A single dataset, which brings together a wide range of core variables from across all sweeps of the Millennium Cohort Study (MCS), is now available. The new resource for researchers will make it easier to conduct longitudinal analysis using information collected at different timepoints.
MCS has been mapping the journeys of over 19,000 young people since they were born in 2000-02. Eight data collection sweeps have taken place so far – the first at age 9 months and the most recent at age 23 – and plans are underway for the next, at age 27. The study has gathered thousands of variables on the different aspects of cohort members’ lives – from early childhood development, education and work, to health, wellbeing, identity and more.
A new MCS core dataset has now been created which consolidates key data from all eight sweeps, including many variables that have been measured repeatedly over the years. The aim is to reduce the time researchers need to spend preparing data from different sweeps for analysis.
A single analysis-ready file
Available from the UK Data Service, the MCS core dataset gives researchers access to a wide range of variables in one place, in a single analysis-ready file. It consists of approximately 170 variables, collected over 23 years, and spanning multiple topics and themes, including:
- Demographics – for example, ethnicity
- Socioeconomic circumstances – for example, social class
- Cognitive skills
- Educational attainment
- Physical and mental health – for example, self-rated health and wellbeing
- Body measurements – for example, height and weight
- Behaviours – for example, physical activity and alcohol intake.
Dr Dominic Kelly, Senior Research Fellow at CLS, and part of the team behind the project, said:
One of the most common challenges researchers face when working with the Millennium Cohort Study is the need to merge and clean data across multiple sweeps. The new MCS core dataset aims to address this by bringing these data together into one single analysis-ready file. We hope this will be a valuable resource for the research community.
Dominic Kelly, Senior Research FellowIn the dataset, each cohort member has eight rows of data, one per sweep. Variables that were previously spread across eight separate datasets, which each had to be downloaded separately, are now consolidated into a single, consistently named variable within one file. The core dataset will be updated as future data collections take place, including the upcoming Age 27 Sweep.
Tailoring the dataset to your own project
Researchers can also tailor the MCS core dataset to meet the needs of their specific projects, using code provided on the CLS GitHub repository. This code enables researchers to replicate and adapt the scripts to add new variables not already included and regenerate the dataset with their own bespoke configuration. Full R code is available and STATA code will be added soon.
Making all the CLS studies easier to use
Equivalent core datasets for three other cohort studies managed by CLS are in development and will be available to researchers by the end of 2026, with plans to release the Next Steps core dataset later this summer. These resources are part of a broader programme of work underway to make the rich data collected across all the CLS studies as accessible and user-friendly as possible. Other steps include simplifying the data deposits, so that main survey data are provided in a single deposit rather than multiple deposits (with age sub folders); and applying a consistent naming format to all datasets.
Feedback
To give feedback on the MCS core dataset or to suggest improvements, including any variables that could be added in future updates, please get in touch: clsdata@ucl.ac.uk.
Access the new resources
Read the user guide to the MCS core dataset on the CLS website.
Access the MCS core dataset from the UK Data Service website [SN 8172].
Tailor the core dataset with code from the CLS GitHub repository.
