NotesFAQContact Us
Search Tips
ERIC Number: ED310151
Record Type: Non-Journal
Publication Date: 1989-Mar
Pages: 13
Abstractor: N/A
Reference Count: N/A
Things Mother Never Told You about Fit Statistics.
Forster, Fred; Karr, Chad
It is postulated that mean square fit statistics do not adequately evaluate the quality of test items adequately. The fit statistics arise from the desire to have statistical (computerized) procedures to assess item quality transcending the group on which they are computed. It appears that none of the current formulations meet these criteria. The limitations of current fit statistics are demonstrated. Problems identified are related to sample size and to "outliers." It is proposed that psychometricians use the plot of the item characteristic curve in place of fit statistics. Random samples of 300 high achievers, 300 low achievers, and 300 sample students were drawn from the test results of 5,000 eighth graders, who took a 65-item statewide mathematics test, in order to demonstrate the "f" coefficient--a new correlation type coefficient. It is concluded that the "f" coefficient may prove useful in analyzing item characteristic curves and assessing item quality, but is not consistent across samples or for problem items. (TJH)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A