I have now been a tutor for the Open University for longer than I would often care to admit. During that time the 'how to's' of assessment and formative feedback have been an integral part of my academic life. We were early on instructed in the wonders of the feedback 'sandwich' as a means to draw the student on from simply reading their mark - first, positive comments and encouragement; followed the real nitty-gritty of comment and critique about the way in which the brief has been attempted; bookended by a final, summative, positive comment. This can seem somewhat artificial, and I have been told by some students that they don't read the framing comments because they know the real 'meat' is the critical comments. Good assessment as a (well-prepared) ham sandwich.
Now I discover that I have also been working to a set of ideas about good assessment practice which I knew nothing about: Bales's (1950) interactional categories. Bales suggests four main categories of interaction within feedback: Category A - positive reactions (shows agreement or solidarity); Category B - attempted answers (gives suggestions, opinion, information); Category C - questions (asks for information, opinions, suggestions); and finally, Category D - negative reactions (shows disagreement, tension, antagonism. The parallels between these and the 'sandwich' are clear to see.
Wheeler notes "evidence of systematic connections between different types of tutor comment and the level of attainment in assessment", by which I suppose she means the greater likelihood of Category A in high scoring assessments, and a higher occurrence of Categories B and C in lesser grades. However, the OU definitely doesn't (supposedly) 'do' the Category D: 'negative' comments, which is surely only right, as demoralising weak(er) students is not a good tactic. The probably need more positive comments, even if sometimes it is pretty difficult to do, apart from phrase a criticism in a positive tone.
We have been introduced to an electronic assessment monitor, 'OpenMentor', based upon Bales' Categories. It would seem straight-forward to use, and to offer some possibilities. However, I did have one big problem with the OpenMentor results. This again is founded very much on my OU experience. It was the implication that there had been too much 'teaching' (B responses) in the higher scoring essays. I wonder at the reason for this? Is this because OU TMAs are a rather different entity to essays for conventional f2f/blended learning courses? Does this suggest that OpenMentor is not, in its present incarnation, not really geared toward e-learning/distance learning, which tends to imply that more 'teaching' will be done via comments to the essays? Does this mean that electronic assessment monitors such as Open Mentor might indeed not be not discipline specific, as claimed - but might have to be more context specific?
Reference
Bales, R.F. (1950) 'A set of categories for the analysis of small group interaction', American Sociological Review, vol. 15, pp. 257-63.