The Cureton-Tukey Method of Equating Test Scores
- Author(s):
- Karon, Bertram P.; Cliff, Rosemary Hayes
- Publication Year:
- 1957
- Report Number:
- RB-57-06
- Source:
- ETS Research Bulletin
- Document Type:
- Report
- Page Count:
- 20
- Subject/Key Words:
- Cureton-Tukey Method, Equated Scores, Equipercentile Equating, Mean and Sigma Method, Statistical Bias
Abstract
Because of its less restrictive assumptions, the equi-percentile method of equating is theoretically preferable to the mean-and-sigma method; unfortunately it involves more sampling error. The Cureton-Tukey procedure for reducing this sampling error by mathematically "smoothing" the samples before equating by the equi-percentile method is described, and empirically evaluated. Although the Cureton-tukey procedure seems promising, none of the differences among the three methods being compared -- the mean-and-sigma method, the equi-percentile method, and the Cureton-Tukey method -- reach statistical significance. It is therefore impossible to draw any definitive conclusions from the limited data employed in this study.
Read More
- Request Copy (specify title and report number, if any)
- http://dx.doi.org/10.1002/j.2333-8504.1957.tb00072.x