NotesFAQContact Us
Search Tips
ERIC Number: ED171757
Record Type: RIE
Publication Date: 1979-Mar
Pages: 61
Abstractor: N/A
Reference Count: 0
Using Beta Distributions to Estimate Percentile Ranks and Accumulate Norms for Student Opinion of Teaching Items. Research Report No. 99.
Kolen, Michael J.; And Others
Members of the beta family of distributions were used to estimate percentile ranks and to accumulate normative data collected in a university-wide system for gathering student opinions about teaching--including the areas of course content, objectives, instructor's behavior, teaching methods and materials, and outcomes of instruction. The fitted distributions were found to be acceptably accurate and to provide as accurate or more accurate percentile ranks than the linear interpolation method often used for this purpose. In addition, the estimation method investigated requires storage of only minimal data in order to accumulate (update) norms from year to year. The nature of the results suggests that the beta family may be a useful modeling distribution for norms or population distribution estimates on Likert-type items in other attitude and opinion research areas. (Author/MH)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers: Likert Scales
Note: Paper presented at the Annual Meeting of the American Educational Research Association (63rd, San Francisco, California, April 8-12, 1979)