The importance of for professional in nursing is increasingly recognized in the USA, UK and elsewhere, but there is no consensus on what is meant by for nursing, nor on how it should be taught, learned and assessed. Meanwhile, studies in various countries indicate a lack of proficiency in of some students and qualified nurses (Hutton, 1997; Meyer, 2004; Sabin, 2001). Against this background, we are undertaking an exploratory interdisciplinary (nursing and education) study to evaluate the teaching, learning and assessment of for nursing in a large pre-qualification program at a school of nursing in England (referred to here as 'the school') and drawing out implications for the teaching, learning and assessment of for nursing in the school and elsewhere. The program leads to registration as a qualified nurse through its recognition by the UK Nursing and Midwifery Council (NMC). Since September 2008, students on such programs must achieve 100% in a test of numeracy in practice (NMC, 2007). However, no national standards for for nursing are specified in the UK (nor elsewhere, as far as we are aware). The body tasked by government with providing educational solutions for workforce development for the National Health Service (NHS) workforce in Scotland, NHS Education for Scotland (NES), is seeking to remedy this situation through the establishment of a benchmark for for nursing at point of registration (Coben et al., 2008). Meanwhile, throughout the UK, Higher Education Institutions (HEIs), including the one in our study, are incorporating tests into their pre-registration nurse education programs in order to meet the NMC requirement. We are concerned that these tests may be of variable quality and, at worst, some may be mathematically and/or professionally inappropriate. In a safety-critical context such as nursing, this could have serious negative consequences for all concerned, not least the patient. Numeracy in Pre-registration Nurse Education in the School The school in our study offers an integrated program in undergraduate nursing studies leading to professional registration after three years full-time study. Numeracy features strongly throughout, in particular through the mandatory 'Calculations for Nursing' element. This focuses on the development of mathematical skills, reviewing some basic mathematics principles and applying these to calculations encountered in clinical practice. Summative assessment of 'Calculations for Nursing' comprises ten multiple-choice questions randomly chosen from the school's item bank and attempted online under examination conditions. Students have 30 minutes to complete the assessment and three opportunities to achieve 100%. In this article we focus on our analysis of the summative assessment. Initial Analysis of Summative Assessment Data We analyzed data from summative tests undertaken by Year One students as part of their normal program: (N [Candidates]=378; N [Items]=41 items). Of the total candidates, 199 (53%) were unsuccessful on their first attempt. In our preliminary analysis, we have examined items facilities (i.e., the percentage of correct answers). These are generally high (between 0.59 and 1; mean: 0.89; median: 0.93), but the tests taken by individual candidates vary considerably in difficulty depending on the 10 items drawn from the item bank. Hence, for purposes of comparison, we have additionally conducted a Rasch analysis. Briefly, the Rasch model is a probabilistic model based on an item response theory (IRT) model (Hambleton, 1993).It can be used to estimate candidate 'abilities' and item difficulties. It assumes that latent variables, ability and difficulty can be measured on the same uni-dimensional interval scale. Given the limitations of both dataset and items, the Rasch model is a good fit. Figure 1 shows the distribution of candidates' ability estimates for pass and fail groups. …