The Third International Mathematics and Science Study (TIMSS) is one of the most comprehensive international studies of schooling and students' achievement in science and mathematics. TIMSS was originally conducted in 1995. Four years later in 1999, the Third International Mathematics and Science Study-Repeat (TIMSS-R) was conducted. This document…

A review of four areas in which the computer has influenced the theory and practice of achievement testing: a) test scoring and item analysis, b) item sampling, c) item generation, and d) the sequencing of items resulting in various types of adaptive testing. Also, reference is made to the impact of computer assisted achievement testing upon the…

A study compared college students' performance on complex multiple-choice tests with scores on multiple true-false clusters. Researchers concluded that the multiple-choice tests did not accurately measure students' knowledge and that cueing and guessing led to grade inflation. (PP)

This study compares multitrait-multimethod analyses (MTMM) performed on items and factor scores derived from the items. Based upon this demonstration, researchers are encouraged to conduct a preliminary factor analysis before analyzing MTMM data whenever there is a doubt about the underlying trait structure. (Author/PN)

A model is presented using examinee judgements in detecting ambiguous/misinterpreted items on teacher-made criterion-referenced tests. A computational example and guidelines for constructing domain categories and interpreting the indices are presented. (Author/PN)

Describes a computer-assisted testing system which produces multiple-choice examinations for a college course in business administration. The system uses SPIRES (Stanford Public Information REtrieval System) to manage a database of questions and related data, mark-sense cards for machine grading tests, and ACL (6) (Audit Command Language) to…

Test materials in English and social studies, reflecting Black and White culture, were developed and administered to Black and White high school students. Analyses of variance indicated that performance by either group was not affected by Black or White culture content. White examinees scored higher on all tests. (SJL)

The construct validity of reading comprehension test items was studied in a two-stage process. Five characteristics of task difficulty were defined and a heterogeneous set of 52 items were rated for these characteristics. Then correlations were obtained between ratings and item difficulty data. (CTM)

The scope and nature of sex differences in the Graduate Record Examination are explored by identifying individual test items that differ from the other items in terms of the magnitude of the difference in item difficulty for the sexes. In general, limited evidence of differences was established. (Author/CTM)

A measure of the extent to which the Kuder Occupational Interest Survey (KOIS) was predictive of occupational membership for an individual was correlated with KOIS item and scale scores. Results indicated that the KOIS was a moderator of its own predictive validity. (Author/JKS)

A computer program for analyzing test results is presented. Item analysis and the point biserial correlation coefficient, both generated by this program, are explained and discussed. (MK)

Investigates comprehension of metaphoric compounds. The study asked children and adults to identify referents of various types of these compounds. Findings reveal that 5-year-olds outperformed 3-year-olds but performed significantly less well than adults and that 3-year-olds did not generally interpret the compounds literally. (37 references)…

A study evaluated the use of a "none-of-these" option on multiple-choice achievement tests in undergraduate dental education. Results indicated this option neither enhanced nor diminished examinee performance stability but did reduce the examinee's opportunity to select correct choices by means unrelated to course objectives, thereby enhancing…

The effects were assessed of a test coaching program for urban Black youth, who intended to take the Scholastic Aptitude Test (SAT), on their performance on quantitative items. Findings for 116 program participants are discussed in relation to the improvement of supplemental instructional programs. (TJH)

Two alternatives to traditional item analysis and reliability estimation procedures are considered for determining the difficulty, discrimination, and reliability of optional items on essay and other tests. A computer program to compute these measures is described, and illustrations are given. (SLD)