This article proposes that sampling design effects have potentially huge unrecognized impacts on the results reported by large-scale district and state assessments in the United States. When design effects are unrecognized and unaccounted for they lead to underestimating the sampling error in item and test statistics. Underestimating the sampling…

Sampling errors limit the accuracy with which forms can be linked. Limitations on accuracy are especially important in testing programs in which a very large number of forms are employed. Standard inequalities in mathematical statistics may be used to establish lower bounds on the achievable inking accuracy. To illustrate results, a variety of…

Investigated empirically through post mortem item-examinee sampling were the relative merits of two alternative procedures for allocating items to subtests in multiple matrix sampling and the feasibility of using the jackknife in approximating standard errors of estimate. The results indicate clearly that a partially balanced incomplete block…

Transformations or equating of raw test scores on two or more forms of the same test are made interchangeable by empirical procedures deriving the standard error of an equipercentile equating for four different situations. Some numerical results are checked by Monte Carlo methods. Numerical standard errors are computed for two sets of real data.…

Item response theory (IRT) model parameter estimates have considerable merit and open up new directions for test development, but misleading results are often obtained because of errors in the item parameter estimates. The problem of the effects of item parameter estimation errors on the test development process is discussed, and the seriousness…

A debate concerning specific issues and the general usefulness of the Rasch latent trait test model is continued. Methods of estimation, necessary sample size, and the applicability of the model are discussed. (JKS)

Statements made in a previous article of this journal concerning the Rasch latent trait test model are questioned. Methods of estimation, necessary sample sizes, several formuli, and the general usefulness of the Rasch model are discussed. (JKS)

Use of clustering methods with data from the 1987 High School Transcript (HST) component of the 1986 National Assessment of Educational Progress (NAEP) is discussed in an attempt to persuade data analysts and researchers that they should be interested in problems traditionally turned over to samplers. Nearly all surveys used by the National Center…

With the increasing globalization of human resources practices, a commensurate increase in demand has occurred for multi-language ("global") personality norms for use in selection and development efforts. The combination of data from multiple translations of a personality assessment into a single norm engenders error from multiple sources. This…

The National Assessment of Educational Progress (NAEP) currently tests seventeen-year-old students enrolled in public and private secondary schools, but it does not test "out-of-school" seventeen-year-olds who have either graduated or dropped out. Estimating that one of five seventeen-year-olds is out of school, the interpretability of…

Linear and equipercentile equating conversions were developed for two forms of the Graduate Record Examinations (GRE) quantitative test and the verbal-plus-quantitative test. From a very large sample of students taking the GRE in October 1981, subpopulations were selected with respect to race, sex, field of study, and level of performance (defined…