Research Matters 18


Recent Submissions

Now showing 1 - 9 of 9
  • ItemPublished versionOpen Access
    Research Matters 18: Summer 2014
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Wilson, Frances
    Research Matters is a free biannual publication which allows Cambridge University Press & Assessment to share its assessment research, in a range of fields, with the wider assessment community. 
  • ItemPublished versionOpen Access
    Students' views and experiences of A level module re-sits
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Gill, Tim; Suto, Irenka
    In this study we obtained over 1,300 A level students' views and experiences of re-sits in Psychology and Mathematics, prior to a reduction in re-sit opportunities taking effect nationally. The aim in collecting the data was to gain an understanding of what the likely effects of a system of reduced re-sits would be on students and their teachers. We focused on two popular but contrasting A level subjects: Psychology and Mathematics. We found that one of the students' most common reasons for re-sitting could be seen as a valid means of getting a higher grade. However, most students who responded to the questionnaire gave multiple reasons for re-sitting a module. In each subject, a majority thought that re-sits had both made them work harder, and increased their knowledge of the subject. These views indicate that module examinations do not only provide summative assessment, but are also used for formative assessment purposes too.
  • ItemPublished versionOpen Access
    Multivariate representations of subject difficulty
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Bramley, Tom
    This study compared the Kelly method, the Rasch partial credit model, and multidimensional scaling (MDS) for representing similarities and differences in the grade distributions of different A level subjects. Although there were differences in the patterns across the MDS representations depending on which index was used to measure similarity among A levels, at a broad level the same findings were observed - that is, STEM subjects, Languages, and Humanities clustered together fairly well in the 2-D and 3-D representations, Expressive and Applied subjects less well. However, the conclusion was that a 2-D plot of difficulty against fit to the Rasch model is the most informative way of visually representing the different subjects.
  • ItemPublished versionOpen Access
    Calculating the reliability of complex qualifications
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Benton, Tom
    Most traditional methods of calculating reliability cannot be applied to complex qualifications that can be completed through multiple routes. For example, for some Maths A levels, candidates can choose when the take the various exam papers that are required, and, in addition, can choose the optional subjects in which they wish to complete exams. This article demonstrates a method by which reliability can be calculated in these instances by applying an optimal method of split halves to each individual assessment that may contribute to the qualification. All of these split halves can then be combined together to create "half qualifications" for each candidate regardless of the route they have taken. Once this have been achieved, traditional methods to calculate reliability can be applied. A full example from a Maths A level is provided.
  • ItemPublished versionOpen Access
    Do Cambridge Nationals support progression to further studies at school or college, to higher education courses and to work-based learning?
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Vidal Rodeiro, Carmen
    Cambridge Nationals are vocationally-related qualifications that take an engaging, practical and inspiring approach to learning and assessment. They are industry-relevant, geared to key sector requirements and popular with schools and colleges because they suit a broad range of learning styles and abilities. The present research set out to investigate the characteristics of the learners who were awarded Cambridge National qualifications in the academic years from 2006/07 to 2008/09 (e.g., age, prior attainment, socio-economic background and centre attended) and where they progressed on completion. The outcomes of this work showed that Cambridge Nationals enabled learners to progress in a variety of ways (to further studies at school or college, to work-based learning and to higher education) and therefore are an important contribution to the 14-19 curriculum.
  • ItemPublished versionOpen Access
    Big data and social media analytics
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Dhawan, Vikas; Zanini, Nadir
    Big data' is fast becoming an area of great importance for businesses in many areas, including education. In simple terms it refers to the combination of data from various sources and understanding patterns in the data which can be used for various purposes such as improving market intelligence and educational research. In this article we give an introduction to big data and some of its applications in various fields, including education. We also describe the use of big data for the monitoring of social media.
  • ItemPublished versionOpen Access
    An investigation of the effect of early entry on overall GCSE performance, using a propensity score matching method
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Gill, Tim
    Previous research has shown (Gill, 2013) that certain groups of students performed worse than expected in some GCSE subjects when they were taken early (i.e. in Year 10). However, one possible reason for taking a GCSE exam early is to 'get it out of the way' to enable increased focus on other subjects in Year 11. This study used a propensity score matching method to investigate whether students entering early for GCSEs performed better or worse across all their GCSEs (or equivalents) than those who did not enter for any GCSEs early. In terms of overall GCSE performance, there did not seem to be any advantage in early entry after accounting for differences in the characteristics of the students in the two groups. However, when looking at all qualifications (including non-GCSEs), early entry students did perform better than those not taking any GCSEs early, to a statistically significant degree. Furthermore, early entry students were more likely to pass the five A* to C threshold measure.
  • ItemPublished versionOpen Access
    An intra-board comparison at syllabus level based on outcomes of rank-ordering exercises at component level
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Yim, Louis
    This article describes a variant comparability methodology which uses the rank-ordering method at component level to derive results at syllabus level for intra-board comparison. In other words, instead of judges holding several components' information about each candidate in their minds and making a holistic evaluation of individual candidates during comparison, judges only rank-order candidates' performance within each prescribed component. The final rank order at syllabus level of each judge is derived based on his/her component level's rank orders.
  • ItemPublished versionOpen Access
    An analysis of the unit and topic choices made in an OCR A level History course
    (Research Division, Cambridge University Press & Assessment, 2014-06-01) Child, Simon; Darlington, Ellie; Gill, Tim
    This study aimed to explore how schools that offer A level History use the options available to them, in terms of unit and topic choices. Specifically, this study aimed to determine which units and topics were most commonly taught. It was intended that this data would help establish how optionality within A level History is used, and whether it meets the desired purpose of exposing students to a broad range of historical periods and topics. Data was collated using a survey of 90 heads of history departments, and from an analysis of topic choices within one A level history unit. Comparisons were made between different school types (state vs independent), and schools with different levels of performance. Approximately 60% of centres sampled taught either a combination of F961B and F964B or F962B and F963B; the two unit combinations which permit Modern History to be studied exclusively. In terms of topic choice, it was found that schools seek to teach in-depth within a historical era, rather than breadth over different historical periods. The findings are discussed in relation to the ongoing reforms to A level history.