This research project investigates the development of the link between number words and number symbols, with the aim to identify potential risk factors for low numeracy. Children entering primary school need to shift from using mainly spoken numbers words (e.g., 'twenty-three') to being able to write down the correct multi-digit Arabic strings (e.g., '23') upon hearing spoken number words. This is called number word transcoding and is fundamental for mathematical development. This project investigates three aspects of transcoding between number words and Arabic digits. In this project we measure the efficiency of the mapping in children of different ages. We also measure children's mathematical ability to see whether the efficiency of the mapping is related to their mathematical ability. We follow children from Year 1 to Year 3 and measure their ability to transcode multi-digit numbers as well as other number skills every year as well as testing their maths performance. This study aims to find out whether a child's ability to transcode multi-digits accurately in Year 1 predicts how their mathematical ability will develop over the next few years better than other measures of numerical skills. Finally, we investigate the influence of language on number transcoding by comparing monolingual children from two language backgrounds: English and German.Numeracy is the Cinderella of literacy, much less researched but at least as important for employment and life success as literacy. This research project will investigate the development of a fundamental principle of numeracy, the link between number words and number symbols, with the aim to identify potential risk factors for low numeracy. Children entering primary school need to shift from using mainly spoken numbers words (e.g., 'twenty-three') to being able to write down the correct multi-digit Arabic strings (e.g., '23') upon hearing spoken number words. This is called number word transcoding and fundamental for mathematical development. This project will investigate three aspects of transcoding between number words and Arabic digits. First, we will analyse the mapping between single Arabic digits (e.g., '8') and the corresponding number words (e.g., 'eight'), because the single Arabic digits are the basic building blocks for all Arabic multi-digit strings (e.g. 28732). Most children in primary school already know the single Arabic digits. We propose that the efficiency with which children map between digits and number words is still increasing in primary school. In this project we will test this by measuring the efficiency of the mapping in children of different ages. We will also measure children's mathematical ability to see whether the efficiency of the mapping is related to their mathematical ability. Secondly, this project will investigate multi-digit number transcoding. This is important because most of the numbers children are dealing with in primary school are multi-digit numbers (e.g., 2015) and instructions are often given verbally (e.g., 'thirty-two'). We will follow children from Year 1 to Year 3 and measure their ability to transcode multi-digit numbers as well as other number skills every year. We will also test their maths performance. This study will enable us to decide whether a child's ability to transcode multi-digits accurately in Year 1 predicts how their mathematical ability will develop over the next few years better than other measures of numerical skills. Thirdly, number words depend on the language a person speaks. Non-native children entering primary school will be much less familiar with native number words and their mapping to the Arabic digits. Given the multilingual context of many children in primary school in the UK and across Europe, it is important to find out whether and if so, how, language factors influence the association between spoken number words and Arabic digits. We will investigate the influence of language on number transcoding by comparing monolingual children from two language backgrounds: English and German. In addition to increasing our knowledge about the process of number transcoding, this research project will lead to new ways of assessing children's number skills and will in the long run help native and non-native children who struggle with maths.

Data were collected in groups and individually. Cognitive tasks in the form of standardised assessments and bespoke tests were used. Assessments were administered by pen and paper as well as using a computer. For a detailed description of the Methodology at each time point see Documentation files. In the UK sample, 257 children took part at T3, of which 232 had complete data. The children were followed from Year 1 to Year 3 of primary school. Of those with complete data, 217 also had complete data from T2, and 195 from T1 and T2. The children at T3 were between seven and eight years old, at T2 they were between six and seven and at T1 between five and six. Of the 196 children with complete data, 99 (50.5%) were males and 97 (49.5%) females. From the sample tested, six of the children who completed all three time points were classed as having English as an additional language (EAL) or as bilingual by the school records. Children were recruited from both suburban and rural primary schools in areas across North and West Yorkshire. Children with developmental disorders and/or neurological disorders were not excluded from the study. For more details about the population please see the Methodology for each time point in the UK Documentation folder. In the Austrian sample, 168 children took part at T3, of which all 168 had complete T3 data. The children were assessed in Year 1, Year 2 and Year 3 of Austrian Primary Schools. Of those with complete data, 167 also had complete data from T2, and 166 from T1 and T2. The children at T3 were between eight and nine years old, at T2 between seven and eight years old and at T1 between six and seven years old. Of the 166 children with complete data at all time points, 88 (53%) were males and 78 (47%) females. Out of the 166 children, 22 spoke a first language other than German. SEN data was not collected for this sample. For more information please see the Methodology at each time point in the Graz Documentation folder. In the UK, 74 primary schools in York, surrounding areas in North Yorkshire, Leeds, Harrogate and Whitby were contacted via email and invited to take part. The schools approached were a mix of community, voluntary aided, voluntary controlled, maintained and independent schools as well as academies. In each of these areas testing was undertaken in all schools that were interested in taking part. In the end, at the first time point, 11 schools agreed to take part in the study. In Austria, the team contacted 13 schools in the city of Graz that were in proximity to the university. Of these, five schools agreed to take part in the study.