Abstract
Frequency estimation, also called poststratification, is an equating method used under the common-item nonequivalent groups design. A modified frequency estimation method is proposed here, based on altering one of the traditional assumptions in frequency estimation in order to correct for equating bias. A simulation study was carried out to compare equating errors for the modified frequency estimation method, the traditional frequency estimation method, and the chained equipercentile method. The results show that the modified frequency estimation method performs better than the traditional frequency estimation method and, under most circumstances, also performs better than the chained equipercentile method.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have