Qualrus – automated essay marking

Writing about web page http://news.bbc.co.uk/1/hi/education/4425423.stm

Responding to student demand, the University of Missouri have introduced an automated essay marking tool. Claims concerning its abilities can be believed. However, does that necessarily mean that it is a good thing?

Main text

The Qualrus system is capable of pattern searching and semantic analysis of an essay, identifying the depth and exactitude of the domain knowledge contained in it. The developers are keen to stress that this only covers one part of the assessment process. A human expert still needs to consider creative and innovative aspects of the essay. That may welll be the case, and indeed could give the markers more opportunity to concentrate on these aspects of the student's work. However, the danger is always there that the machine's assessment will be trusted too much, and the human dimension will be forgotten due to a lack of time.

Unsurprisingly, students in the US have responded positively to the system. After all, it the aspects of their work that the system rewards, repetition of fatcs and content, are much easier to understand and perfect than the higher level skills that the machine cannot analyse. Worse still, it seems that the students are allowed to submit their essays into the system as many times as they like until the deadline is reached. Hence they may be adapting their work to meet the assessment criteria of the machine, which are inevitably limited.

My conjecture is this. Lecturers often do not want to be too explicit about assessment criteria. There is a chance that such over-determination stifles student creativity and innovation, precisely the characteristics of intelligence that higher education is supposed to develop. Machine marking may well act against this.

Robert O'Toole : 09 Apr 2005 22:09 |  Tags: E-Learning Warwick University |  Comments (0) |  Report a problem