What measures do we need to build an electronic monitoring tool for postgraduate tutor marked assignments?

This paper reports the findings from a preliminary study, which set out to test a number of metrics and heuristics that will form the basis of an electronic monitoring system for postgraduate level assignments. Such a (web-based) system would augment the University’s electronic ‘Tutor Marked Assignment’ (e-TMA) infrastructure with new services. These services would support tutors, monitors, and students as they learn to grade and write essays and reports. This study has illustrated that some readability measures can act as excellent diagnostics, such as picking up over-length essays, but that sentence length is not a good one because of problems concerning the complexity and style of the text. Other heuristics such as ‘indicators of analytical thinking’ formed promising measures that could be included in our system. Overall this study has provided some benchmark metrics to begin to construct a monitoring system that will include both readability metrics and content heuristics from both tutors and students