The University of California (Irvine) has automated its scoring of placement tests to incoming freshmen and women with specialized administrative software that interfaces a microcomputer with an optical card reader. The university's placement testing program, the computerized system, and advantages of the system are explained. (JN)

Maximum likelihood and Bayesian estimates of proficiency, typically used in adaptive testing, use item weights that depend on test taker proficiency to estimate test taker proficiency. In this study, several methods were explored through computer simulation using fixed item weights, which depend mainly on the items difficulty. The simpler scores…

An approach to scoring reading comprehension based on the concept of the testlet is described, using models developed for items in multiple categories. The model is illustrated using data from 3,866 examinees. Application of testlet scoring to multiple category models developed for individual items is discussed. (SLD)

Within large scale educational assessment agencies in the UK, there has been a shift towards assessors marking digitally scanned copies rather than the original paper scripts that were traditionally used. This project uses extended essay examination scripts to consider whether the mode in which an essay is read potentially influences the…

A recent innovation in the area of educational measurement is MDT multi-digit testing, a machine-scored near-equivalent to "fill-in-the-blank" testing. The MDT method is based on long lists (or "Answer Banks") that contain up to 1,000 discrete answers, each with a three-digit label. Students taking an MDT multi-digit test mark…

The capabilities and hardware requirements of four microcomputer software packages produced by the Office of Educational Testing, Research and Service at the University of Illinois at Urbana-Champaign are described. These programs are: (1) the Scan-Tron Forms Analysis Package Version 2.0, an interface between an IBM-compatible and a Scan-Tron…

The third in a series on Navy Computer Managed Instruction (CMI), this report describes how the automated scoring of teletypewriting tests affects training in a system for automated performance testing (APT) which was implemented in the teletypewriter (TTY) portion of the Radioman "A" School in San Diego. The system includes a computer-generated…

Identifies a number of practical decisions that must be made when designing and developing criterion referenced tests as part of a larger system of computer assisted instruction. The topics discussed include test construction, test security, item presentation, and response capturing and scoring. (19 references) (CLB)

Test Pac, a test scoring and analysis computer program for moderate-sized sample designs using dichotomous response items, performs comprehensive item analyses and multiple reliability estimates. It also performs single-facet generalizability analysis of variance, single-parameter item response theory analyses, test score reporting, and computer…

This report documents progress at the Center for Research on Evaluation, Standards, and Student Testing (CRESST) on the feasibility of scoring concept maps using technology. CRESST, in its integrated simulation approach to assessment, has assembled a suite of performance assessment tasks (the integrated simulation) onto which they have mapped the…

Discusses multiple choice questions and presents a statistical approach to post-test correction for guessing that can be used in spreadsheets to automate the correction and generate a grade. Topics include the relationship between the learning objectives and multiple-choice assessments; and guessing correction by negative marking. (LRW)

Performed generalizability analyses of expert ratings and computer-produced scores for a computer-delivered performance assessment of physicians' patient management skills. The two automated scoring systems produced scores for the 200 medical students that were approximately as generalizable as those produced by the four expert raters. (SLD)

This book contains a series of research articles about computerized adaptive testing (CAT) written for advanced psychometricians. The book is divided into sections on: (1) item selection and examinee scoring in CAT; (2) examples of CAT applications; (3) item banks; (4) determining model fit; and (5) using testlets in CAT. (SLD)

Differential item functioning (DIF) assessment attempts to identify items or item types for which subpopulations of examinees exhibit performance differentials that are not consistent with the performance differentials typically seen for those subpopulations on collections of items that purport to measure a common construct. DIF assessment…

A two-stage adaptive test and a conventional peaked test were constructed and administered on a time-shared computer system to students in undergraduate psychology courses. (The two-stage adaptive test consisted of a routing test followed by one of a series of measurement tests.) Comparison of the score distributions showed that the two-stage test…