Ever have the experience of having this huge urge to change answers during the last few minutes of an examination, especially those Multiple Choice Questions ones? It is usually at such moments when panic sets in and many answers seems right suddenly.
I have been wanting to understand why this happens to me, psychologically, and a Google search actually brought me contrasting views:
- From Tips on Taking Multiple-Choice Tests:
- Even though first answers are often correct, you shouldn’t be afraid to change your original answer if, upon reflection, it seems wrong to you. Dozens of studies over the past 70 years have found that students who change dubious answers usually improve their test scores. For example, a May, 2005, study of 1,561 introductory psychology midterm exams found that when students changed their answers , they went from wrong to right 51% of the time, right to wrong 25% of the time, and wrong to a different wrong answer 23% of the time (Journal of Personality and Social Psychology, Vol. 88, 725-735).
- From Suggestions for Coping with Multiple Choice Questions on… (Long title :S)
- A good rule of thumb to follow once you have selected an answer alternative is: BE WARY OF CHANGING YOUR MIND. There is evidence to suggest that students more frequently change right answers to wrong ones than wrong answers to right ones.
Alright, I didn’t manage to find the psychological explanation to that bad habit of mind; but the statistics from the first quoted page is actually quite encouraging (and official since its reference is properly cited; unlike the suspicious “there is evidence…” from the second article). To think a research was actually conducted on 1561 students on the effects of changing their MCQ answers!
That research reminded me of a MCQ-tests-related research performed by one of my lecturer (an entertaining one to be exact) from last semester. “Secure Internet examination system based on video monitoring” by C.C. Ko (this is the entertaining lecturer) and C.D. Cheng is an article on how to authenticate a student’s identity in a web-based examination. In that research, 450 students
were exploited volunteered for a randomized multiple choice test in a course on analog and digital signals.
Sounds familiar, fellow EE2009 people? We were in some similar tests, weren’t we?! Now we know how all the students in various tests were obtained…