Does the multiple choice question structure in examinations have an effect on student performance?

The existence of multiple-choice questions (MCQs) as an examination method has been around for more than a century. The attraction of MCQ tests/examinations is twofold. They allow a large amount of material to be tested over a short period of time and many cognitive abilities can be assessed. These tests can be marked relatively quickly and accurately. The majority of MCQs are structured with a stem (the question) with options (containing a single key and multiple distractors) (Cohen & Wollack, 2000). In addition the test can allocate negative marks for incorrect choices to avoid guessing or it can have no penalty for guesses. There are no hard and fast rules to determine the optimal number of distractors (Nitko, 2004). The focus of this project is to investigate the effect on student results of changing the structure of MCQs in examinations. In particular the effect of replacing 4 optioned MCQs with negative marking by 6 optioned MCQs with no penalties will be evaluated. The method used in this research was to change the format of the MCQs. Originally the format was 4 alternatives with a mark of +3 for the key, 0 for no answer and -1 for any of the distractors. The new format was to have 6 alternatives with +3 for the key and 0 for no answer or any of the distractors. The data is collected through end of semester final written examinations. The data collected has equal samples of years of average examination results of both negative marking MCQs and non-negative marking MCQs. The data is analysed on the performance of the various year's students and then compared and contrasted. The comparison is carried on both the MCQ section results as well as the total examination result. A student survey is conducted to attain the students' views. The results show that there was an improvement in the performance of the students in the MCQ section, but overall across the whole exam, which includes the MCQ section as well as a short answer section, there was only a marginal improvement. A student survey was conducted to determine the student's views on the different MCQ formats in the examination. The survey shows that 60% prefer to have the 6 alternative MCQs with no penalties. When negative marking could not be avoided 80% of the students chose to have the largest differential possible ie (+4,-1), thereby reducing the impact of negative marking. The effect of changing the structure of MCQs in examinations has shown that the overall results in the examination have not changed greatly statistically. In fact the overall marks improved slightly as you would expect with no negative marking, but of no great significance and yet the students believe that the new format is better for them. The students feel less stressed in doing the examination.