As a teacher of the IB Diploma, I supervise a couple of Extended Essays each year and my classes Theory of Knowledge essays. The criteria used for the TOK essay is given below and is very much in line with the ‘adverb soup’ problem described by many others (thoughtful v discerning and ‘focus and ‘sustained focus’). The IBO advocate a global impression marking approach, but is it possible to assess and essay across grade bands for differences such as relevance, lucidity, organization and coherence? I’ve always felt, but never really pushed the idea, that I’m not exactly sure how many of those terms I would be confident in defining clearly (in the context of the task) let alone objectively assessing them together within an essay. Also, how to manage if some seem to be present and some don’t- which is ‘worth’ more or weighted more favorably? What is interesting is that the essays themselves are ‘blind marked’ by the IBO once uploaded to their system- we don’t actually submit any marks to be moderated. The essays are treat as an exam- the mark scheme is used, in combination with a presentation mark, to give an estimate for a TOK grade.
I’ve read with interest the posts looking at NoMoreMarking, which is an essay comparison tool, and was keen to try this out. The site allows a user to force a comparison between two essays by judging (rather than marking) the two side by side- which is best (our concept of quality)? Once we have a rough ordering of the essays, we can start to ask: What makes particular essays better than others? I thought I’d try this with my Y9 science class, who had just submitted short essays as a summative assessment piece.
The set up and upload process was quite straight forward (although I had to switch to Chrome as the upload wouldn’t work in Firefox). I share the group with a colleague so I set us both up as judges. I managed to rattle through the essays (26 students) in about 50 minutes. I will update once my colleague has completed his judging. I will then then insert some moderated grades into the system and try to generate a mark for each. We will then compare this to marking them by criteria (which two colleagues have kindly offered to do).