The Mantel-Haenszel approach for investigating differential item functioning (DIF) was applied to United States history items within the 1986 National Assessment of Educational Progress administered to 7,812 11th graders. DIF analyses were based on responses of 7,743 11th graders. Results concerning sex and racial differences and ethnicity are…

The available evidence indicates that home school youth of compulsory education age have been scoring equal or better than their conventional school peers on measures of cognitive achievement and desirable affective traits; the significance of this is discussed. (BJV)

A study evaluated the feasibility of using students' self-reports of academic development and college experiences as substitutes for standardized test scores (proxies) or policy indicators. Results suggested self-reports should be used with care as proxies for a national test but may be used as policy indicators for improvement of undergraduate…

The average overall Scholastic Assessment Test (SAT) score for high school seniors in 1994 remained at 902 of a possible 1600. Math scores showed a one-point gain whereas the average verbal scire fell one point. Despite recent improvements in minority group scores, these dropped or remained stable in 1994. Math and verbal test scores for specific…

Maintains that, in the wake of unfavorable court decisions, many school boards in the South attempted to legitimize wage discrimination against black teachers through use of the National Teacher Examination (NTE). Examines questions of racial bias within the testing procedure. Profiles the NTE's leading proponent, Ben Wood. (MJP)

In 43 rural Maine communities that pay tuition to send their students to high schools in larger towns, achievement test scores appeared to influence parental school choice but explained only 20% of the variability in enrollment trends. Enrollment trends were not related to school socioeconomic status or expenditure per pupil. Contains 40…

Discusses specific instruments and procedures that university-level foreign languages and literatures departments can use in outcomes-assessment of their undergraduate majors, including standardized tests, oral assessments, questionnaires, and portfolios. Case studies of outcomes-assessment programs in language departments at the University of…

The first "recentering" of Scholastic Assessment Test (SAT) scores in 50 years is described. As of April 1995, test results are reported in terms of a new reference group, high school students graduating in 1990. The history of SAT scoring, suggested reasons for test score declines, and implications of the change are discussed. (MSE)

A study investigated the usefulness of a new ambulatory medicine clerkship design by randomly assigning students to it or to an existing traditional general medicine rotation, then comparing student performance on the licensing examination, a laboratory interpretation test, written work, and a problem-solving exam. Performances were the same…

In Texas, which has minimal child care standards, extensive alternative care arrangements were associated with pervasive negative effects in third graders. Children with the alternative care background received more negative ratings from parents and teachers, poorer grades, lower standardized test scores, and more negative sociometric nominations…

Effectiveness of the Mantel-Haenszel statistic in detecting differentially functioning test items was studied when the internal criterion was varied, using a dataset from a life skills examination taken by 1,000 Anglo-American and 1,000 Native American high school students. Choice of criterion variable has a substantial effect on detection. (SLD)

Median polishing, an exploratory data statistical analysis technique, was used to study achievement patterns for men and women on the Pharmacy College Admission Test over a six-year period. In general, a declining trend in scores was found, and males performed better than females, with the largest differences found in chemistry and biology.…

Examines the differences between norm-referenced tests (standardized assessments of intelligence, aptitude, achievement, and personality) and criterion-referenced tests. Until school districts improve their potential to develop meaningful criterion-referenced tests, norm-referenced tests will be the major yardstick for measuring student…

There is presently much disagreement about how students should be evaluated effectively to determine their progress. Some common methods are reviewed, including standardized tests and performance-based evaluation such as portfolios. Continuous experimentation is necessary in the effort to develop better measurement approaches. (SLD)

Examines Scotland's increasingly standardized national curriculum and testing reforms. Discusses centrally determined innovations, relationships between curriculum and assessment, reforms' effects on classroom, and teachers' role. Compares elements of encroaching "market model" reforms in England, Wales, and United States, suggesting…