Have Socio-Economic Inequalities in Childhood Cognitive Test Scores Changed? A Secondary Analysis of Three British Birth Cohorts

Lead Research Organisation: University of Edinburgh
Department Name: Sch of Social and Political Science

Abstract

Research has consistently demonstrated that there is a relationship between parents' social and economic positions and their children's scores on tests which measure cognitive performance. There are marked social inequalities and children from less advantaged social backgrounds generally perform less well on cognitive tests. Theorists have suggested that children born more recently tend to perform better on cognitive tests than children born in earlier generations. The overall aim of the proposed research is to examine the extent to which the relationship between parents' socio-economic positions and their children's performance on cognitive tests has changed. The UK is unique because it has a long tradition of collecting data on children and following them into adulthood. These studies are known as birth cohort studies and in the proposed research we will undertake advanced statistical analyses of three of these studies.

We will be using data from three birth cohort studies, the National Child Development Study (NCDS) which began in 1958, the British Cohort Study (BCS) which began in 1970, and the Millennium Cohort Study (MCS) which began in 2000. There are two main research challenges associated with the proposed work. The data in these studies was not primarily collected for making comparisons, and over the course of time there have been changes in how the data are collected and how things are measured. The first challenge is the practical one of constructing a set of measures that are suitably consistent, and that allow genuine comparisons to be made.

The second challenge is more statistical. The NCDS and the BCS are older studies and they were collected using a relatively straightforward selection technique, which selected all of the babies born in one week of the year. The MCS has a more complex design. Children are selected because they live in different parts of the UK, and because of the type of area that they were born in. At the same time extra children from areas with high levels of ethnic minority families were also selected. More complex statistical methods are required to analyse data from the MCS. At the current time there is no recognised method for undertaking combined statistical analyses using data from the NCDS and the BCS along with data from the MCS. Providing a solution to this problem is the second challenge. Drawing on advanced statistical thinking we will seek to develop an effective and practicable technique to perform combined statistical analyses of the three datasets. We will also develop resources to help researchers who wish to combine data from the MCS with other studies with complex designs (examples include the new ESRC funded Life Study and international studies such as Growing up in Australia).

Planned Impact

We are fully aware of the requirement to generate impact from ESRC funded activities. We also understand that impact embraces the diverse ways that research related skills benefit individuals, organisations and nations. We are confident that this project can achieve both social science excellence and high impact outside of academia. We conceptualise that this project will be orientated towards valuable methodological innovation. The proposed project is relatively short in duration and the funding requested is modest by current ESRC standards. We have therefore scaled our impact ambitions accordingly, and we are mindful that the aims of the project must be realistic and achievable. The proposed project has the potential to deliver substantial benefits to non-academic knowledge users.

The substantive aim of the proposed research is to provide an informed evidence base addressing the question 'have socio-economic inequalities in childhood cognitive test scores changed between three British birth cohorts?' This new evidence on social inequalities in childhood will be of interest to policy makers, the third sector, the media and the general public. The dissemination of the substantive outputs of this research will result in conceptual impact by building awareness of the inequalities which can develop in early childhood, and the extent to which these inequalities have changed over time.

The new substantive evidence provided by the proposed project will make an important contribution by providing information that is directly relevant to our understanding of social inequalities, the processes of social stratification, and social mobility. Potential knowledge users who will benefit from this new substantive evidence include third sector organisations concerned with social inequalities (e.g. the Social Mobility Foundation, the Sutton Trust and the Joseph Rowntree Foundation). This new evidence will also be of relevance to government researchers and policy makers when building evidence reviews to inform government policies and priorities regarding childhood inequalities, education and social mobility.

We anticipate that the methodological aims of the proposed research will be of benefit to government and third sector researchers. The methodological element of the proposed project will result in clear prescriptions for cross-cohort comparisons, and the suite of new parental socio-economic measures developed will directly shape future non-academic research. We will develop a toolkit based on these methodological insights that will be made available online. This toolkit will be designed to enable straightforward use by a range of knowledge users (e.g. government researchers, third sector researchers and academic researchers). The methodological insights of the proposed project will bring substantial benefit through facilitating future comparative analyses of the British birth cohort studies, and comparative analyses of other surveys with complex sample designs. This will contribute to capacity building amongst government and third sector researchers, and the increased use of ESRC funded data resources by these groups.
 
Description There is no impact to date, however this will be updated as evidence of impact emerges.
 
Description Analysing and Comparing Large-Scale and Complex Social Survey Data: A Brief But Practical Introduction for Non-Academic Researchers 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Third sector organisations
Results and Impact Analysing Large-Scale and Complex Social Survey Data: A Brief But Practical Introduction for Non-Academic Researchers

This was a free event for non-academic researchers who would like to get a brief practical introduction to analysing large-scale complex social surveys (e.g. Understanding Society and the Millennium Cohort Study).

The UK has a wealth of high quality large-scale social survey data resources. These resources include Understanding Society (the UK Household Longitudinal Study), which is the largest household panel study in the world, and the British Birth Cohort Studies. A positive feature of many modern social surveys is that they have complex designs and selection strategies. These datasets are general infrastructural resources and they are designed so that they can support the widest range of research investigations (for example the analysis of sub-samples and special groups). Complex selection strategies are used to ensure that the datasets provide good representations of populations. The challenge of using surveys with complex designs and selection strategies is that they require techniques that go beyond standard analysis methods. In our experience researchers often find the structure of complex large-scale surveys initially difficult to understand. Researchers who are new to analysing these surveys also report that understanding how best to represent the sample designs and selection strategies is far from obvious.

The overall goal of the event was to provide researchers with a brief, accessible and practical introduction to analysing survey datasets with complex survey structures.

Attendees at this event included researchers from the Department for Education, the Department for Work and Pensions, the NHS and Historic England.
Year(s) Of Engagement Activity 2017
URL http://www2.warwick.ac.uk/fac/soc/sociology/staff/connelly/cognitiveinequalities/events/
 
Description Analysing and Comparing Large-Scale and Complex Social Survey Data: Insights from the Analysis of Socio-Economic Inequalities in Childhood 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Analysing and Comparing Large-Scale and Complex Social Survey Data: Insights from the Analysis of Socio-Economic Inequalities in Childhood

This was a free event for academic researchers who would like to gain a better understanding of analysing and comparing complex large-scale social survey data. The event showcases insights from an ongoing project analysing socio-economic inequalities in childhood.

The UK has a wealth of high quality large-scale social survey data resources which include the famous British birth cohort studies. The older birth cohorts e.g. the National Child Development Study (NCDS) and British Cohort Study (BCS) were collected using a relatively straightforward sampling technique, which selected all of the babies born in one week of the year. The more recent birth cohort the Millennium Cohort Study (MCS) has a more complex design. The MCS was designed to support the widest range of research investigations (for example the analysis of sub-sample and special groups). Complex selection strategies were used to ensure that the dataset provides a good representation of the population. The challenge of using surveys with complex designs and selection strategies is that they require techniques that go beyond standard analysis methods. The focus of this event is the challenges that researchers face when undertaking research that combines data from the older birth cohort studies and the Millennium Cohort Study and showcasing insights from an ongoing project that uses these data sources for analysing socio-economic inequalities in childhood.

Attendees at this event included PhD students from UCL, LSE, Imperial College, Oxford, Goldsmiths, Leeds and Edinburgh. The event also attracted participants from the National Literacy Trust, and NHS England.
Year(s) Of Engagement Activity 2017
URL http://www2.warwick.ac.uk/fac/soc/sociology/staff/connelly/cognitiveinequalities/events/
 
Description MZES Open Science Conference 2019 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact https://www.youtube.com/watch?v=f0j88u1_J5U&t=1166s
Year(s) Of Engagement Activity 2019
URL https://www.mzes.uni-mannheim.de/openscience/
 
Description Tackling Socio-Economic Inequalities in Childhood Test Scores 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Event: 'Tackling Socio-Economic Inequalities in Childhood Test Scores' (March 21st 2017).

This event showcased contemporary data and research findings and explored the opportunities for policy formulation and social change. The event featured research presentations from Dr Roxanne Connelly, Professor Alice Sullivan, Dr Liz Washbrook, and Dr Terry Ng-Knight. Professor Vernon Gayle chaired a discussion featuring non-academic stakeholder key note discussants Claire Harding from the Family and Childhcare Trust and Dr Wanda Wyporska from the Equality Trust.

Attendees at this event included representatives from the Social Mobility Commission, the Department for Education, the Family and Childcare Trust, the Equality Trust and the National Literacy Trust.
Year(s) Of Engagement Activity 2017
URL http://www2.warwick.ac.uk/fac/soc/sociology/staff/connelly/cognitiveinequalities/events/