One of the things I’ve found time and time again in my investigations into student engagement with e-assessment is that little things can make a difference. Therefore the research done by Matt Haigh of Cambridge Assessment into the impact of question format, which I’ve heard Matt speak about a couple of times, most recently at CAA 2011, was well overdue. It’s hard to believe that so few people have done work in this area.
Matt compared the difficulty (as measured by performance on the questions) of ten pairs of question types e.g. with or without a picture, drag and drop vs tick box, drag and drop vs drop-down selection, multiple-choice with only single selection allowed vs multiple-choice with multiple selections enabled, when adminstered to 112 students at secondary schools in England. In each case the actual question asked was identical. The quantitative evaluation was followed by focus group discussions.
This work is very relevant to what we do at the OU (since, for example, we use drop-down selection as the replacement for drag and drop questions for students who need to use a screen reader to attempt the questions). Happily, Matt’s main conclusion was the variations of item format explored here had very little impact on difficulty – even when there appeared to have been some difference this was not statistically significant. The focus group discussions led to general insight into what makes a question difficult (not surprisingly ‘lack of clarity’ came top) and also to some suggestions for the observed differences and lack of differences in difficulty in the parallel forms of the questions.
I’d very much like to do some work in this area myself, looking at the impact of item format on our rather different (and vast) student population. I’d also like to observe people doing questions in parallel formats, so see what clues that might give.