ERIC Number: ED400299
Record Type: Non-Journal
Publication Date: 1990-Apr
Reference Count: N/A
An Item-Level Analysis of "None of the Above."
Rich, Charles E.; Johanson, George A.
Despite the existence of little empirical evidence for their effectiveness, many techniques have been suggested for writing multiple-choice items. The option "none of the above" (NA) has been widely used although a recent review of empirical studies of NA suggests that, while generally decreasing the difficulty index, NA also decreases discrimination and may decrease reliability. It is suggested that most of the studies of the effect of NA on these item parameters have been flawed by methodological inconsistencies and by a disregard for the finding that discrimination is restricted when corresponding item difficulties have been extremely high or low values. By examining the effects of NA on difficulty and discrimination indices in light of optimal difficulty for a 100-item test taken by 300 undergraduates, this study found that when following reasonable guidelines: (1) difficulty tended to approach the optimal level; (2) discrimination tended to increase; and (3) reliability was unaffected. (Contains 6 tables and 24 references.) (Author/SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A