The overall purpose of this research was to compare a maximum likelihood based tailored testing procedure to a Bayesian tailored testing procedure. The results indicated that both tailored testing procedures produced equally reliable ability estimates. Also an analysis of test length indicated that reasonable ability estimates could be obtained…

The development of the Computer Assisted Testing (CAT) System at the University of Delaware is described. The introduction presents the background leading up to interactive terminal testing in 1973. Documentation for the system includes CAT System programs, format of questions for constructing test item pools, format for entering class lists,…

Tailored testing increases the efficiency of tests by individually selecting for each person a set of items from an item pool so that the difficulty of the items selected will be such as to maximize the information provided by the score. The tailored testing procedure designed by Cliff orders persons and items on a common ordinal scale and…

Describes a study which investigated the feasibility of interactive computer assisted testing (ICAT) as an effective instructional method and its effects on attitudes towards computer assisted instruction (CAI). (VT)

Describes measurement problems associated with computer based testing (CBT) programs when they are part of a computer assisted instruction curriculum. Topics discussed include CBT standards; selection of item types; the contamination of items that arise from test design strategies; and the non-equivalence of comparison groups in item analyses. (8…

School psychologists and education students and faculty were asked to distinguish between computer-generated and individually written psychological test comments and indicate cues used in making identifications. Even though results indicate correct identification did not differ significantly from chance, school assessment reports must provide more…

The computer assisted testing (CAT) program at Florida State University's Assessment Resource Center is described. Three measurement issues (test quality, confidence in mastery decisions, and maintenance of test validity) and two instructional design issues (quality of instructional objectives and extended feedback following testing) important for…

FITNESSGRAM, a computerized program for scoring and reporting students' physical fitness status based on specific tests, has been a resounding success in the Tulsa, Oklahoma, public schools. The program provides feedback to students and permits evaluation of physical fitness efforts on a school-by-school basis. (PP)

To reduce the cost of item writing and to enhance the flexibility of item presentation, items can be generated by item-cloning techniques. An important consequence of cloning is that it may cause variability on the item parameters. Therefore, a multilevel item response model is presented in which it is assumed that the item parameters of a…

Score comparability is important to those who take tests and those who use them. One important concept related to test score comparability is that of "equity," which is defined as existing when examinees are indifferent as to which of two alternate forms of a test they would prefer to take. By their nature, computerized adaptive tests…

In recent years many researchers, clinicians, and testing companies have shifted from paper-and-pencil to computer administrated tests of psychological measures. This paper explores the feasibility of using the Internet to collect data from the Inventory of Parent and Peer Attachment (IPPA), the Negative Mood Rating Scale (NMRS), and the Trait…

The example of a medical licensure test is used to demonstrate situations in which complex, integrated content must be balanced at the total test level for validity reasons, but items assigned to reportable subscore categories may be used under a multidimensional item response theory adaptive paradigm to improve subscore reliability. (SLD)

Provides a personal view of the history of Anglo-American language testing over the last half-century. Argues that major developments in the field have tended to be embraced too enthusiastically so that they have led to unbalanced views concerning the construct definition of language, the scope of test impact, and the value of new methods of test…

A polychotomous nominal response model-based computerized adaptive test (CAT) was simulated using data from 1,093 University of Texas students. The ability estimation of this model and its overall performance were compared with those of a dichotomous three-parameter logistic model-based CAT. Advantages and drawbacks of nominal response CAT are…

Using the constant information model, constant amounts of test information, and a finite interval of ability, simulated data were produced for 8 ability levels and 20 numbers of test items. Analyses suggest that it is desirable to consider modifying test information functions when they measure accuracy in ability estimation. (SLD)