Although International Association for the Evaluation of Educational Achievement-pioneered international large-scale assessment (ILSA) of education is now a well-established science, non-practitioners and many users often substantially misunderstand how large-scale assessments are conducted, what questions and challenges they are designed to…

Use of Tukey's Jackknife in establishing a confidence interval around the population coefficient alpha is explored and the robustness of Feldt's procedure along with ten variants of the Jackknife when the data do not conform to the necessary normality requirements are evaluated. Only two of the variants compared to Feldt's approach. (RC)

This investigation had two major purposes. The first was to explore the use of an inferential technique called Tukey's Jackknife in establishing a confidence interval about cooefficient alpha reliability. The second purpose was to study the robustness of the Feldt and the jackknife procedures when the data fails to satisfy usual normality…

The purpose of this paper is to prove that one currently recommended method of obtaining the reliability of an instrument defined on a population of aggregate units is invalid. This method randomly splits the aggregate into two halves, correlates the two half unit scores by a Pearson product moment correlation coefficient, and corrects the…

A rederivation of Lord's formula for estimating variance in multiple matrix sampling is presented as well as the ways Cronbach's coefficient alpha and the Spearman-Brown prophecy formula are related in this context. (Author/JKS)

Details of a reasonably precise normalization technique for coefficient alpha are outlined, along with methods for estimating the variance of the normalized statistic. These procedures lead to the K-sample significance test. (RC)

This research examined 2 elements of multiple-choice test construction, balancing the key and optimal number of options. In Experiment 1 the 3 conditions included a balanced key, overrepresentation of a and b responses, and overrepresentation of c and d responses. The results showed that error-patterns were independent of the key, reflecting…

The adequacy of equipercentile, linear, one-parameter (Rasch), and three-parameter logistic item-response theory procedures for equating 12 forms of five tests of general educational development were compared. Results indicated the equating method adequacy depends on a variety of factors such as test characteristics, equating design, and sample…

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…

Generalizability theory was used to analyze the dependability of elementary school student ratings of attitudes toward school subjects. The rating scales under investigation have been developed to measure the attitudes of students toward four school subjects at both the primary and intermediate levels. Two generalizability coefficients, differing…

This study attempted to determine the effectiveness of cloze procedures as norm-referenced instruments by comparing the differential responses of four groups of college students of English as a second language on two identical cloze passages. The responses were scored using both exact-answer and acceptable-word methods. The results indicate that…

This manual outlines the rationale for using the Criterion Referenced Test (CRT) approach and suggests specific guidelines for test developers to use in constructing test items. Methods for assessing the adequacy of a CRT are also provided. (Author/RC)

The statistical properties of two methods of estimating gain scores for groups in quasi-experiments are compared: (1) gains in scores standardized separately for each group; and (2) analysis of covariance with estimated true pretest scores. The fan spread hypothesis is assumed for groups but not necessarily assumed for members of the groups.…

The existence of criterion-referenced (CR) measurement is questioned in this paper. Despite beliefs that differences exist between two alternative forms of measurement, CR and Norm Referenced (NR), an analysis of philosophical and psychological descriptions of measurement, as well as a growing number of empirical studies, reveal that the common…

This interim report of a pilot year impact study on evaluation methodology is part of a series of documents on the evaluation of Project Developmental Continuity, a Head Start demonstration program aimed at promoting educational and developmental continuity between children's Head Start and primary school experiences. This report deals with…