How to Design and Interpret a Multiple-Choice-Question Test: A Probabilistic Approach.

Autor: Zhao, Y.
Předmět:
Zdroj: International Journal of Engineering Education; 2006, Vol. 22 Issue 6, p1281-1286, 6p, 5 Charts, 1 Graph
Abstrakt: Multiple choice question (MCQ) tests involve an element of guesswork, which affects the reliability and interpretation of test scores. This article studies the probability of obtaining a certain score by pure guesswork and introduces a conversion scheme which converts raw test scores into standard percentage marks. The probabilistic analysis shows that the optimum number of choices of answers for MCQ questions is four, and for a four-choice question test, increasing from 8 questions to 18 and 48 questions reduces the probability of obtaining a converted mark above 40 by pure guesswork from about 5% to below 1% and 0.01%, respectively. [ABSTRACT FROM AUTHOR]
Databáze: Supplemental Index