Children Learning Group

The Children Learning Research Group has worked on two approaches to quantitative research, drawing on different data sets.

The first draws on the ALSPAC data (Avon Longitudinal Study of Parents and Children) and uses structural equation modelling and regression analyses to predict achievement in Key Stage 2 and Key Stage 3 as well as standardized measures of abilities given in adolescence.

The aim of these studies is to investigate specific predictors of achievement in educational and psychological measures and to tease out the relative importance of different measures in these longitudinal predictions. The specific predictors of interest are consistently measures of knowledge that can be taught; they are considered specific in so far as they predict achievement after controlling for general cognitive abilities as measured  by intelligence and working memory tests. Ultimately, the aim is to identify what sort of teaching might be more effective at earlier ages in promoting achievement later in primary and secondary school. Nunes and Bryant designed the measures that were given to the children when they were between 8 and 10 years. The Key Stage achievement measures were obtained at 11 and 14, and standardised tests of reading comprehension and fluency were given at 10, 12 and 13. In the domain of mathematics achievement, these longitudinal studies showed that mathematical reasoning and number knowledge measured at 8-9 independently predicted the children’s achievement in KS2 and KS3 mathematics after controlling for measures of intelligence and working memory. Reasoning was consistently the more important of the two types of mathematical knowledge. In the domain of literacy, the longitudinal studies showed that children’s use of phonological and morphological knowledge in word reading and spelling predicted their KS2 and KS3 English achievement, after controlling for intelligence. This was also true of outcomes in standardised measures of reading comprehension and fluency. The use of morphological knowledge was consistently a better predictor than the use of phonological knowledge. In science, the analyses showed that children’s understanding of the importance of controlling variables in an experiment predicted their KS2 and KS3 science achievement, after controlling for intelligence. These studies identify possible targets for instruction that could enhance educational outcomes. (References 1, 2 ,4, 5)

The second data set draws on studies about children’s knowledge of fractions, carried out during an ESRC-TLRP project.

We designed a new methodology for identifying profiles of children who are relatively better at procedural skills, relatively better at conceptual knowledge or have a balance in their conceptual and procedural knowledge in the domain of fractions. The method consists in using cluster analysis on the residuals obtained from regressing procedural knowledge against conceptual knowledge and vice versa. These analyses showed that, if the knowledge is unbalanced, conceptual knowledge is a better predictor than procedural skills of positive outcomes in a measure of understanding of intensive quantities, which are measured by rational numbers. This method has already been replicated by other researchers.

Together, the two lines of investigation demonstrate the relative independence of procedural skills and reasoning in mathematics, as well as the consistently stronger effect of reasoning than procedural skills measures in the prediction of educational achievement. (reference 3 & 7)


  1. Bryant, P., Nunes, T., & Barros, R. (2013) The connection between children’s learning about grapho-phonic and morphemic units spelling rules and their achievements at school. British Journal of Educational Psychology, in press. DOI:10.1111/bjep.12030
  2. Bryant, P., Nunes, T., Hillier, J., Gilroy, C., Barros, R. (2013). The importance of being able to deal with variables in learning science. International Journal of Science and Mathematics Education, in press. DOI: 10.1007/s10763-013-9469-x
  3. Hallett, D., Nunes, T., Bryant, P., & Thorpe, C. M. (2012). Individual Differences in Conceptual and Procedural Fraction Understanding: The Role of Abilities and School Experience. Journal of Experimental Child Psychology, 113, 469-486.
  4. Nunes, T., Bryant, P., Barros, R. (2012). The development of word recognition and its significance for later reading skills. Journal of Educational Psychology, 104, 959-973. doi: 10.1037/a0027412.
  5. Nunes, T., Bryant, P., Barros, R., & Sylva, K. (2011). The relative importance of two different mathematical abilities to mathematical achievement. British Journal of Educational Psychology, 82, 136–156.
  6. Gottardis, L., Nunes, T., & Lunt, I. (2011). A Synthesis of Research on Deaf and Hearing Children’s Mathematical Achievement. Deafness & Education International, 13, 131-150.
  7. Hallett, D., Nunes, T., & Bryant, P. (2010). Individual Differences in Conceptual and Procedural Knowledge When Learning Fractions. Journal of Educational Psychology 102, 395-406.