cent. The figures relate to May/June 2024. 12 DEIS is a programme of supports for schools that serve disadvantaged populations. Urban Band 1 DEIS schools have a greater concentration of pupils from disadvantaged backgrounds than those in Urban Band 2. 13 It should be noted that SCP provision is designed to promote school engagement rather than provide care per se. Chapter 1 | 7 for providing evidence-based educational interventions as well as offering a chance for staff to interact with the parents of at-risk children (ibid). Pobal (2025) reports that the median hourly fee (before subsidies) for school-age childcare services is €5 during term time. This varies from €6.26 in Dún Laoghaire– Rathdown to €4 in Monaghan. Most children attend these services for up to 10 hours per week, which means parents are paying a median of approximately €50 weekly. Comparative data on the cost of afterschool childcare in Ireland are limited. In 2025, a survey of parents (Loscher and Mulcahy, 2025) found the average monthly cost for afterschool centre-based services was €262 per child. For parents using any form of afterschool childcare, 22 per cent found it difficult to afford. Additionally, 17 per cent of parents who did not use afterschool care said this was because of the cost. Furthermore, there was a lack of awareness among parents as to the availability of financial supports from the National Childcare Scheme, with 69 per cent saying they had no awareness that the NCS provides financial support towards the cost of registered childminders. More generally, 44 per cent of parents and guardians surveyed were not aware of financial support towards the cost of registered childcare generally. This suggests a higher awareness of subsidies for afterschool centres. An independent review of the operating model of school-age care and ELC in Ireland (Indecon, 2021) found that the current operating model needs to be updated. Concerns were raised regarding fragmentation and duplication of decision-making, creating confusion and administrative burden for providers as well as an insufficient level of engagement with parents as to their needs. This review recommended the creation of a new statutory agency, which would address the key issues in the current system. 1.2.2 Afterschool care and provision in Europe The provision of afterschool care and the policies underpinning it vary widely across Europe. Research on this topic is limited, and available comparative research (Plantenga and Remery, 2013; 2017) is outdated. Eurofound (2020) provides the most recent review of afterschool care in Europe, examining the period before the COVID-19 pandemic.14 The pandemic brought about a persistent increase in working from home, which may allow parents of older children to manage without any additional afterschool care.15 Eurofound (2020) identified key drivers of and barriers to afterschool care provision and take-up across Europe. Affordability and availability of adequate hours were key drivers of afterschool care take-up by parents. The report identified a gap between average working hours and formal childcare hours (including school hours), meaning that there was a disconnect between the requirements of parents and state provision. In Ireland 14 This is relevant given the pandemic’s effect on flexible working and the provision of afterschool care during lockdowns. The findings of this report may not hold up in the current year (2025) for some countries. 15 Alamir et al. (2024) outline the significant increase in working from home in Ireland preand post-pandemic. 8 | Afterschool Care for School-Aged Children in Ireland this was listed as approximately 7 hours per week. This gap was lowest (<3 hours) for Sweden, Denmark, Luxembourg, Italy and Portugal (Eurofound, 2020). Sweden, Denmark and Portugal, particularly, have formalised co-ordinated afterschool care driving this trend, as all-day coverage is provided for parents (Plantenga and Remery, 2013). However, even in these systems, flexibility remains an issue with parents who do not have set hours, and services are closed outside of term time. The highest gaps between provision and parental needs were in Romania (>20 hours), Slovakia (>15 hours) and Finland (>12 hours) (Eurofound, 2020). Barriers identified by Eurofound included regional differences in provision and costs, including in Ireland (Eurofound, 2020; Indecon, 2021). Italy and Croatia were highlighted as systems with strong regional inequalities in provision. Eurofound (2020) also noted that a key challenge in researching afterschool care across Europe is the co-existence of formal and informal childcare. It is hard to capture those who would use formal childcare if barriers were reduced, though take-up in countries where there is universal provision (such as Sweden or Denmark) suggests there would be increased participation in formal afterschool care if financial barriers were entirely removed. Additionally, Plantenga and Remery (2013) identified issues in comparing afterschool care across Europe, particularly in a lack of disaggregated data by care type, as well as the lack of a formal framework for assessing service quality between countries. They noted little regulation in many European countries regarding child-staff ratios, education level of staff, and group size. Since that report in 2013, progress has been made. In Ireland, the National Quality Guidelines for School Age Childcare Services were published in 2020 (see Section 1.4 below), and new regulations were introduced in 2019. There are important policy developments across Europe that could inform future decision-making in Ireland. These include an aim in the 2018 German coalition agreement to introduce the legal right to out-of-school childcare for primary school children. This will be in place from the 2026 school year.16 This expands an existing provision for pre-primary school children of the right to early childhood education and care (Eurofound, 2020). Several countries have developed policies which extend school hours towards ‘all-day’ schools, including Austria, Denmark, France, Hungary, Latvia, Lithuania and Portugal. 1.3 OUTCOMES FOR CHILDREN IN AFTERSCHOOL CARE Afterschool care can have differing outcomes for children depending on the specific characteristics of the service/setting and of the families and children. Much existing research on this topic has focused on meta-analyses of existing studies in the US context. Scott-Little et al. (2002) identified 23 studies from 1996–2001 16 www.bundesregierung.de/breg-en/service/archive/all-day-care-at-primary-schools-to-be-expanded-1911210 Chapter 1 | 9 which evaluated services for school-aged children between kindergarten (5 years old) and grade 12 (18 years old). Data were collected on student outcomes in 15 of these studies, but most suffered from design or reporting problems. Only six studies included effect sizes, using cognitive or psychological scores. These studies saw that programme participation was positively correlated with achievement on standardised maths, reading and language tests. One cited study found that targeted programmes closed the gap in maths performance between nonparticipants and participants, and another found that programme participants with the lowest prior cognitive scores saw the greatest improvement. Lauer et al. (2006) examined the quality and findings of 35 studies concerning outof-school-time programmes17 for school-aged at-risk students between 1985 and 2003, specifically examining the effectiveness of programmes for at-risk students. This analysis explored effect sizes between studies, finding a positive impact of afterschool or summer programmes for at-risk students. The controls for these studies were at-risk students not involved in these programmes. The effectiveness of the programmes varied by different characteristics. The use of tutoring in afterschool programmes was beneficial, and grade level had a significant association with performance. More recently, Christensen et al. (2023) published work examining effect sizes of 56 studies on the outcomes of afterschool care on marginalised young people.18 This research found a small positive effect of out-ofschool care on youth outcomes for this group, but gains were seen overall rather than in sub-groups. They also found that youthor teacher-reported outcome measures presented greater effect sizes than official reports, reflecting the importance of direct student engagement in this research context. In the Irish context, research using Growing up in Ireland (GUI) data has found limited effects of afterschool care on children’s performance and socio-emotional outcomes (Byrne, 2016; Byrne and O’Toole, 2015; Russell et al., 2016). Afterschool care was found to matter for reading in high-achieving students, where centrebased care had a negative relationship with reading scores for the highest quintiles of reading performance (Byrne and O’Toole, 2015). This research finds that while afterschool care is associated with worse cognitive or socio-emotional outcomes, this relationship is largely due to selection, as it becomes non-significant once family background or child characteristics are added (Byrne, 2016; Byrne and O’Toole, 2015). Byrne (2016) also notes that the GUI data are limited by not providing detailed information on afterschool care type, such as pedagogy or public/private provision. Furthermore, research at the time was limited by the low 17 Defined as an education intervention occurring outside of school hours (Lauer et al., 2006). Note that this differs from the definition used in Ireland. 18 Christensen et al. (2023) and Lauer et al. (2006) refer to at-risk youth and marginalised youth. In both cases this refers to young people with lower test scores or with characteristics associated with earlier school leaving or issues in school, such as socio-economic status or race. 10 | Afterschool Care for School-Aged Children in Ireland take-up of afterschool care, and low levels of provision more broadly, which prevent more detailed results on outcomes. Another type of outcome considered in research is health outcomes, which Byrne and O’Toole (2015) found had significant associations with childcare placements. Children in non-parental care were rated as less healthy across all age groups in this study (9 months (GUI ’08 Cohort Wave 1), 3 years (GUI ’08 Cohort Wave 2) and 9 years (GUI ’98 Cohort Wave 1)). Centre-based care was particularly associated with lower health outcomes in early childhood, while relative care was associated with lower health outcomes in middle childhood. However, little evidence is found that indicates childcare in one life stage will impact on health outcomes in the next, while stability of childcare across life stages is associated with more positive outcomes. Qualitative research asking parents about their child’s development has found that parents and staff perceive a positive impact on child wellbeing from afterschool care (Lacey, 2016; Russell et al., 2016). This was seen to stem from positive interactions between children and staff, the development of friendships and the variety of activities provided in afterschool care environments. Russell et al. (2016) additionally found that being cared for by relatives at age 3 led to lower socioemotional difficulties at age 5 (as reported by parents and teachers). Children at age 3 who had attended centre-based care were rated by teachers as having greater socio-emotional difficulties at age 5; this included hyperactivity and concentration. Parents agreed that their children had marginally higher conduct problems if they had attended centre-based care, but rated their emotional and friendship issues as lower. While these findings apply to early childhood education, they may indicate similar trends could be observed in school-aged children. 1.4 KEY POLICIES FOR AFTERSCHOOL CARE IN IRELAND This section provides an overview of relevant policies for childcare in Ireland. While it includes recent policies, it also addresses key policies that existed at the time of the Growing Up in Ireland cohort studies at 9 years old, which took place when Cohort ’98 turned 9 in 2007–2008, and Cohort ’08 turned 9 in 2017–2018. The policy context is relevant when considering take-up and outcomes from afterschool care (Gambaro et al., 2015; Morrissey, 2017). The policies discussed below must be seen in the wider economic context of the period, with rising employment during the Celtic Tiger era, followed by the economic downturn and high unemployment during the financial crash, followed by the COVID-19 pandemic in 2020–2021. A commitment to ‘expanding the provision of afterschool care’ has been included in each Programme for Government since 2016 (Department of the Taoiseach, 2016; 2020; 2025). This has led to a more substantive level of policymaking than was the case prior to 2016. The foundation of childcare legislation in Ireland is the Chapter 1 | 11 Child Care Act 1991, with subsequent key policy developments identified in Table 1.2. TABLE 1.2 POLICY DEVELOPMENTS FOR SCHOOL-AGE CHILDCARE FROM 2000 Year Development 2000 Publication of the National Children’s Strategy (DoHC) Equal Opportunities Childcare Programme introduced 2001 Child Care (Amendment) Act 2001 (amending Child Care Act 1991) Establishment of the National Children’s Office 2003 Foundation of the Family Support Agency Appointment of the Children’s Ombudsman 2008 Launch of Community Childcare Subvention Scheme 2011 Establishment of the Department of Children and Youth Affairs (DCYA) 2012 The Children’s Referendum on a constitutional amendment to strengthen children’s rights 2013 Launch of After School Childcare Scheme 2014 Publication of Better Outcomes, Brighter Futures: The National Policy Framework for Children and Young People 2014–2020 (DCYA) Establishment of The Child and Family Agency (Tusla) 2015 Establishment of the Inter-Departmental Group on Future Investment in Early Years and School-Age Care and Education (DCYA) 2016 Publication of Houses of the Oireachtas Joint Committee on Health and Children’s Report on Affordable and Quality Childcare 2017 Publication of Action Plan on School Age Childcare (DCYA) 2018 Childcare Support Act 2018 Publication of First 5 – Strategy for Babies, Young Children and Families (2019–2028) 2019 Establishment of National Childcare Scheme Child Care (Amendment) Act 2019 2020 Publication of National Quality Guidelines for School Age Childcare Services 2021 Publication of National Action Plan for Childminding (2021–2028) Publication of Partnership for the Public Good: A New Funding Model for Early Learning and Care and School-Age Childcare Publication of Nurturing Skills: The Workforce Plan for Early Learning and Care and SchoolAge Childcare 2022–2028 2022 Launch of Together for Better, new funding model for early learning and care and schoolage childcare 2023 Publication of Young Ireland: National Policy Framework for Children and Young People 2024 Launch of Equal Start Programme Publication of Childminding Regulations Child Care (Amendment) Act 2024 Source: Government of Ireland (various years). Prior to the economic crash in 2008, limited policy had been enacted to regulate and provide for childcare in the State (Barry and Sherlock, 2008). Irish provision was below EU targets for school-age childcare, and there was no standardised definition of quality in childcare provision. The National Children’s Strategy (Department of Health and Children (DoHC), 2000) was published in 2000, providing a ten-year action plan to improve children’s lives in Ireland. This included reducing child poverty as well as providing quality childcare services. This strategy provided for a framework for qualification and accreditation in the childcare 12 | Afterschool Care for School-Aged Children in Ireland sector, and the establishment of County and City Childcare Committees. This created a decentralised environment for childcare provision, with 33 committees established the following year. Funding for this strategy and the Equal Opportunities Childcare Programme was established in the National Development Plan (2000–2006) (Government of Ireland, 1999). However, these funds were for capital expenditure, so while the funding increased building of childcare centres, it did not provide for training and accreditation of staff. Childcare in Ireland has developed substantively since the 1991 Child Care Act, which enshrined the regulation of early childhood education and care services. This section discusses developments occurring during the 2000s. In the early 2000s, several key organisations were established to improve the rights and welfare of children in Ireland. These include the Family Support Agency (now amalgamated into Tusla), the National Children’s Office, and the Office of the Children’s Ombudsman. Additional funding was provided to parents of young children under the age of 6 through the Early Childcare Supplement (McGinnity et al., 2015) between 2006 and 2009, and capital grants were provided to encourage private or community childcare provision by services. The National Development Plan (2000– 2006) also provided for funding for childcare facilities. In 2010 the free pre-school year was introduced, though costs of childcare for families in Ireland remained among the highest in the EU. Other targeted schemes established at this time included the Community Childcare Subvention Scheme and the After School Childcare Scheme.19 In 2011 the Department of Children and Youth Affairs was first established, which included in its aims to support the unified provision of affordable childcare in the State (DCYA, 2012). Then in 2014 Tusla (the Child and Family Agency) became an independent legal entity which amalgamated the HSE Children and Family Services, The Family Support Agency and the National Educational Welfare Board. In the same year, the National Policy Framework for Children and Young People 2014– 2020 was published. This policy framework focused on children’s early years, including care and education, but referred to afterschool services in the context of additional supports for disadvantaged children and reducing barriers to employment. In 2015 and 2016, an inter-departmental group on Future Investment in Early Years and School-Age Care and Education was established and a report on Affordable and Quality Childcare was released. The former used data from the Growing Up in Ireland study to evaluate needs in the sector. Its report noted that just 3 per cent of children were in centre-based care at 9 years old. It also noted that Irish childcare fees were one of the highest in the OECD as a percentage of the average family wage. The report from this group also noted the importance of afterschool care to maternal labour force participation. 19 Most subsidies for parents were for young children and the number of beneficiaries of the After School Childcare Scheme was small (only 800 in 2014, www.oireachtas.ie/en/debates/question/2014-07-15/403/). Chapter 1 | 13 The Action Plan on School Age Childcare was published in 2017, which established the groundwork for quality guidelines for school-age childcare introduced in 2020. In 2018, First 5 – A Whole-of-Government Strategy for Babies, Young Children and their Families (2019–2028) was published (Government of Ireland, 2018). This committed to stronger regulation and registration of paid childminders, and to undertaking additional research on early learning and care, including school-age childcare. It also committed to making school-age childcare more affordable. Preliminary regulations were established in 2019 that required all school-age childcare services to register with Tusla (DCYA, 2019). The Childcare Support Act 2018 defines school-age childcare as any service that caters to children under the age of 15 years enrolled in a school providing primary or post-primary education, and provides a range of activities that are developmental, educational and recreational that take place outside of school hours. The definition includes termtime and holiday services, and excludes services that solely provide activities relating to the arts, youth work, competitive/recreational sport, tuition or religious teaching. The National Childcare Scheme was established in 2019 as a result of the Childcare Support Act 2018. In particular, this provided financial support for childcare and early learning costs for children between the ages of 24 weeks and 15 years. The National Childcare Scheme replaced previous childcare programmes with a single scheme and a centralised application portal (Government of Ireland, 2025). There are two subsidies that can be applied for. The first is a universal subsidy providing €2.14 per hour for a maximum of 45 hours per week towards childcare in a registered provider for children aged over 6 months and up to 15 years old.20 The second is income-assessed, and varies based on characteristics such as family income, the age of the child and their educational stage, and the number of children in the family. The universal scheme was extended to all children under 15 years old in 2022. In 2020, the National Quality Guidelines for School Age Childcare Services were published by the Department of Children and Youth Affairs (DCYA, 2020). They emphasised that school-age childcare should be underpinned by the principles of children’s right to a voice in decision-making that affects them, as well as their right to play. The guidelines cover areas such as leadership and governance, staff and professional practice, the environment, health and wellbeing, activities, and partnerships with families and communities. The National Action Plan for Childminding (2021–2028) also establishes a pathway to improve regulation, support and subsidies for paid childminders, extending state support on a phased basis by 2028. Its core objectives are to increase the take-up of subsidies, improve quality assurance, provide for greater recognition and regulation of childminders, develop resources to support childminding provision (including financial incentives 20 Figures for 2023–2024. 20 | Afterschool Care for School-Aged Children in Ireland TABLE 2.1 MAIN TERM-TIME CARE TYPE AT AGE 9 AND AGE 13 BY COHORT Care Type Age 9 Age 13 Cohort ’98 (%) Cohort ’08 (%) Cohort ’98 (%) Cohort ’08 (%) Parent 76.4 71.2 86.2 86.3 Self/sibling 0.5 0.7 6.0 8.6 Relative 11.4 13.8 4.0 3.8 Childminder 8.2 9.7 1.8 0.7 Afterschool club or centre 3.4 4.5 2.0 0.5 100.0 100.0 100.0 100.0 Unweighted N 8,562 7,995 7,414 6,555 Source: GUI Research Microdata Files (RMFs). Notes: The age 13 surveys for Cohort ’08 were conducted during the COVID-19 pandemic. This is likely to have influenced the take-up of care options as schools and childcare services were closed for some of the period. ‘Other’ categories are excluded. Columns may add up to just above or below 100 due to rounding. The primary caregiver was also asked to provide information on the hours of care at age 9. No further questions were asked about care hours at age 13. The number of hours spent by 9-year-olds in any non-parental term-time care was relatively low: 5–9 hours per week was the most common pattern in both cohorts (see Table 2.2). For just under half of the group, care was for less than 10 hours per week in both cohorts. The proportion of children being cared for 20 or more hours per week declined from 15 per cent in Cohort ’98 to 9 per cent in Cohort ’08, suggesting a decline in care hours over time. TABLE 2.2 HOURS OF ANY CARE PER WEEK AT AGE 9 (CHILDMINDER, AFTERSCHOOL CLUB/CENTRE OR RELATIVE) Number of hours per week Cohort ’98 (%) Cohort ’08 (%) 1–4 12.2 11.8 5–9 35.0 34.6 10–14 21.6 28.6 15–19 16.1 15.7 20–24 8.2 6.0 25+ 6.9 3.3 Total 100.0 100.0 N 2,051 2,401 Source: GUI RMF Cohort ’98 and Cohort ’08. The mean number of care hours fell from 11.9 to 10.7 hours per week between the cohorts (Table 2.3). Comparing across the different forms of care, mean hours are lowest for afterschool care and highest for childminder. This pattern holds for both cohorts. The average time per week spent in afterschool care fell from 10.7 hours for Cohort ’98 to 9 hours for Cohort ’08. Chapter 2 | 21 TABLE 2.3 MEAN HOURS OF CARE PER WEEK AT AGE 9 BY CARE TYPE AND COHORT Care type Cohort ’98 (hrs/week) Cohort ’08 (hrs/week) Relative 11.6 10.7 Childminder 12.9 11.4 Afterschool club/centre 10.7 9.0 All 11.9 10.7 N 2,051 2,401 Source: GUI RMF Cohort ’98 and Cohort ’08. Parents were also asked to report the cost of term-time care26 for their 9-year-old. Table 2.4 presents the mean hourly cost of care across different care types. It excludes cases where no payment was made, which is more common where relatives provide care. Among those using relative care, 59 per cent provided no payment. The figures for Cohort ’98 have been adjusted for inflation over the tenyear period27 (the Consumer Price Index (CPI) inflation rate of 2.5 per cent for the period). For Cohort ’98 (in 2007) the mean hourly cost of afterschool club/centre care was €6.18; this was lower than the hourly cost of childminder care at €7.33, but not substantially different from that of paid care provided by a relative. Similarly for Cohort ’08 in 2017, childminder care was most expensive at €9.28 per hour, while afterschool club/centre care cost €7.12 per hour on average. There was a 15 per cent increase in the real hourly cost of afterschool centre care over the ten-year period. The rate of increase was even higher for childminders, at 27 per cent. TABLE 2.4 MEAN HOURLY CARE COSTS FOR PARENTS AT AGE 9 BY CARE TYPE AND COHORT Care Type Cohort ’98 Cohort ’08 Mean (€) N Mean (€) N Relative 6.09 295 8.25 269 Childminder 7.33 847 9.28 904 Afterschool club/centre 6.18 273 7.12 383 All 6.81 1,415 8.55 1,556 Source: GUI Cohort ’08 and Cohort ’98 at age 9. Notes: Means are weighted, Ns are unweighted. Excludes cases where there is no payment. There was a 2.5 per cent increase in the CPI from September 2007 to September 2017 (CPI inflation calculator, CSO). Figures for ’98 are adjusted for inflation. The cost of afterschool care differed between urban and rural areas (Table 2.5). The gap in cost was wider for Cohort ’98 than Cohort ’08. The difference between prices in urban and rural areas was widest for childminders in Cohort ‘98 and relatives in Cohort ’08. 26 These figures pre-date the NCS subsidies; they represent the out-of-pocket expenses by parents. 27 The CPI inflation rate for September 2007 to September 2017 was 2.5 per cent (CSO CPI inflation calculator). 22 | Afterschool Care for School-Aged Children in Ireland TABLE 2.5 COMPARISON OF URBAN AND RURAL AFTERSCHOOL CARE COSTS (AT AGE 9) Source: GUI Cohort ’08 and Cohort ’98 at age 9. Notes: Means are weighted. Excludes cases where there is no payment. There was a 2.5 per cent increase in the CPI from September 2007 to September 2017 (CPI inflation calculator, CSO). Figures for ’98 are adjusted for inflation. 2.1.2 Adequacy of school-based afterschool provision As outlined in Chapter 1, choice of care type is influenced by the availability of services. While parents were not asked about the availability of afterschool services in the local area when their child was age 9, some information was collected from school principals about the perceived adequacy of school-based afterschool provision. There was a notable increase in the proportion of principals who rated the afterschool facilities as ‘good’ or ‘excellent’ between the ’98 and ’08 cohorts, suggesting that there had been an increase in quality of provision over time (Table 2.6). However, the perception of adequacy may reflect the quality of facilities as well as the availability. In Cohort ’08, 58 per cent of principals rated the afterschool facilities as ‘good’ or ‘excellent’ compared to 35.5 per cent in Cohort ’98. TABLE 2.6 PRINCIPAL-RATED ADEQUACY OF AFTERSCHOOL FACILITIES Cohort ’98 Age 9 (%) Cohort ’08 Age 9 (%) Poor 40.6 22.4 Fair 23.9 19.8 Good 25.5 42.0 Excellent 10.0 15.8 Total 100.0 100.0 N 7,633 6,973 Source: GUI Cohort ’08 and Cohort ’98 at age 9, School Principals’ Survey. The reported adequacy of afterschool facilities varied across school types. Schools that are part of the DEIS programme can use grants to support afterschool services; such supports can also be provided as part of the School Completion Programme which operates in DEIS and some non-DEIS schools (Smyth et al., 2025; see Chapter 1). In Cohort ’98, principals in urban DEIS schools were more likely to report ‘good/excellent’ afterschool facilities than non-DEIS schools (Table 2.7). However, rural DEIS schools were perceived to have the poorest facilities. A much higher proportion of rural DEIS schools reported ‘good’ or ‘excellent’ facilities in Cohort Care type Cohort ’98 Cohort ’08 Urban (€) Rural (€) Urban (€) Rural (€) Relative 6.50 5.81 6.36 9.16 Childminder 8.11 6.82 9.90 8.81 Afterschool club/centre 6.40 5.93 7.43 6.77 All 7.31 6.43 8.64 8.49 Chapter 2 | 23 ’08 (44 per cent) than in Cohort ’98 (12 per cent); however, they remained the school type with the poorest (perceived) facilities. TABLE 2.7 PRINCIPAL-RATED ADEQUACY OF AFTERSCHOOL FACILITIES BY SCHOOL DEIS STATUS Cohort ’98 Cohort ’08 UB1 (%) UB2 (%) Rural (%) NonDEIS (%) UB1 (%) UB2 (%) Rural (%) Non-DEIS (%) Poor 7.3 27.7 53.3 44.8 8.0 33.0 35.1 25.0 Fair 29.1 20.3 34.6 23.0 17.1 21.1 19.5 Good 40.6 32.9 12.1 23.8 58.3 46.3 31.0 40.5 Excellent 23.1 19.1 8.3 16.6 20.7 12.8 15.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: GUI Cohort ’08 and Cohort ’98 at age 9, School Principals’ Survey. Notes: UB1 is urban band 1, UB2 is urban band 2, Rural is rural DEIS school. Where cell sizes are too small, they are merged. Some columns may not add to exactly 100 due to rounding. Despite the differences in reported adequacy of afterschool facilities across schools, they had little association with afterschool club/centre participation rates (Table 2.8). Where such facilities were deemed to be poor, 3 per cent of parents reported using them as their main form of care, compared to 5 per cent where they were rated as excellent. This lack of association may be related to the age profile of participants (for example, take-up may have been mostly among the youngest pupils), a mismatch between the hours needed and the hours provided, or because participation was not viewed by parents as a form of out-of-school-hours care. Adequacy may also reflect quality rather than number of places. Further, as stated above, children may have been taking part in afterschool clubs and groups without parents classifying this as childcare. TABLE 2.8 CARE TYPE AT AGE 9 BY PRINCIPAL-RATED ADEQUACY OF AFTERSCHOOL FACILITIES (POOLED ACROSS COHORTS) Care type Poor (%) Fair (%) Good (%) Excellent (%) All (%) Parents 75.7 76.4 72.5 74.2 74.6 Relative 12.0 12.1 13.5 11.6 12.5 Childminder 9.2 7.5 9.3 9.4 8.9 Afterschool club/centre 3.0 4.0 4.7 4.9 4.0 Total 100.0 100.0 100.0 100.0 100.0 Source: Authors’ analysis using Growing Up in Ireland data. Note: Columns may not add exactly to 100 due to rounding. 2.2 WHAT FACTORS INFLUENCE THE USE OF GROUP AFTERSCHOOL CARE AT AGES 9 AND 13? Here we consider the factors that influence take-up of afterschool centre/club care at age 9, combining the data for both cohorts. Section 2.2.1 below considers where these factors changed over time. We first estimate a logit model of using afterschool care compared to not using such care, which includes children cared 24 | Afterschool Care for School-Aged Children in Ireland for by parents, childminder or relatives. The results are presented as average marginal effects, which for categorical variables show the change in the probability of using afterschool care compared to the reference group. These are interpreted as the percentage point increase/decrease in the likelihood of using afterschool care compared to the reference group. To test whether the factors influencing take-up have changed over time, we estimate the interaction between cohort and each explanatory variable. The results of these tests are outlined in Table 2.9. 2.2.1 Factors influencing participation in afterschool clubs/centres at age 9 Model 1 shows that after controlling for relevant compositional factors between cohorts, there was no significant difference between Cohort ’08 and Cohort ’98 in the use of centre-based care after school at age 9. The gender of the child and the migrant status of the parent(s) were not significant, nor was the disability status of the child. While the children were the same age, they may have been in different school classes, which might have influenced the norms around afterschool care; however, this was not found to be significant. Controlling for income and other characteristics, lone parenthood was not related to participation. Where there were older siblings in the household the probability of afterschool club/centre participation was reduced by 2.8 percentage points. The presence of younger siblings was not significant. Parents from the professional social class were more likely to use afterschool care than the semi/unskilled manual group. The strongest predictor of participating in an afterschool centre/club was maternal employment hours. For example, the likelihood of using such afterschool care was 6.9 percentage points higher when mothers were employed for 40 hours or more per week, compared to where the mother was not employed. Finally, there was a small increase in the probability of participating in an afterschool club/centre among those in urban areas, which is likely to reflect differences in the availability of these services in rural areas. We additionally tested the effect of whether the school attended had DEIS status, as such schools were more likely to have afterschool provision (see Section 2.1.2). We found no significant effect (results not presented). As there was a significant number of missing values for school information in Cohort ’08, we do not include this in the final specification. The school principals’ evaluation of adequacy of afterschool facilities was also tested and found not to be significant. Chapter 2 | 25 TABLE 2.9 LOGIT MODEL OF PARTICIPATION IN AFTERSCHOOL CENTRE/CLUB AT AGE 9 AND AGE 13 (AVERAGE MARGINAL EFFECTS) Age 9 Age 13 Cohort ’98 (ref.) Cohort ’08 0.004 -0.020*** Male child (ref.) Female 0.004 -0.003 Child not disabled (ref.) LLC -0.005 0.003 Hampered 0.006 0.000 Two-parent household (ref.) Lone parent 0.011 0.016 No older sibling (ref.) Any older sibling -0.028*** 0.000 No younger sibling (ref.) Any younger sibling -0.006 -0.001 Born in Ireland (ref.) Migrant 0.003 -0.011** Semi/unskilled (ref.) Professional 0.034*** 0.020** Managerial 0.013 0.007 Non-manual 0.003 0.007 Skilled 0.006 0.007 Class missing 0.035 0.004 Mother not employed (ref.) 1–15 hours 0.021* -0.004 16–29 hours 0.028*** 0.000 30–39 hours 0.054*** 0.005 40+ hours 0.069*** 0.014* Lowest income quintile (ref.) Income quintile 2 -0.005 0.001 Income quintile 3 -0.004 -0.004 Income quintile 4 0.005 0.004 Income quintile 5 0.015 0.011 Income missing -0.009 0.006 Age 9: Second class (ref.) Third class 0.005 Fourth class 0.003 Other class 0.011 Age 13: First year (ref.) Second year 0.008** Urban (ref.) Rural 0.012** 0 Observations 15,756 13,359 Source: GUI Cohort ’98 and Cohort ’08 at age 9 and 13 years. Notes: *** p<.001, ** p<.01, * p<.05. Model showing odds ratios for all interactions at age 9 is shown in Appendix Table A2.1. LLC = long-lasting condition. 26 | Afterschool Care for School-Aged Children in Ireland 2.2.2 Factors influencing participation in afterschool clubs/centres at age 13 Factors influencing participation in afterschool clubs/centres at age 13 are shown in the final column of Table 2.9. There was a small decline in participation of 1.5 percentage points in Cohort ’08 compared to Cohort ’98. It is likely that this decline is related to COVID-19 pandemic-related restrictions still being in place for part of the period of fieldwork for the Cohort ’08 wave at age 13. Participation at age 13 did not vary by family structure; neither lone parenthood nor presence of younger/older siblings had a significant effect. Unlike the model for participation at age 9, participation at age 13 was lower among children of migrant background. In line with the findings at age 9, children in the professional social class were also more likely to participate in afterschool clubs/centres at age 13. Participation was more common where the mother was employed for 40 or more hours per week. However, unlike the pattern for 9-year-olds, there was no difference at age 13 between mothers working less than 40 hours per week and mothers who were not employed. This is likely to reflect the longer school days in post-primary school, which can facilitate more parental employment without the need for afterschool cover. Moreover, while school class made no difference at age 9, children in second year at 13 were less likely to participate in afterschool clubs compared to their agemates in first year, suggesting that norms around participation are influenced by schooling stage. Given that participation was so low at age 13 and the potential distorting effect of the COVID-19 pandemic, the focus of the remainder of the chapter is on patterns of care at age 9. 2.2.3 Changes in factors influencing afterschool centre/club care at age 9 between cohorts To test change between cohorts, each of the explanatory factors in Model 1 are interacted with the cohort variable. The model results are presented as odds ratios in Appendix Table A2.1. However, as these interactions cannot be easily interpreted, we present the predicted probabilities from the model in the graphs below. The results are based on models in which each interaction is separately tested. Chapter 2 | 27 2.2.3.1 Social class Figure 2.1 presents the interaction between social class and cohort. For Cohort ’98, we see that those who cannot be assigned to a social class because of their lack of employment history (‘no class’) had the highest level of participation in afterschool centres/clubs; however, among Cohort ’08 participation had declined, while participation for other groups remained static. The model results show that the professional classes were more likely to participate than the semi/unskilled manual group, even controlling for income and employment intensity of the mother. FIGURE 2.1 PROBABILITY OF AFTERSCHOOL CENTRE/CLUB PARTICIPATION AT AGE 9: COHORT BY SOCIAL CLASS INTERACTION Source: GUI Cohort ’98 and Cohort ’08. Note: Results from model with all controls listed in Table 2.9 plus the class-by-cohort interaction. 2.2.3.2 Income quintile There is a different pattern of change across income groups (Figure 2.2). Among the more advantaged groups (top income quintile – group 5), participation in afterschool care increased. In contrast, participation fell among the bottom income quintile. This means that the income gaps in participation are wider for Cohort ’08 than in Cohort ’98. The error bars show there are wide confidence intervals around these figures due to the small number of cases; therefore, the estimates should be treated with caution. 28 | Afterschool Care for School-Aged Children in Ireland FIGURE 2.2 PROBABILITY OF AFTERSCHOOL CENTRE/CLUB PARTICIPATION AT AGE 9: COHORT BY INCOME QUINTILE INTERACTION Source: GUI Cohort ’98 and Cohort ’08 at age 9. Note: Quintile 1 is the bottom income quintile, quintile 5 is the top income quintile. 2.2.3.3 Mother’s employment In both cohorts, the children of mothers with the longest employment hours were most likely to be in afterschool centres/clubs, and those with mothers not in paid employment were least likely to participate (see Figure 2.3). However, the gap between those with more and fewer hours of paid work widened over time. Participation of the children of those not employed declined between cohorts; this is consistent with the pattern found for income quintile and social class. Chapter 2 | 29 FIGURE 2.3 PROBABILITY OF AFTERSCHOOL CENTRE/CLUB PARTICIPATION: COHORT BY MOTHER’S EMPLOYMENT INTERACTION Source: GUI Cohort ’98 and Cohort ’08 at age 9. Note: Quintile 1 is the bottom income quintile, quintile 5 is the top income quintile. 2.2.3.4 Family structure The association between family structure and afterschool centre/club participation also shifted over time (Figure 2.4). In Cohort ’98 at age 9, children of lone parents were more likely to participate in an afterschool centre/club than children of twoparent households; however, in Cohort ’08 no difference was observed.28 The participation of children in lone-parent families declined, while that of children in two-parent families rose slightly. This may be due to cost limitations for loneparent households. 28 The significance of this difference was confirmed with contrast margins. 36 | Afterschool Care for School-Aged Children in Ireland Participation in afterschool clubs/centres at age 13 was lower than at age 9. There was also a decline between cohorts, which is likely due to the onset of COVID-19 pandemic-related restrictions, some of which were still in place during fieldwork with Cohort ’08 at age 13. Despite the differences in rates at ages 9 and 13, similar factors influenced participation, with those from professional backgrounds and with mothers employed for 40 or more hours per week being more likely to participate. Unlike the situation at age 9, children with a migrant background were less likely to participate at age 13. Chapter 3 | 37 CHAPTER 3 Afterschool care and activities among 9and 13-year-olds 3.1 INTRODUCTION There has been little research on the influence of type of care on children’s day-today activities. However, the type of care that children and young people receive outside of the school day might be expected to influence the kinds of activities they engage in. For example, children being cared for by their grandparents may be in another neighbourhood and not have easy access to their usual friends. Alternatively, children may form larger friendship networks in formal care settings. This chapter compares a range of activities at 9 and 13 years of age, including number of, and contact with, friends; involvement in organised sports and cultural activities; and screentime. Chapter 2 has indicated significant differences in the profile of families using different types of afterschool provision. The analyses in this chapter therefore take account of child and family background characteristics in looking at the net effect of afterschool provision on social activities. 3.2 FRIENDSHIPS AND PEER INTERACTION Analyses in this chapter are based on nested logit models and ordered logit models of the relationship between type of care provision and the activity in question (see Chapter 1). In all tables, Model 1 looks at the raw differences between children receiving different kinds of afterschool care. Model 2 examines whether any differences found relate to the profile of children and their families, taking account of gender, disability, parental education, social class, family structure, migrant background, urban/rural location and class or year level in school. Model 3 adds in mother’s employment hours and whether the (resident) father is employed or not. This is a more stringent test of the potential impact of care provision, as parental employment patterns and type of afterschool care are closely related (see Chapter 2). Analyses presented in the appendix tables test whether any relationship between care provision and activities changed between Cohort ’98 and Cohort ’08. Table 3.1 looks at the number of friends 9-year-olds had and how often they saw them outside school. According to mothers’ responses, 9-year-olds typically had two or three close friends and saw them two or three days a week (see Table A3.1 for all descriptive patterns). Model 1 indicates that those in relative care had slightly more friends than those in parental care. However, this difference is no longer significant in Model 2, indicating that this pattern reflects the profile of families using relative care. The frequency with which 9-year-olds saw their friends does not vary by type of afterschool provision. Over the period between Cohorts ’98 and ’08 the number of friends that children had increased, while the frequency of seeing them reduced. These changes occurred across types of afterschool care (Table A3.3). The only exception is that those who looked after themselves or were 38 | Afterschool Care for School-Aged Children in Ireland cared for by siblings had fewer friends in Cohort ’98 but resembled other groups ten years later. TABLE 3.1 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION, NUMBER OF FRIENDS AND FREQUENCY OF SEEING FRIENDS AMONG 9-YEAR-OLDS (ODDS RATIOS) Number of friends Frequency of seeing friends Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 1.324*** 1.424*** 1.450*** 0.870*** 0.878** 0.885** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.671 1.125* 1.074 0.995 0.637 1.056 1.015 0.979 0.617 1.038 1.002 0.965 1.303 1.019 0.967 1.049 1.164 0.987 1.052 0.963 1.158 0.994 1.070 0.962 N 15,665 15,665 Pseudo R2 0.003 0.011 0.011 0.001 0.029 0.029 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. For number of friends: cut points for Model 1 are -2.356, 0.020 and 1.491; for Model 2, -2.597, -0.186 and 1.303; for Model 3, -2.582, -0.169 and 1.322. For frequency of seeing friends: cut points for Model 1 are -2.789, -1.249, 0.194 and 1.016; for Model 2, -3.032, -1.456, 0.079 and 0.960; for Model 3, -2.967, -1.380, 0.156 and 1.038. At age 13 young people typically reported having two or three close friends, but the size of friendship groups declined between cohorts. Within both cohorts, young people being cared for by relatives tended to have fewer friends and close friends, while those in afterschool care tended to have more friends overall (though this is significant only at the 10 per cent level)29 (Table 3.2). Comparing the two cohorts, the number of friends and close friends was found to have declined over time, a trend that was evident across all care provision groups (Table A3.4). 29 The small number in this group (see Chapter 2) may make it difficult to discern a significant difference. Chapter 3 | 39 TABLE 3.2 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION, NUMBER OF FRIENDS AND NUMBER OF CLOSE FRIENDS AMONG 13-YEAR-OLDS (ODDS RATIOS) Number of friends Number of close friends Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.523*** 0.570*** 0.570*** 0.601*** 0.640*** 0.624*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.908 0.767* 1.344 1.575± 0.893 0.758* 1.324 1.535± 0.873± 0.759* 1.312 1.535± 0.928 0.752** 1.229 1.042 0.900 0.753** 1.143 0.987 0.858* 0.725** 1.087 0.958 N 12,749 12,749 Pseudo R2 0.013 0.017 0.018 0.008 0.015 0.016 Source: GUI Cohorts ’98 and ’08 at age 13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. For number of friends: cut points for Model 1 are -2.415, -0.198 and 1.458; for Model 2, -2.302, -0.069 and 1.598; for Model 3, -2.303, -0.066 and 1.603. For number of close friends: cut points for Model 1 are -2.691, -0.380 and 1.024; for Model 2, -2.464, -0.127 and 1.293; for Model 3, -2.483, -0.142 and 1.278. 3.3 ORGANISED SPORTS AND CULTURAL ACTIVITIES This section looks at whether involvement in sports clubs, reading for pleasure and structured cultural activities, such as drama or dance classes or clubs, varied by type of afterschool provision.30 The results for sports clubs (Table 3.3) are reported as marginal effects. The majority of 9-year-olds in both cohorts were involved in a sports club (Table A3.1). Model 1, Table 3.3 shows that involvement in organised sports tended to be higher among those cared for by relatives or childminders. The pattern for relative care is due to the profile of children and their families using this type of provision (Model 2). However, there was a net difference of 4–5 percentage points in participation between those with childminders and those in other care types, a pattern that holds taking account of child and family characteristics as well as parental employment patterns. Participation declined over time, but this happened equally across types of afterschool care (Table A3.5). 30 It should be noted that information is not available on whether these activities took place in the evenings or at weekends, so we cannot determine the extent to which they overlapped with time in afterschool care. 40 | Afterschool Care for School-Aged Children in Ireland TABLE 3.3 LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND INVOLVEMENT IN A SPORTS CLUB AMONG 9-YEAR-OLDS (AVERAGE MARGINAL EFFECTS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) -0.095*** -0.104*** -0.097*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.009 0.037* 0.118*** 0.002 0.034 0.006 0.047** -0.026 0.025 -0.007 0.035* -0.036 N 15,665 Pseudo R2 0.014 0.107 0.109 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. Most 9-year-olds said that they read at least a few times a week (Table A3.1). In descriptive terms, reading for pleasure was more frequent among those in afterschool clubs/centres or cared for by childminders, and lowest among those who took care of themselves or were minded by siblings and, to some extent, those cared for by relatives (Model 1, Table 3.4). The reading patterns for those in paidfor care (childminders and centres) reflected the profile of children and their families. However, reading frequency was lower among the self/sibling group, even taking account of family characteristics and parental employment patterns (Model 3). This gap was evident for Cohort ’98 but narrowed over time (Table A3.5). Involvement in structured cultural activities, such as music or dance classes, was also higher among those in more formal care (childminders or centres), but again this is explained by differences in the family characteristics of those using different care types. However, involvement among those in relative care was lower, by about 5 percentage points, than might be expected given their profile (Models 2 and 3). Further investigation indicates that this difference was evident among those in Cohort ’98 but not in Cohort ’08 (Table A3.5). Both reading frequency and engagement in cultural activities declined over time, but these trends did not vary by type of afterschool provision (Table A3.5). Chapter 3 | 41 TABLE 3.4 MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND FREQUENCY OF READING AND INVOLVEMENT IN STRUCTURED CULTURAL ACTIVITIES AMONG 9-YEAROLDS Frequency of reading (Odds ratio) Cultural activities (Average marginal effects) Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.917* 0.783*** 0.777*** -0.038*** -0.075*** -0.072*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.509** 0.904± 1.303*** 1.381** 0.574* 0.875* 1.028 1.170 0.578* 0.908 1.076 1.213± -0.025 -0.024 0.086*** 0.083** 0.023 -0.047** 0.008 0.027 0.025 -0.048** 0.008 0.027 N 15,665 15,665 Pseudo R2 0.002 0.022 0.023 0.004 0.141 0.142 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Frequency of reading models are ordered logits, while cultural activity is a logit model. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. For frequency of reading: cut points for Model 1 are -2.739, -2.272, -1.559, -0.884 and 0.799; for Model 2, -2.104, -1.627, -0.894, -0.192 and 1.569; for Model 3, -2.137, -1.657, -0.924, -0.225 and 1.536. At age 13 the majority were still involved in sports clubs, but involvement was not as high as at age 9 (Tables A3.1 and A3.2). Those in the care of childminders or afterschool clubs/centres were significantly more likely to be involved in a sports club than those in parental or other family care (Table 3.5). This difference holds even when the profile of families and parental employment patterns are taken into account, with differences of 10 to 14 percentage points in the level of involvement (Models 2 and 3). In contrast to the pattern at age 9, involvement in sports increased by about 7 percentage points between the cohorts, though this trend did not vary by type of afterschool provision (Table A3.5). 42 | Afterschool Care for School-Aged Children in Ireland TABLE 3.5 LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND INVOLVEMENT IN A SPORTS CLUB AMONG 13-YEAR-OLDS (AVERAGE MARGINAL EFFECTS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.078*** 0.076*** 0.073*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) -0.010 -0.015 0.141** 0.179*** -0.025 -0.008 0.095* 0.140** -0.029 -0.011 0.093* 0.139** N 12,749 Pseudo R2 0.008 0.070 0.073 Source: GUI Cohorts ’98 and ’08 at age 13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. The measure of frequency of reading used differed between the cohorts, so patterns over time cannot be directly compared.31 Table 3.6 shows few systematic differences in either cohort, though in Cohort ’08 13-year-olds looking after themselves or being cared for by siblings were much less likely to read, even taking account of a range of other characteristics. TABLE 3.6 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND FREQUENCY OF READING AMONG 13-YEAR-OLDS (ODDS RATIOS) Frequency of reading – Cohort ’98 Frequency of reading – Cohort ’08 Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.920 0.892 0.780 1.130 0.863 0.895 0.634± 1.138 0.882 0.937 0.644± 1.198 0.773* 0.757 1.347 1.304 0.774* 0.806 1.084 1.117 0.803* 0.825 1.087 1.094 N 7,243 5,501 Pseudo R2 0.000 0.014 0.015 0.001 0.020 0.021 Source: GUI Cohorts ’98 and ’08 at age 9=13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. For Cohort ’98: cut points for Model 1 are -1.363, 0.188 and 1.525; for Model 2, -1.084, 0.509 and 1.875; for Model 3, -1.147, 0.450 and 1.816. For Cohort ’08: cut points for Model 1 are -0.251, 0.558 and 1.886; for Model 2, 0.366, 1.224 and 2.606; for Model 3, 0.423, 1.303 and 2.713. 31 Cohort ’98 were asked ‘on a normal weekday during term-time, about how many hours do you spend reading for pleasure?’, with respondents allowed to give an answer in terms of hours/minutes or ‘none’. Cohort ’08 were asked how many times they read for fun per week, with response categories including ‘every day’, ‘3-6 times a week’, ‘once or twice a week’ and ‘less than once a week/never’. Chapter 3 | 43 Young people being cared for by themselves or their siblings were also less likely to be involved in structured cultural activities, such as music or dance classes, at age 13 (Table 3.7). Furthermore, this group of 13-year-olds saw less of an increase in cultural involvement over time than those using other types of care provision (Table A3.6). TABLE 3.7 MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND INVOLVEMENT IN STRUCTURED CULTURAL ACTIVITIES AMONG 13-YEAR-OLDS (AVERAGE MARGINAL EFFECTS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.100*** 0.080*** 0.088*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) -0.033± 0.016 0.081 -0.021 -0.044* 0.013 0.004 -0.029 -0.042* 0.017 0.007 -0.026 N 12,749 Pseudo R2 0.011 0.095 0.096 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. 3.4 SCREENTIME This section looks at the relationship between type of care provision at age 9 and 13, ownership of a mobile phone and length of screentime, distinguishing between television and other digital activities. The span between the two cohorts showed an expansion in mobile phone ownership among 9-year-olds, with an increase of 10 percentage points over the decade (Table 3.8). The pattern of growth differed somewhat by care provision (Table A3.7), with an initially higher level of ownership among those in relative care but a narrowing of this gap over time. In contrast, the growth in ownership was somewhat greater among those with childminders. 44 | Afterschool Care for School-Aged Children in Ireland TABLE 3.8 LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND THE 9-YEAROLD HAVING THEIR OWN MOBILE PHONE (AVERAGE MARGINAL EFFECTS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.101*** 0.103*** 0.104*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.117± 0.060*** -0.038* -0.016 0.073 0.058*** 0.011 -0.004 0.058 0.034* -0.016 -0.028 N 15,665 Pseudo R2 0.009 0.040 0.042 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. Nine-year-olds typically spent one to three hours watching TV per weekday, though the proportion watching more than three hours declined between cohorts (Table A3.1). In descriptive terms, time spent watching TV was lower among those with childminders or in afterschool clubs or centres (Model 1, Table 3.9). However, this difference was related to the characteristics of children and their families. Once we account for the profile of families, those in the care of relatives were found to spend more time watching television. On closer investigation (Table A3.7), this difference was evident among 9-year-olds in Cohort ’98 but reduced a decade later for Cohort ’08. TABLE 3.9 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND AMOUNT OF TIME SPENT WATCHING TV AMONG 9-YEAR-OLDS (ODDS RATIOS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.321*** 0.350*** 0.345*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.895 1.103 0.751*** 0.741** 0.802 1.120± 0.913 0.855 0.814 1.161* 0.949 0.886 N 15,665 Pseudo R2 0.041 0.052 0.053 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. Cut points for Model 1 are -3.922, -1.194 and 2.079; for Model 2, -4.458, -1.701 and 1.625; for Model 3, -4.463, -1.700 and 1.634. Chapter 3 | 45 Time online increased significantly between cohorts. Nine-year-olds being cared for by a childminder spent less time online than those in other forms of care (Table 3.10). There is some tendency for those in afterschool clubs or centres to spend less time online, though this is only significant at the 10 per cent level. Not surprisingly, given that the timing of the two cohorts coincided with a period of rapid digital expansion, time spent online doubled over time. This increase was less among the group being cared for by childminders (Table A3.7). TABLE 3.10 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND AMOUNT OF TIME SPENT ONLINE AMONG 9-YEAR-OLDS (ODDS RATIOS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 2.153*** 2.012*** 1.961*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 1.199 1.021 0.816** 0.847± 1.222 1.040 0.862* 0.862 1.194 1.018 0.839** 0.843± N 15,665 Pseudo R2 0.020 0.025 0.025 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. Cut points for Model 1 are -0.542 and 1.677; for Model 2, -0.474 and 1.763; for Model 3, -0.430 and 1.804. These analyses do not explore mobile phone ownership among 13-year-olds, as almost all of this age group owned a phone (Smyth, 2024). Table 3.11 shows a marked reduction in television watching between Cohorts ’98 and ’08 but no marked variation by type of care provision. The reduction in TV watching does not vary by care provision either (Table A3.8). TABLE 3.11 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND TIME SPENT WATCHING TV AMONG 13-YEAR-OLDS (ODDS RATIOS) Model 1 Model 2 Model 3 Cohort ’08 (ref. Cohort ’98) 0.273*** 0.283*** 0.273 Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 1.050 1.144 0.675± 1.168 1.068 1.110 0.814 1.230 1.061 1.075 0.808 1.218 N 12,749 Pseudo R2 0.044 0.051 0.051 Source: GUI Cohorts ’98 and ’08 at age 13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Model 3 controls for mother’s employment hours and whether the resident father is employed or not. Cut points for Model 1 are -1.994, -0.378 and 0.767; for Model 2, -2.471, -0.831 and 0.334; for Model 3, -2.423, -0.780 and 0.385. 52 | Afterschool Care for School-Aged Children in Ireland Female students performed worse than their male counterparts on maths tests at age 13, holding all else constant (Table A4.3 in the appendix). The migrant status of parents is not significant for any of the three models. The parent’s educational attainment is significant across all three models, and the coefficients increase as parental education increases when compared to children of parents with only a Junior Certificate level of education. Children of middleto high-income quintile families also performed better on maths tests at age 13 relative to children of families in the lowest income quintiles, all else equal. This is still significant when controlling for maths scores at age 9 in Model 4. Additionally for Model 4, those in second year of post-primary school performed worse on maths tests at age 13 compared to their counterparts in first year of post-primary school, significant at p<0.05. Those who performed better in their maths tests at age 9 also performed better on maths tests at age 13, all else equal. 4.3 SOCIO-EMOTIONAL OUTCOMES This final section addresses the socio-emotional outcomes of children at age 13, as a cross-cohort comparison between Cohort ’08 and Cohort ’98. Previous research found that parents and teachers reported differing socio-emotional outcomes depending on the type of afterschool care that children were in (Byrne, 2016). In the GUI study the behaviour of the young person was rated by the SDQ, which creates a ‘total difficulties’ score from emotional symptoms, conduct problems, peer relationship problems, and hyperactivity/inattention. The ‘total difficulties’ scale ranges from 0 to 40, with higher values indicating greater difficulties. TABLE 4.4 CROSS-COHORT COMPARISON OF SDQ SCORES AT AGE 13: OLS REGRESSION COEFFICIENTS Model 1 Model 2 Model 3 Model 4 Care type at age 9 (ref. parents) Self/sibling 0.103 -0.327 -0.322 -0.826 Relative -0.004 0.263 0.263 0.195 Childminder -0.915*** -0.104 -0.110 -0.150 Afterschool club/centre 0.124 0.565 0.544 0.228 Cohort ’08 (ref. Cohort ’98) -0.395** -0.386** 0.008 In afterschool care at age 13 0.482 0.626 Total SDQ score at age 9 0.575*** Adjusted R2 0.002 0.183 0.186 0.453 N 13,154 Source: Authors’ analysis using Growing Up in Ireland data for Cohorts ’98 and ’08 at age 13 and age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for cohort, gender, parent migrant status, educational attainment, disability status, household type, income quintile, and year of post-primary school. Model 3 adds in whether the child is in afterschool care at age 13, and Model 4 adds in the child’s total SDQ score at age 9. Chapter 4 | 53 Table 4.4 is a cross-cohort comparison of total SDQ scores at age 13. Being cared for by a childminder at age 9 is significant in Model 1, with this group scoring lower on the SDQ measure relative to those cared for by parents (meaning lower socioemotional difficulties). However, this relationship becomes non-significant once controls are added, suggesting it is due to the selective nature of those participating as outlined in Chapter 2. Cohort ’08 have lower SDQ scores relative to the baseline of Cohort ’98, taking account of a number of other factors, including disability.32 Cohort ’08 turned 13 in 2021, during the COVID-19 pandemic, which is likely to have affected socio-emotional outcomes, but research suggests that the effects differed by gender (Smyth and Russell, 2024). Being in afterschool care at age 13 is not significant, but those with higher SDQ scores at age 9 had similarly high socio-emotional difficulties on their tests at age 13, all else equal. Overall, it is evident that type of afterschool care has little association with SDQ, explaining less than 1 per cent of variance (Model 1). Moreover, the full models show that family factors – such as household income and mother’s education level – and child characteristics, especially the presence of a disability or long-lasting condition, have a much stronger association with SDQ (see table A4.3). 4.3.1 SDQ Subscales Table 4.4 shows the relationship between the aggregate SDQ score and afterschool care, but the SDQ ‘total difficulties’ score consists of two key subscales, externalising behaviours and internalising behaviours, grouped according to Goodman et al. (2010). Externalising behaviours refer to behavioural and hyperactivity issues, while internalising behaviours refer to emotional and peer issues. A third scale, the pro-social skills scale, is measured but not included in the aggregate SDQ ‘total difficulties’ score. In Table 4.5 we present the coefficients from the OLS regression for the three SDQ subscales. There is no significant effect of afterschool care type at age 9 on children’s prosocial scores, suggesting that pro-social behaviour at age 13 is caused by other factors (Table 4.5, Models 1–4). For externalising behaviour, referring to behavioural and hyperactivity issues, attending an afterschool club/centre is associated with a higher SDQ score. This means that attending afterschool care at age 9 is associated with a small increase in behavioural and hyperactivity problems at age 13. This is not significant in Model 1 with no controls, but becomes significant once family and child characteristic controls are added in Model 2 (p<.001). Adding in SDQ score at age 9 for Model 4 decreases significance slightly (p<.05), with a coefficient of 0.460. However, while statistically significant the effect size is small and afterschool care type explains less than 0.5 per cent of variance in externalising scores. 32 SDQ scores increased overall in the wake of the pandemic (from an average of 6.4 for Cohort ’98 to 7 for Cohort ’08, a pattern driven by the increase in emotional difficulties), but this largely reflected an increase among those with a disability (see Smyth and Russell, 2024). 54 | Afterschool Care for School-Aged Children in Ireland TABLE 4.5 SDQ SUBSCALES – CROSS-COHORT COMPARISON Model 1 Model 2 Model 3 Model 4 Pro-Social Afterschool care at age 9 (ref. parents) Self/sibling 0.345 0.383 0.383 0.348 Relative 0.046 0.057 0.057 0.049 Childminder -0.050 -0.024 -0.024 -0.028 Afterschool club/centre 0.004 0.030 0.030 0.018 Afterschool club/centre at age 13 0.005 0.038 Adjusted R2 0.000 0.033 0.033 0.195 Externalising Afterschool care at age 9 (ref. parents) Self/sibling 0.386 0.131 0.136 -0.113 Relative -0.053 0.127 0.127 0.19 Childminder -0.366* 0.103 0.096 0.109 Afterschool club/centre 0.263 0.565*** 0.540*** 0.460* Afterschool club/centre at age 13 0.563 0.671* Adjusted R2 0.001 0.133 0.134 0.284 Internalising Afterschool care at Age 9 (ref. parents) Self/sibling -0.283 -0.457 -0.458 -0.341 Relative 0.049 0.137 0.137 0.085 Childminder -0.549*** -0.207* -0.206* -0.193 Afterschool club/centre -0.140 0.000 0 003 -0.046 Afterschool club/centre at age 13 -0.082 0.145 Adjusted R2 0.002 0.160 0.160 0.344 N 13,154 Source: Authors’ analysis using Growing Up in Ireland data for Cohorts ’98 and ’08 at age 13 and age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 2 controls for cohort, gender, parent migrant status, educational attainment, disability status, household type, income quintile, and year of post-primary school. Model 3 adds in whether the child is in afterschool care at age 13, and Model 4 adds in the child’s total SDQ score at age 9. Internalising behaviours, referring to peer and emotional issues, are only significantly different for those cared for by childminders. This means that childminder care is associated with lower peer and emotional issues at age 13. This is significant across Models 1–3 but becomes non-significant once controlling for internalising behaviour scores at age 9. 4.4 CONCLUSIONS Reading scores at age 13 and at age 9 in Cohort ’98 were higher for those minded by a childminder than those minded by a parent, even controlling for family background, but this positive relationship becomes non-significant once reading scores at age 9 are added in as a control. In Cohort ’98, attending an afterschool club/centre at age 13 is associated with an increase in reading scores at age 13 (+3.1), controlling for earlier reading scores. In the case of Cohort ’08, only relative Chapter 4 | 55 care is significant in the model, associated with slightly lower reading scores at age 13. The results for both cohorts cannot be directly compared, as different tests were used across each cohort. Due to the shift to home-based (rather than school-based) test administration at age 9 and constraints caused by the COVID-19 pandemic at age 13, Cohort ’08 were not administered a maths test. In Cohort ’98 there is a positive relationship between care by a childminder at age 9 and maths scores at age 13. The magnitude of this relationship reduces as controls are added in, but it remains significant across models. This includes maths scores at age 9. The significant effect of the cognitive scores at age 9 on the same scores later suggests that early educational interventions and support in reading and maths will have a greater impact on later outcomes. In both cohorts, type of afterschool care explains little of the variation in cognitive scores, and effect sizes are much greater for family characteristics, especially mother’s education level but also income and family structure. For example, for Cohort ’98 the increase in reading scores at age 13 associated with having a mother with degree-level education is seven times greater than the increase associated with attending a childminder at age 9. For both cohorts combined, there is no relationship between total SDQ scores at age 13 and type of afterschool arrangements at age 9 once child and family characteristics are taken into account. However, we find a weak association between afterschool centre/club attendance and increased externalising behaviour. The findings in this chapter are similar to those of Byrne (2016), who assessed outcomes for Cohort ’98 and found higher socio-emotional difficulties at age 13 for those who had attended afterschool centres at age 9, before controls for relationship with mother and mother’s depression. Overall, afterschool care explains very little of the variation in cognitive and socio-emotional outcomes. This suggests that the characteristics of children and their families remain more relevant than afterschool care choices in determining outcomes for children. The lack of significant effects of afterschool care for most of the outcomes analysed may disguise differences between services of different quality. The literature on early childhood education and care underlines that the effects are differentiated by the quality of provision (Camilli et al., 2010; Ulferts et al., 2019). We currently lack detailed information within the GUI study on the nature of afterschool provision, including the quality of indoor and outdoor facilities, the range and type of activities offered, and the background and experience of staff. Being able to disaggregate data on care by a range of quality metrics in the new GUI birth cohort (Cohort ’24) would therefore be very beneficial. 56 | Afterschool Care for School-Aged Children in Ireland CHAPTER 5 Conclusions and Policy Implications This research focuses on the Growing Up in Ireland ’98 and ’08 Cohorts at ages 9 and 13, examining both their participation in afterschool care and the relationship between that participation and children’s friendship networks, activities (including screen-based activity, reading and sports), and cognitive and socio-emotional outcomes. The research also assesses whether these patterns have changed across cohorts. This research takes place in the context of ongoing policy development for afterschool care in Ireland, which has improved quality guidelines and started consultations to establish professional qualifications in school-age childcare. Ireland has also developed new regulation requirements for childminders and has created a more centralised portal for parents and service providers. 5.1 PATTERNS OF AFTERSCHOOL CARE AT AGE 9 AND AGE 13 At age 9, the majority of children were looked after by their parents after school, though this dropped somewhat from 76 per cent for Cohort ’98 to 71 per cent for Cohort ’08. A minority of children aged 9 years old in Ireland participated in afterschool clubs or centres as their main form of care, with little change in takeup between the ’98 and ’08 Cohorts. For Cohort ’98 in 2007–2008, 3.4 per cent were in afterschool clubs/centres compared to 4.5 per cent of Cohort ’08 in 2017. In both cohorts, the most common non-parental care type was relative care, with childminders the most frequently used formal (paid-for) provision. Hours of participation tended to be relatively short, at an average of 11–12 hours per week. There was a small decline in average hours of afterschool care (of all types) between cohorts, but a consistent pattern of shorter hours for afterschool clubs/centres, and longer hours spent in childminder care. Patterns of take-up were different for 13-year-olds, with 86 per cent being cared for by their parents after school, a figure that remained stable over time. The main non-parental care type was being looked after by siblings or looking after themselves, with only a small number (2 per cent for Cohort ’98 and 0.5 per cent for Cohort ’08) attending formal afterschool centres/clubs. The cost of afterschool care for 9-year-olds increased significantly between the two periods, above the rate of inflation.33 This increase was greatest for childminders but was also present for afterschool clubs/centres. Maternal employment was the strongest predictor of afterschool club/centre use, which is consistent with the findings of comparative research in Europe and the OECD (Brandon and Hofferth, 2003; Cartmel and Hayes, 2016; OECD, 2007). The likelihood of participation in afterschool clubs/centres at age 9 was 7 percentage 33 No information was collected on the cost of afterschool care for 13-year-olds. Chapter 5 | 57 points higher where the mother worked 40 hours or more per week compared to non-employed mothers. Maternal employment hours was also the strongest predictor of other non-parental care types, and longer employment hours was particularly strongly associated with using childminder care. Income became a more important predictor of afterschool club/centre participation across cohorts, with the gap between the lowest and highest income quintiles widening over time. Children of lone parents were more likely to participate in afterschool care compared to children of two-parent households for Cohort ’98, but this was no longer the case for Cohort ’08. Those living in urban areas were more likely to participate in afterschool clubs and centres, suggesting that availability of places, or of transport, may influence takeup in rural areas. Instead, families in rural areas were more likely to use childminders than those in urban areas. There was no difference in participation in afterschool clubs/centres by DEIS school status, though the survey of school principals shows that the perceived adequacy of afterschool facilities was higher in urban DEIS schools. The proportion of school principals reporting ‘excellent’ or ‘very good’ facilities increased significantly between the cohorts, potentially reflecting increased investment over the period 2007 to 2017. However, this improvement was not associated with a significant increase in take-up, most likely because the responses relate to extracurricular provision in general rather than care provision per se. 5.2 AFTERSCHOOL CARE AND ACTIVITIES AMONG 9AND 13-YEAR OLDS The care settings that children and young people attend outside of the school day might be expected to influence the kinds of activities they engage in, with potential consequences for their mental and physical wellbeing. We found that 9-year-olds cared for by a childminder were more involved in sports clubs than their peers and, along with those attending afterschool clubs/centres, spent less time watching TV or being online. Young people who were cared for by relatives at age 9 were less involved in reading and cultural activities, and those looking after themselves or being cared for by siblings were less likely to read for pleasure. This pattern for relative care continued at age 13. Research finds that young people at age 13 tend to spend more time with their peers and less time with their families (Larson and Richards, 1991), but their network is affected by the type of care they receive. Those cared for by relatives had fewer friends, and fewer close friends, than their peers, suggesting that this care type may disrupt local networks. Young people attending childminders and afterschool clubs/centres were more likely to be involved in sports clubs. There 58 | Afterschool Care for School-Aged Children in Ireland was little variation in time online or on other devices by care type, most likely reflecting the ubiquity of mobile phone ownership at age 13. 5.3 AFTERSCHOOL CARE TYPE AND SOCIO-EMOTIONAL AND COGNITIVE OUTCOMES Different measures of cognitive outcomes at age 13 were used due to the disruption of the COVID-19 pandemic for the 13-year-old wave of the ’08 cohort. For Cohort ’98, reading scores were higher for those cared for by a childminder compared to those in parental care only; this remains significant with a range of controls, but becomes non-significant once reading scores at age 9 are added in as a control. This suggests that childminder care at age 9 does not influence the change in reading scores between ages 9 and 13. There was no difference between the reading scores of those attending afterschool centres/clubs at age 9 compared to those in parental care. However, there was a positive effect for those attending afterschool clubs/centres at age 13 (measured at the same time as the outcome), when controlling for earlier reading scores. For Cohort ’08, relative care was associated with lower cognitive scores at age 13 compared to parental care only, but there was no significant difference among those attending afterschool clubs/centres or childminders. The negative effect of relative care remains significant with a range of controls, including reading scores at age 9. These lower scores may reflect the differences in activities engaged in by young people cared for by a relative (e.g. less reading and fewer cultural activities). It may also reflect the lower educational qualifications of these groups. Only Cohort ’98 were tested for maths achievement, and there was a positive relationship between maths scores at age 13 and care by a childminder at age 9 compared to parental care. This remains significant, though the magnitude reduces, when individual and family characteristics are taken into account. There was no relationship between maths scores and afterschool club/centre participation or relative-provided care. It should be acknowledged that any relationship found between type of care and adolescent outcomes is not necessarily causal and may reflect other unmeasured factors that influence both choice of care type and the outcome examined. The absence of effects on cognitive outcomes for afterschool clubs/centres may be due to the relatively short number of hours children spend in these settings, as well as the absence of dedicated educative objectives of these services. There are equally a myriad of confounding factors that may contribute to cognitive outcomes beyond what is controlled for here, such as the school setting of the child. The null finding may also disguise differential effects for settings of different quality or among sub-groups of the population. We lack detailed information on the nature of activities engaged in while children are in these settings, or the quality of Chapter 5 | 59 provision. Research showing positive impacts on cognitive outcomes has tended to come from programmes with a specific educational focus (see Chapter 1), which is not the goal of school-age childcare provision in Ireland. Given the small proportion of children attending afterschool care, we do not present sub-group analyses34 and significant effects might be detected in a larger sample of attendees – for example, those from an economically disadvantaged background or those whose first language is not English. Turning to social and emotional outcomes, we find that children cared for by childminders at age 9 had somewhat lower socio-emotional difficulties at age 13, but this becomes non-significant when child and family characteristics are controlled, suggesting that the initial effect is due to selection. Children attending afterschool clubs/centres at age 9 had slightly higher externalising scores at age 13, even when earlier SDQ scores are controlled. This is consistent with prior research on early years settings (Russell et al., 2016). The overall effect sizes for afterschool care arrangements in the SDQ models is small and explains very little of the variation in socio-emotional outcomes. Child characteristics, especially the presence of a disability, and family factors such as maternal education and income, have a stronger association with these outcomes. 5.4 POLICY IMPLICATIONS Census figures for 2022 indicate that 30 per cent of 5to 12-year-olds and 7 per cent of 13/14-year-olds are in afterschool care in group settings. Pobal figures indicate that provision has continued to expand since then, with the numbers aged 8 years or more in school-age care increasing from 15,030 in 2021/22 to 24,934 in 2023/24. The most substantial policy developments for school-age childcare, and resulting increase in take-up, therefore occurred after the children in Cohorts ’98 and ’08 were aged 9 and in school-age childcare. This report finds little change in patterns of afterschool care and outcomes between cohorts, which may reflect a lack of policy change in the years between 2008 and 2017. The type of afterschool arrangements that families have in place was found to be related to a range of factors including working hours, rural/urban location and income. The influence of such factors suggests that choice was limited for lowincome parents, those in rural areas and those working longer hours. Significant developments in the area of school-age care have taken place since the data were collected. These include the NCS which provides additional subsidies for parents, and the significant increase in the number of support places for school-aged 34 Models tested the interaction of care type and social background but these interactions were not significant. This may reflect the fact that the small numbers make it difficult to detect difference rather than the absence of variation of the effect of care type across social groups. 60 | Afterschool Care for School-Aged Children in Ireland children. It is important that the impact of these developments is tracked to see if they have resulted in increased choice for parents. This research finds that the use of afterschool childcare centres remained very low in 2017, and the gap between take-up for lowand high-income households had widened. Childcare costs for childminders and afterschool centres had increased beyond inflation, which is likely to influence the care decisions and labour market choices of parents. Improvements to afterschool care will also have a gender equality impact, allowing mothers to engage more fully in the labour market. Recent policy initiatives have aimed to address cost and quality issues. The National Childcare Scheme (2019), which replaced existing childcare programmes with a single scheme and centralised application portal, provides subsidies to parents of school-aged children (up to age 15 years) and comprises both a universal element and an income-related element. In terms of quality, research examining the workforce in school-age childcare in Ireland (Pobal, 2025) identifies a low percentage of staff with higher-level qualifications, as well as low pay. A workforce plan for the sector was released in 2021, which should support increased standards in professional qualifications for care providers. Nurturing Skills: The Workforce Plan for Early Learning and Care and School-Age Childcare 2022–2028 will be crucial in improving retention, pay and qualifications in this workforce (DCEDIY, 2021b). There is currently a lack of information on the nature and quality of afterschool provision in Ireland, including the quality of indoor and outdoor facilities, the background and experience of staff and the quality of the relationships they cultivate with children, and the range and type of activities offered in these settings. Children themselves emphasise the importance of play in afterschool care (Horgan et al., 2017; 2018) so it is important to provide access to a broad range of play activities, including creative and physical activities, that enhance wellbeing and holistic development. Continued evidence of greater externalising difficulties among those attending afterschool centres or clubs is concerning and suggests the need for providers to pay greater attention to fostering a positive peer climate and to enhancing interpersonal and coping skills among children in their settings. The majority of school-age childcare providers are still privately owned for profit: figures for 2023 suggest that three-quarters of provision is in the private sector and one-quarter in the community sector. The privatised nature of the provision restricts the State’s ability to standardise the system and implement change. Regulations setting minimum standards and eligibility rules can link state support to the implementation of practices, yet market forces will remain a constraint on change in a privatised system. Chapter 5 | 61 Childminders remain a significant part of school-age care; yet this care type is more expensive and heretofore less regulated, though the outcomes outlined here for children attending childminders are generally positive. Current policy direction is to encourage childminders into the regulated childcare system, but low pay and poor working conditions remain key barriers. The results also suggest a disparity in childcare provision and take-up regionally, with lower take-up in rural areas. Examining childcare centrally is beneficial, but city and county childcare committees could also identify specific needs in their regions that could be missed on a national level, or childcare needs could be examined regionally. Additional incentives to support childcare workers to stay in the sector and infrastructural supports for providers, particularly in underserviced areas, may be useful to address gaps in provision. 5.5 FUTURE RESEARCH The available data do not include assessments of care quality, activities offered or service characteristics. Future data collection or data linkages connecting information at the provider level to child-level data would create a more nuanced understanding of the role of afterschool care in child outcomes. The higher level of externalising behaviour among children attending afterschool clubs/centres is also important and would benefit from further research into the nature of child–child and child–carer interactions and relationships in these settings, and, for example, whether there are approaches in place to promote inclusion and prevent bullying behaviours (Smyth and Darmody, 2025). However, increased peer-to-peer interaction benefits the development of children’s peer and emotional skills, which are harder to develop in relative care. While the cohorts examined here (Cohorts ’08 and ’98) turned 9 before the introduction of many new policies discussed in Section 1.4, there is a new cohort in 2024 who will turn 9 in 2033. This provides an excellent opportunity to examine the effect of the rapid policy change in Ireland, particularly with regards to the use of paid childcare and outcomes for children following workforce and quality guideline changes, as well as increased regulation of childminders and school-age childcare more generally. In conclusion, GUI data on childcare, and indeed on young people’s activities and experiences, generally relate to the term-time period. There has been no research to date in Ireland on the types of arrangements made by parents during school holidays, particularly during the long summer break. This topic would benefit from future research to examine the costs and consequences of holiday provision for young people and their families. 68 | Afterschool Care for School-Aged Children in Ireland APPENDIX I TABLE A2.1 PARTICIPATION IN AFTERSCHOOL CARE AT AGE 9 – ODDS RATIOS Model 1 Model 2 Cohort ’98 (ref.) 1 1 Cohort ’08 1.102 0.388 Male 1 1 Female 1.117 1.119 Child not disabled (ref.) 1 1 LLC 0.874 0.756 Hampered 1.171 1.12 Two-parent household (ref.) 1 1 Lone parent 1.328 1.631* No older sibling (ref.) 1 1 Any older sibling 0.463*** 0.463*** No younger sibling (ref.) 1 1 Any younger sibling 0.854 0.833 Parents non-migrant (ref.) 1 1 Migrant 1.092 1.065 Semi/unskilled social class (ref.) 1 1 Professional class 2.414*** 3.539*** Managerial class 1.507 1.87 Non-manual 1.111 1.555 Skilled 1.25 1.728 Class missing 2.479* 2.518 Mother not employed (ref.) 1 1 1–15 hours 2.668** 2.274* 16–29 hours 3.260*** 1.637 30–39 hours 5.594*** 3.843*** 40+ hours 7.030*** 5.250*** Lowest income quintile (ref.) 1 1 Income quintile 2 0.853 0.822 Income quintile 3 0.875 0.648 Income quintile 4 1.158 0.712 Income quintile 5 1.461 0.895 Income missing 0.75 0.667 Second class (ref.) 1 1 Third class 1.155 1.142 Fourth class 1.091 1.087 Other 1.338 1.383 Rural (ref.) 1 1 Urban 1.396** 1.348 Cohort ’08 # Professional class 0.388 Cohort ’08 # Managerial class 0.527 Cohort ’08 # Non-manual class 0.432 Cohort ’08 # Skilled 0.407 Appendix I | 69 Model 1 Model 2 Cohort ’08 # Semi/unskilled class (ref.) 1 Cohort ’08 # Class missing 0.55 Cohort ’08 # Mother not employed (ref.) 1 Cohort ’08 # 1–15 hours 1.248 Cohort ’08 # 16–29 hours 4.990*** Cohort ’08 # 30–39 hours 2.621* Cohort ’08 # 40+ hours 2.248 Cohort ’08 # Not disabled (ref.) 1 Cohort ’08 # LLC 1.248 Cohort ’08 # Hampered 1.094 Cohort ’08 # Non-migrant (ref.) 1 Cohort ’08 # Migrant 1.072 Cohort ’08 # Two-parent household (ref.) 1 Cohort ’08 # Lone parent 0.621 Cohort ’08 # Income quintile 1 (ref.) 1 Cohort ’08 # Income quintile 2 1.133 Cohort ’08 # Income quintile 3 2.323 Cohort ’08 # Income quintile 4 3.194* Cohort ’08 # Income quintile 5 3.254** Cohort ’08 # Income missing 1.734 Cohort ’08 # Urban = 0 (ref.) 1 Cohort ’08 # Urban = 1 1.077 Observations 15,756 15,756 Source: Authors’ analysis using Growing Up in Ireland data for Cohorts ’98 and ’08 at age 13 and age 9. Note: Dependent variable 1 = afterschool participation, 0 = all other care arrangements including parental care. 70 | Afterschool Care for School-Aged Children in Ireland TABLE A3.1 FREQUENCIES OF INVOLVEMENT IN ACTIVITIES AMONG 9-YEAR-OLDS, COHORTS ’98 AND ’08 Cohort ’98 Cohort ’08 (%) (%) No. of close friends: 0/1 2/3 4/5 6+ 8.4 41.1 33.2 17.4 7.1 37.2 31.1 24.5 Sees friends per week: Never 1 day 2–3 days 4–5 days 6–7 days 6.1 15.7 33.6 18.1 26.6 6.8 19.1 31.9 18.1 24.1 Attends sports club 75.3 84.0 Attends cultural class/club 47.3 43.6 Time spent reading: Every day A few times a week Once a week A few times a month Less than once a month Never 30.7 40.6 12.6 7.8 2.6 5.7 30.5 38.7 11.2 8.6 4.1 6.9 Time spent watching TV per weekday: None <1 hour 1–<3 hours 3 or more hours 2.1 21.4 65.7 10.8 6.0 43.6 46.3 4.1 Time spent on computer per weekday: None <1 hour More than 1 hour 36.2 50.0 13.8 23.6 46.3 30.1 Own mobile phone 43.7 54.3 N 8,556 7,487 Source: GUI Cohorts ’98 and ’08. Appendix I | 71 TABLE A3.2 FREQUENCIES OF INVOLVEMENT IN ACTIVITIES AMONG 13-YEAR-OLDS, COHORTS ’98 AND ’08 Cohort ’98 Cohort ’08 (%) (%) No. of close friends: 0/1 2/3 4/5 6+ 6.4 34.4 32.8 26.3 7.5 45.2 33.6 13.7 Attends sports club 64.7 69.9 Attends cultural class/club 24.0 33.7 Time spent reading per weekday: <1 hour 1–<2 hours 2–<3 hours 3+ hours 54.8 27.5 9.6 8.1 - Time spent watching TV per weekday: <1 hour 1–<2 hours 2–<3 hours 3+ hours 11.5 29.8 27.4 31.5 35.2 33.7 20.2 10.9 Time spent on computer per weekday: <1 hour 1–<2 hours 2–<3 hours 3+ hours 66.1 18.7 7.3 7.9 39.5 29.3 18.4 12.9 Time spent video/computer gaming per weekday: <1 hour 1–<2 hours 2–<3 hours 3+ hours 36.3 31.0 16.2 16.5 64.1 19.0 0.5 6.4 N 7,383 5,577 Source: GUI Cohorts ’98 and ’08. Note: The measurement of reading is not comparable between Cohorts ’98 and ’08 at age 13. 72 | Afterschool Care for School-Aged Children in Ireland TABLE A3.3 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION, NUMBER OF FRIENDS AND FREQUENCY OF SEEING FRIENDS AMONG 9-YEAR-OLDS, WITH INTERACTION TERMS BETWEEN COHORT AND PROVISION TYPE (ODDS RATIOS) Number of friends Frequency of seeing friends Model 2 Model 2 Cohort ’08 (ref. Cohort ’98) 1.438*** 0.856** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.373** 1.122 0.987 1.040 0.936 0.959 0.964 0.974 Self/sibling* Cohort ’08 Relative* Cohort ’08 Childminder* Cohort ’08 Afterschool club/centre* Cohort ’08 (ref. parents Cohort ’08) 2.849± 0.886 1.055 0.888 1.469 1.060 1.192 0.984 N 15,665 15,665 Pseudo R2 0.003 0.029 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. These models resemble Model 2 in Table 3.1 and control for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. TABLE A3.4 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION, NUMBER OF FRIENDS AND NUMBER OF CLOSE FRIENDS AMONG 13-YEAR-OLDS, WITH INTERACTION TERMS BETWEEN COHORT AND PROVISION TYPE (ODDS RATIOS) Number of friends Number of close friends Model 2 Model 2 Cohort ’08 (ref. Cohort ’98) 0.571*** 0.640*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.882 0.786 1.378 1.486 0.896 0.812 1.126 0.887 Self/sibling* Cohort ’08 Relative* Cohort ’08 Childminder* Cohort ’08 Afterschool club/centre* Cohort ’08 (ref. parents Cohort ’08) 1.023 0.919 0.843 1.217 1.008 0.842 1.064 2.040 N 12,749 12,749 Pseudo R2 0.017 0.015 Source: GUI Cohorts ’98 and ’08 at age 13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. These models resemble Model 2 in Table 3.1 and control for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Appendix I | 73 TABLE A3.5 LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION, INVOLVEMENT IN SPORTS CLUBS, READING AND CULTURAL ACTIVITIES AMONG 9-YEAR-OLDS, WITH INTERACTION TERMS BETWEEN COHORT AND PROVISION TYPE (ODDS RATIOS) Sports club (Logit) Reading frequency (Ordered logit) Cultural activities (Logit) Text Model 2 Model 2 Model 2 Cohort ’08 (ref. Cohort ’98) 0.561*** 0.782*** 0.679*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 1.136 1.020 1.273± 0.902 0.394* 0.925 1.026 0.993 1.736 0.756** 1.068 0.958 Self/sibling* Cohort ’08 Relative* Cohort ’ֽ08 Childminder* Cohort ’08 Afterschool club/centre* Cohort ’08 (ref. parents Cohort ’08) 1.116 1.024 1.059 0.935 1.894 0.893 1.004 1.358 0.451 1.094 0.951 1.400 N 15,665 15,665 15,665 Pseudo R2 0.107 0.023 0.142 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. These models resemble Model 2 in Table 3.1 and control for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. The results for sports club and cultural activity involvement are presented here as odds ratios rather than AMEs (as in Chapter 3) because interaction terms are included. TABLE A3.6 LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND INVOLVEMENT IN SPORTS CLUBS AND CULTURAL ACTIVITIES AMONG 13-YEAR-OLDS, WITH INTERACTION TERMS BETWEEN COHORT AND PROVISION TYPE (ODDS RATIOS) Sports club Cultural activities Model 2 Model 2 Cohort ’08 (ref. Cohort ’98) 1.482*** 1.636*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 0.806 1.195 1.961* 2.280* 0.993 1.263 1.170 0.898 Self/sibling* Cohort ’08 Relative* Cohort ’08 Childminder* Cohort ’08 Afterschool club/centre* Cohort ’08 (ref. parents Cohort ’08) 1.211 0.592± 0.488 0.946 0.633* 0.700 0.605 0.791 N 12,749 12,749 Pseudo R2 0.071 0.096 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. These models resemble Model 2 in Table 3.1 and control for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. The results for sports club and cultural activity involvement are presented here as odds ratios rather than AMEs (as in Chapter 3) because interaction terms are also included. 74 | Afterschool Care for School-Aged Children in Ireland TABLE A3.7 LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION, MOBILE PHONE OWNERSHIP AND TIME SPENT ON TV AND OTHER ONLINE ACTIVITIES AMONG 9-YEAROLDS, WITH INTERACTION TERMS BETWEEN COHORT AND PROVISION TYPE (ODDS RATIOS) Mobile phone TV (Ordered logit) Other online activities (Ordered logit) Model 2 Model 2 Model 2 Cohort ’08 (ref. Cohort ’98) 1.553*** 0.378*** 2.077*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 1.140 1.520*** 0.902 0.880 0.981 1.400*** 0.980 0.993 1.051 0.957 1.100 0.914 Self/sibling* Cohort ’08 Relative* Cohort ’08 Childminder* Cohort ’08 Afterschool club/centre* Cohort ’08 (ref. parents*Cohort ’08) 1.394 0.707** 1.329* 1.229 0.705 0.659** 0.866 0.756 1.296 1.175 0.602*** 0.886 N 15,665 15,665 15,665 Pseudo R2 0.041 0.053 0.025 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. These models resemble Model 2 in Table 3.1 and control for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. TABLE A3.8 ORDERED LOGIT MODELS OF THE RELATIONSHIP BETWEEN CARE PROVISION AND TIME SPENT ON TV, COMPUTER GAMING AND OTHER ONLINE ACTIVITIES AMONG 13-YEAROLDS, WITH INTERACTION TERMS BETWEEN COHORT AND PROVISION TYPE (ODDS RATIOS) TV Computer gaming Other online activities Model 2 Model 2 Model 2 Cohort ’08 (ref. Cohort ’98) 0.281*** 0.275*** 4.167*** Self/sibling Relative Childminder Afterschool club/centre (ref. parents) 1.167 0.968 0.699 1.189 1.431*** 0.868 0.688± 0.794 0.855 1.081 0.737 1.016 Self/sibling* Cohort ’08 Relative* Cohort ’08 Childminder* Cohort ’08 Afterschool club/centre* Cohort ’08 (ref. parents* Cohort ’08) 0.842 1.394 1.972± 1.222 0.666** 1.168 1.384 1.363 1.361± 1.256 1.432 0.778 N 12,749 12,749 12,749 Pseudo R2 0.051 0.058 0.074 Source: GUI Cohorts ’98 and ’08 at age 9. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. These models resemble Model 2 in Table 3.1 and control for gender, disability, parental education, social class, family structure, migrant background, urban/rural and class/year level in school. Appendix I | 75 TABLE A4.1 COHORT ’98 READING SCORES AT AGE 13 Model 1 Model 2 Model 3 Model 4 Care type at age 9 (ref. parents) Self/sibling -3.506 -2.387 -2.363 -0.473 Relative 0.101 -0.769 0.772 -0.559 Childminder 5.149*** 1.627* 1.595* 0.638 Afterschool club/centre -0.190 -1.349 -1.413 0.02 Child gender (ref. male) Female -3.275*** -3.270*** -3.599*** Parent non-migrant (ref.) 1.514 1.455 Migrant 1.480 1.514 1.455 Mother’s educ. (ref. Junior Cert.) Leaving Cert. 5.002*** 4.986*** 2.05** Post-secondary qualification 7.290*** 7.279*** 2.761** Third-level degree 12.312** 12.293*** 5.431*** Lowest income quintile (ref.) Income quintile 2 2.459** 2.455** 1.318* Income quintile 3 2.965*** 2.981*** 0.843 Income quintile 4 4.654*** 4.645*** 1.446** Income quintile 5 (top) 6.315*** 6.288*** 2.129*** Income not reported 3.950*** 3.940*** 1.02 Two-parent household (ref.) Lone parent -2.106* -2.136* -1.414 Child not disabled (ref.) LLC 0.727 0.730 0.330 Disability -2.087 -2.113 -1.035 School class (ref. First year) Second year 2.282*** 2.254*** -2.117*** In afterschool care at age 13 1.531 3.101* Reading score at age 9 0.628*** Constant 97.476*** 88.176*** 88.181*** 34.766*** Observations 6,898 6,898 6,898 6,898 Source: GUI Cohort ’98 at age 13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Model 4 includes a time-lag control variable of reading score at age 9. Mean score of 100 and a standard deviation of 15. 76 | Afterschool Care for School-Aged Children in Ireland TABLE A4.2 COHORT ’08 READING SCORES AT AGE 13 Model 1 Model 2 Model 3 Model 4 Care type at age 9 (ref. parents) Self/sibling 0.151 1.346 1.342 1.362 Relative -1.371 -2.253** -2.255** -2.322* Childminder 1.475 -1.066 -1.053 -1.292 Afterschool club/centre 0.890 -1.158 -1.056 -1.242 Gender (ref. male) Female 0.280 0.279 0.104 Parent migrant status (ref. non-migrant) Migrant -1.387 -1.401 -0.642 Parent education attainment (ref. Junior Cert.) Leaving Cert. 1.051 1.050 0.677 Post-secondary qualification 4.199** 4.208** 3.523** Third-level degree 6.713** 6.724*** 5.222*** Family income quintile: (ref. income quintile 1 – lowest) Income quintile 2 2.198* 2.200* 1.82 Income quintile 3 1.905* 1.915* 1.168 Income quintile 4 2.490* 2.496* 1.683 Income quintile 5 3.748*** 3.765*** 2.709** Income not reported 3.448** 3.460** 2.670* Household type (ref. two-parent household) Lone parent 0.737 0.750 0.563 Child disability status (ref. not disabled) LLC, not hampered -0.205 -0.191 0.195 LLC, hampered -1.751* -1.766* -1.057 Year of post-primary school at age 13 (ref. first year) Second year -1.431* -1.434* -2.624*** In afterschool care at age 13 -2.994 -3.134 Reading score at age 9 0.181*** Constant 98.967*** 93.934*** 93.932*** 78.249*** Observations 5,366 5,366 5,366 5,366 Source: GUI Cohort ’08 at age 13. Notes: *** p<.001, ** p<.01, * p<.05, ± p<.10. Mean cognitive score = 100, standard deviation = 15. Appendix I | 77 TABLE A4.3 COHORT ’98 MATHS OUTCOMES AT AGE 13 Model 1 Model 2 Model 3 Model 4 Care type at age 9 (ref. parents) Self/sibling -3.605 -2.373 -2.400 -1.810 Relative -0.675 -1.286 -1.282 -1.229 Childminder 4.805*** 1.808* 1.843* 1.704* Afterschool club/centre 1.798 0.929 1.003 0.662 Gender (ref. male) Female -3.656*** -3.663*** -2.781*** Parent migrant status (ref. non-migrant) Migrant -0.738 -0.775 -1.166 Parent education attainment (ref. Junior Cert.) Leaving Cert. 4.838*** 4.857*** 4.701*** Post-secondary qualification 6.089*** 6.102*** 5.795*** Third-level degree 10.911*** 10.933*** 10.083*** Family income quintile: (ref. income quintile 1 – lowest) Income quintile 2 0.52 0.525 0.327 Income quintile 3 2.777*** 2.758*** 2.323** Income quintile 4 3.412*** 3.422*** 3.009*** Income quintile 5 4.202*** 4.234*** 3.591*** Income not reported 2.541* 2.553* 1.869 Household type (ref. two-parent household) Lone parent -1.273 -1.240 -1.003 Child disability status (ref. not disabled) LLC, not hampered -0.791 -0.795 -0.996 LLC, hampered -2.089* -2.058* -1.622 Year of post-primary school at age 13 (ref. first year) Second year -0.898 -0.866 -1.138** In afterschool care at age 13 -1.756 -1.792 Drumcondra maths score at age 9 (standardised) 88.835*** Constant 97.559*** 92.289*** 92.283*** 59.098*** Observations 6,940 6,940 6,940 6,940 Source: GUI Cohort ’98at age 13. Note: *** p<.001, ** p<.01, * p<.05, ± p<.10.