Retrieval Blocking Supports the Testing Effect with All-of-the-Above Questions on a Multiple-Choice Test

Abstract

Poorly constructed multiple-choice-questions (MCQ) can have dire consequences on students' exam performance (e.g. negative testing effect). For example, either increasing the answer options (Roediger & Marsh, 2005) or including the correct "none-of-the-above" option (Odegard & Koen, 2007) on a MCQ can lead to poor performance. However the impact of “all- of-the-above” (AOTA) option on the testing effect has not been adequately researched. The present study examined the role of the “all-of-the-above" option on the testing effect. The findings were discussed in light of two types of retrieval hypotheses: retrieval blocking and retrieval fluency. Participants read a set of passages and took an intervening MCQ test for half of the passages and reread half of the passages prior to taking a final MCQ test. Results revealed that previous testing with the correct-AOTA option promoted the testing effect whereas the wrong-AOTA option attenuated the testing effect. This finding is consistent with the retrieval blocking theory. This result will shed light on developing test questions that can help students' retention.