Evaluation of Negation in Stems of Multiple-Choice Items

P. H. Harasym, P. G. Price, R. Brant, C. Violato, F. L. Lorscheider

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

The opinion of some test experts is that negation (e.g., not, except) should be used sparingly or avoided in the stems of multiplechoice items. In these items, examinees are required to select the one alternative that is false from among those that are true. However, negation continues to be used and frequently accounts for 15-20% of the multiple-choice items within a given medical examination. A 90-item multiple-choice examination was administered in various formats to approximately 200 nursing students in each of two consecutive academic years. The findings of the present study provide evidence that the use of negation in stems should be limited and that the single-response, negatively worded (SRNW) item should often be converted to a multiple-response, positively worded (MRPW) item. This recommendation is based on the inherent cuing (ie., eliminating some alternatives by comparative deduction) within SRNW items and on an increase in the reliability and concurrent validity of MRPW test scores.

Original languageEnglish (US)
Pages (from-to)198-220
Number of pages23
JournalEvaluation & the Health Professions
Volume15
Issue number2
DOIs
StatePublished - Jun 1992

Fingerprint

Dive into the research topics of 'Evaluation of Negation in Stems of Multiple-Choice Items'. Together they form a unique fingerprint.

Cite this