The impact of two multiple-choice question formats on the problem-solving strategies used by novices and experts

BACKGROUND Pencil-and-paper examination formats, and specifically the standard, five-option multiple-choice question, have often been questioned as a means for assessing higher-order clinical reasoning or problem solving. This study firstly investigated whether two paper formats with differing number of alternatives (standard five-option and extended-matching questions) can test problem-solving abilities. Secondly, the impact of the alternatives number on psychometrics and problem-solving strategies was examined. METHODS Think-aloud protocols were collected to determine the problem-solving strategy used by experts and non-experts in answering Gastroenterology questions, across the two pencil-and-paper formats. RESULTS The two formats demonstrated equal ability in testing problem-solving abilities, while the number of alternatives did not significantly impact psychometrics or problem-solving strategies utilized. CONCLUSIONS These results support the notion that well-constructed multiple-choice questions can in fact test higher order clinical reasoning. Furthermore, it can be concluded that in testing clinical reasoning, the question stem, or content, remains more important than the number of alternatives.