Concern over the validity of statistical tests performed on data that may not satisfy underlying assumptions has prompted methodological researchers to perform Monte Carlo studies for frequently used tests. Unfortunately, these studies appear to have had little impact on methodological practice. One reason is the lack of an overarching framework…

A balanced appraisal of holistic scoring of writing is presented via: examination of the present popularity of holistic scoring; analysis of several weaknesses associated with the holistic scoring method; and recommendations for remedying these weaknesses. Six reasons for the popularity of holistic scoring are: (1) relative lack of expense; (2)…

This paper summarizes two related sets of data. First, the findings of an inquiry titled, "The Appropriateness of Forced Choice True/False Response Formats for Selected Learning Style Inventories" is reported. This inquiry compared data from two forms of the Dunn/Price Learning Style Inventory. The first form was the real Dunn/Price…

The report describes a federally funded study of the application of language proficiency concepts, developed for commonly taught languages, to less commonly taught languages. It summarizes the project's background and origins and the activities of the project, including four separate studies on the feasibility of proficiency-based instruction and…

The problems of validity and fairness involved in multiple-choice critical thinking tests can be lessened by using verbal reports of examinees' thinking during the process of developing such tests in order to retain only those items which rely on critical thinking skills to obtain the correct answer. Multiple-choice testing can lead to unfair…

The problem of sex bias in mathematics word problems is discussed, with references to the appropriate literature. Word problems are assessed via cognitive science analysis of word problem solving. It has been suggested that five basic semantic relations are adequate to classify nearly all story problems, namely, change, combine, compare, vary, and…

Although research on bilingualism in the schools shows the importance of measuring both English and Spanish language skills regularly, the testing of Spanish falls behind in both quality and frequency. Concerted multidisciplinary cooperation among bilingual educators, linguists, and psychometricians is now needed to produce a variety of valid…

The occurrence of discrepant, or non-contiguous, scores during holistic scoring of writing proficiency is usually blamed on variables of the test reader or external factors. It is hypothesized that characteristics of the essays themselves sometimes may generate discrepancies in essay scoring. Data from 152 essays of college sophomores in teacher…

An attempt was made to determine whether a non-response bias exists in the annual follow-up surveys of graduates of teacher education programs when the surveys focus on more than employment. Subjects were 291 graduates of teacher certification programs at a major university in the southeastern United States from Fall 1986 through Summer 1987. In…

The paper states that quantification is neither ethical nor unethical, but is ethically neutral. It is the behavior or intent of the human being that is clearly a matter of ethical concern. Like numerology and the sects of inumerates and qualitatives, there is not so much an unethical practice that is supported as there is a lack of vision and…

To determine how much prior knowledge affects students' performance on a reading comprehension test, a study was undertaken to examine the prior knowledge factor in a reading competency test developed for college freshmen in New Jersey. The passages used in this test are based on common, everyday topics. Subjects were students in two sections of a…

Problems in validating ability tests for handicapped students and research approaches to predictive validity are discussed. Validity for handicapped persons tested under regular conditions; for applicants to special programs, and for tests taken under special administrative conditions are considered. Item analysis and the construction of new…

A common problem arises in scale transformation when independent estimates of item parameters from two separate data sets must be expressed in the same metric. These item parameter estimates will be different because the metric or scale defined by each independent calibration of the items is different. The problem is frequently confronted in…

Differential item performance (DIP) is discussed as a concept that does not necessarily imply item bias or unfairness to subgroups of examinees. With curriculum-based achievement tests, DIP is presented as a valid reflection of group differences in requisite skills and instruction. Using data from a national testing of the ACT Assessment, this…

This study determines whether Black students who have been placed in an EMR program on the basis of WISC full scale IQ scores as the primary criterion measure would be ruled out when tested on the BITCH. Results indicate that students obtain similar scores on both tests. (Author)