Score equating based on small samples of examinees is often inaccurate for the examinee populations. We conducted a series of resampling studies to investigate the accuracy of five methods of equating in a common-item design. The methods were chained equipercentile equating of smoothed distributions, chained linear equating, chained mean equating,…

Binomial models for domain-referenced testing are compared, emphasizing the assumptions underlying the beta-binomial model. Advantages and disadvantages are discussed. A proposed item sampling model is presented which takes the effect of guessing into account. (Author/CM)

Both criterion-referenced testing and behavioral assessment share the basic assumption that test behavior is a sample rather than a sign. In addition, both types of assessment focus on response capabilities and performance in specified content domains. Although content validity has been traditionally recognized as essential to criterion-referenced…

The usefulness of item sampling with unobtrusive measures and observational data is demonstrated. Particular emphasis is given to the question: Is item sampling appropriate for estimating the differences between means as well as the means themselves? A descriptive analysis is made of one aspect of a junior high school curriculum. Twenty four…

As an alternative to a classical test theory basis for criterion-referenced test construction, it is proposed that a strict item-sampling model be used. The computer's role in such a model is outlined. The assumptions of the model are carefully defined and its properties reviewed. The relationship between mastery criteria and such sampling plans…

Representing a set of required readings for those who would keep abreast of developments in the field of educational evaluation, this book presents nine chapters, each written by a recognized scholar on the topic being considered. Several chapters are also available as separate booklets, thereby increasing the flexibility with which the materials…

The measurement information generated by CRT's is designed for use in instructional management systems where classifications of pupils for treatment are to be decided on the basis of minimal data consistent with predetermined limits for the errors of misclassification. The measures obtained are content specific estimates of proficiency useful for…

The central assumption in domain-referenced testing (DRT), as presented in this book, is that a domain may be determined which adequately represents a particular universe of knowledge. After a domain has been established, the technological and practical problem of using domain-referenced testing must be solved. This book contains a collection of…

A central problem for the user of domain-referenced tests in instruction is deciding who has passed and who has failed. Two procedures were presented and discussed. The first, employing classical test theory, was found to be more useful for larger domains and where the passing standard is 70 percent or less. The sampling procedure suggested by…