In 1969, more than 10,000 11-year-olds taking part in the National Child Development Study wrote an essay imagining what their lives would be like by the time they were 25.

Fast forward 50 years, and a team of researchers from the Centre for Longitudinal Studies in London, Stonybrook University in New York, and University of Melbourne has been undertaking a major project to analyse the content of these essays to see how the language of children can provide vital clues to their future lives.

Using machine learning tools, and drawing on other data collected from the study members over the decades, the researchers have been looking at the predictive power of the words and concepts in the children’s essays. They have explored the extent to which the language used points to their economic status, their physical activity, physical and mental health and cognitive function in middle age.

This is the first time this type of analysis has been done on open responses collected from a large national birth cohort study.

A key part of the project was to digitally transcribe all 10,000 plus essays written in 1969. The essay transcriptions are now available to the wider research community for the first time, from the UK Data Service.