ERIC Number: ED120231
Record Type: Non-Journal
Publication Date: 1976-Apr
Reference Count: N/A
A Study of Omitted Responses Under the Conventional Correction for Guessing.
Cross, Lawrence H.; Frary, Robert B.
It has been demonstrated that corrected-for-guessing scores will be superior to number-right scores in providing estimates of examinee standing on the trait measured by a multiple-choice test, if it can be assumed that examinees can and will comply with the appropriate directions. The purpose of the present study was to test the validity of that assumption using data collected on 407 students enrolled in a freshman chemistry course at a large state university. The results showed that examinees who omitted items under formula scoring directions were omitting items for which, on the average, they were subsequently able to get significantly more correct than would be expected by chance when forced to guess. Moreover, there was a slight, but statistically significant relationship between success in guessing the correct answer to previously omitted items and two personality measures used in the study. (Author)
Publication Type: Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A