The Nuffield Learning to Read Project

Do language skills in infancy predict how well children will learn to read? Are children who are slow to learn to produce and comprehend words at risk of later problems with language and literacy, once they reach primary school? Kate Nation and Kim Plunkett set out to explore these questions in the Nuffield Learning to Read Project. Fiona Duff and Gurpreet Reen worked hard on the project, completing it in 2015. We are grateful to the many families who participated in our study, and to The Nuffield Foundation who funded this research.

What was the purpose of this research?

The Oxford Communicative Development Inventory (OCDI) is a parental checklist of how many words infants can understand and say. Many parents have filled in these checklists when taking part in research at the Oxford University BabyLab. We followed-up some of these babies to look at their reading and language skills once they were at primary school. We wanted to see whether vocabulary skills measured in infancy, as measured by the OCDI, can predict school-age language and literacy skills. If it did, this would support the idea that OCDI vocabulary scores could be used to identify children who might be at risk for later language or literacy difficulties.

What did we do and what did we find out?

Between 2012 and 2014 we managed to visit 300 4-9 year children in 150 different schools in and around Oxfordshire. All these children were previously seen in the BabyLab when they were 1-2 years old. They completed a battery of reading and language tasks. Our data analyses supported these conclusions:

  1. In general, those infants who had larger vocabularies in infancy went on to achieve higher levels of language and literacy in primary school.
  2. However, the relationship between infant skills and school-age performance was not strong enough to support the recommendation that the OCDI be used to identify individual infants who might be at significant risk for later language and literacy difficulties.
  3. Importantly, we found that the prediction of which infants might go on to have reading difficulties was significantly improved by considering their family history. Infants with smaller vocabularies who came from a family where there was a history of reading or language difficulties were more likely to show reading difficulties themselves than children without family risk.
  4. In collaboration with Dorothy Bishop, we took a closer look at those children identified in infancy as late talkers. In general, these children were at no greater risk than average talkers. However, those children who continued to show language difficulties at 4 years of age were at significant risk for reading and language impairments later on in primary school.


The Nuffield Summary is a more general overview of our study and a summary of its main findings is downloadable here.

Fiona's talk presented at a meeting of the Society for the Scientific Study of Reading can be downloaded here.

A more general talk "The Language Foundations of Reading" can be downloaded here.

For parents and families - thank you for your support. Our research is not possible without you.

And visit our page at the Nuffield Foundation for links to other research projects addressing important questions about children's language, literacy and learning.

Leave a Reply