By Sidney H. Irvine, Peter L. Dann (auth.), Peter L. Dann, Sidney H. Irvine, Janet M. Collis (eds.)
Read Online or Download Advances in Computer-Based Human Assessment PDF
Best assessment books
Enormous quantities of useful rules simply built-in Into Your classes and ActivitiesAre your scholars note shrewdpermanent? Math clever? humans shrewdpermanent? faucet into your scholars' various intelligences and increase your lesson plans with this useful source. Your scholars will love studying in ways in which play to their strengths.
This ebook brings jointly a set of present examine at the review of oral talent in a moment language. Fourteen chapters specialise in using the language skillability interview or LPI to evaluate oral skillability. the quantity addresses the primary factor of validity in skillability overview: the ways that the language skillability interview is entire via discourse.
- A Handbook of Children and Young People's Participation : Perspectives from Theory and Practice
- Key concepts in educational assessment
- The Action Research Planner: Doing Critical Participatory Action Research
- Developing Teacher Assessment
Additional resources for Advances in Computer-Based Human Assessment
Air Force Office of Scientific Research funded a programme of research initiated and directed by Dr Raymond Christal. The project, known as the Learning Abilities Measurement Project (LAMP), is located at the Air Force Human Resources Laboratory at Brooks Air Force Base, San Antonio, Texas (Note 2). 51 P. L. Dann et al. ): Advances in Computer-Based Human Assessment, 51-101. © 1991 Kluwer Academic Publishers. 52 Computer-based Cognitive Tasks One of the goals of the LAMP research has been to determine whether it is possible to identify ECTs with reliable individual difference scores, and then to determine whether or not such scores are related to the abilities traditionally measured by aptitude batteries.
One is whether the between-form reliability (the correlation between the results of standard and computer-based administrations) is adequate, and ideally at least as high as the test-retest reliability reported in the original test manual. The other is whether there is any difference between the mean score obtained under the two forms of administration ideally there should be no difference. If both these conditions are met, there is some justification for employing the automated procedure in clinical practice as if it were the original standard form, and for making use of existing normative data for the test.
The loss of reliability on this scale was found to be due entirely to the more variable results of patients detained in a Special Hospital. The patients at this site are, of course, expected to produce abnormal results on the P scale of the EPQ. Reliability on the Road-Map Test was acceptable, although that on the two subtests of the Differential Aptitude Test was disappointing. No sensible reliabilities can be calculated for the Wisconsin Card Sorting Test, and other research has shown that the Digit Span task is fundamentally changed by conversion to an automated format (Beaumont, 1985).