DIGITAL LIBRARY
WHAT IS THE BEST TYPE OF MULTIPLE CHOICE QUESTIONS FOR EPIDEMIOLOGY?
Department of medical education and simulation of faculty the Medicine of Porto (PORTUGAL)
About this paper:
Appears in: INTED2016 Proceedings
Publication year: 2016
Pages: 6756-6759
ISBN: 978-84-608-5617-7
ISSN: 2340-1079
doi: 10.21125/inted.2016.0593
Conference name: 10th International Technology, Education and Development Conference
Dates: 7-9 March, 2016
Location: Valencia, Spain
Abstract:
Introduction:
Previous studies have shown differences in the effect of three types of multiple choice questions (MCQs): standard Multiple Choice, Multiple True/False with negative marking and Extending Matching in the quality of medical knowledge assessments. However none of studies focuses on Epidemiology.

Aim:
Study which type of test format (Standard Multiple Choice (SMC) vs Multiple True/False with negative marking (MTF) vs Extended Matching (EM)) is the best to assess students in Epidemiology.

Materials and Methods:
The analysed data are the results of two Epidemiology assessments divided into two parts each one. The 2010 examination had 25 standard multiple choice questions (first part) and 25 questions MTF each having 5 true or false choices with negative marking (second part).
The 2013 examination had 40 questions of SMC (first part) and 15 questions of Extended Matching (second part). The examination of 2010 assessed 212 students and the 2013 assessed 294 students.
The psychometric quality of the test was analysed according to the Classical Test Theory (CTT).

Results:
The first part of the 2010 test had, in average, a difficulty index of 0.47±0.23 and the second part 0.34±0.13 (p=0.002). In addition, the SMC questions of the assessment had, in average, a discrimination index of 0.22±0.11, while in the MTF questions the discrimination index was 0.23±0.10 (p=0.772). In concern to the Cronbach's alpha, the first part had a Cronbach's alpha of 0.67, while in the second it had a value of 0.68.
In concern to the 2013 examination, the difficulty index was 0.68±0.18 in the SMC part and 0.68±0.25 in the EM part (p=0.616). The discrimination index of the SMC part was 0.28±0.11 and of the EM part was 0.22±0.11 (p=0.034). In respect to the Cronbach's alpha adjusted to 25 items, the first part had a Cronbach's alpha of 0.73, while in the second had a value of 0.68.

Conclusions:
The best type of MCQ for epidemiology was the SMC question. The SMCQ were easier to answer compared to the MTF and had better discrimination compared to EM questions.
Keywords:
MTF test format, SMC test format, EM test format, CTT, Difficulty Index, Discrimination Index and Cronbach's alpha.