The comparability between modular and non-modular examinations at GCE Advanced level
The prime concern of this thesis is the comparability of two types of assessment now prevalent in Advanced level GeE examinations. The more conventional linear scheme assesses all candidates terminally, and the only way to improve the grade awarded is to re-take the whole examination. In contrast, the relatively new modular schemes of assessment include testing opportunities throughout the course of study. This not only has formative effects but allows quantifiable improvements in syllabus results through the medium of the resit option. There are obvious differences between the two schemes, but this does not necessarily imply that they are not comparable in their grading standards. It is this standard which the thesis attempts to address by considering the different variabilities of each of the schemes, and how these might impinge upon the outcomes of the grading process as evidenced in the final grade distributions. A key issue is that of legitimate and illegitimate variabilities - the former perhaps allowing an improvement in performance while maintaining grading standards; the latter possibly affecting the grading standard because its effect was not fully taken into account in the awarding process. By looking at a linear and modular syllabus in mathematics, the differences between the two are investigated, and although not fully generalisable, it is clear that many of the worries which were advanced when modular schemes were first introduced are groundless. Most candidates are seen to use the testing flexibility to their advantage, but there is little evidence of over-testing. Perhaps the major finding is a negative one - that there is no clear evidence for any difference in grading standards between modular and linear schemes of assessment, although there are variabilities which go some way to explaining what appears to be enhanced performances by some of the weaker modular candidates.