A unique style of computer‐assisted assessmentReport as inadecuate






Author: Mike Thelwall

Source: https://core.ac.uk/

This paper examines a project at the University of Wolverhampton that has been producing its own unique style of computerized test for several years. The tests are all designed to deliver a different set of questions each time they are run, a fact which enables many of them to double as learning resources. Most of the tests are used for both formative and summative assessments on Level 1 modules, in conjunction with more traditional assessment methods


Teaser



A unique style of computer-assisted assessment Mike Thelwall School of Computing and Information Technology, University of Wolverhampton.
Email: cm 1993@wlv.ac.uk This paper examines a project at the University of Wolverhampton that has been producing its own unique style of computerized test for several years.
The tests are all designed to deliver a different set of questions each time they are run, a fact which enables many of them to double as learning resources.
Most of the tests are used for both formative and summative assessments on Level 1 modules, in conjunction with more traditional assessment methods. Introduction The Computer-Based Assessment project at the University of Wolverhampton has produced a number of tests built around a common framework.
There are three main tests that are used by about 800 students per year, and they replace written tests.
There are also two diagnostic tests that are not part of any formal assessment but are used by about 400 students per year.
We have produced numerous special versions of these tests for short courses, different teaching techniques, and one for a Broadnet Online training module (Broadnet, n.d.).
The tests are all written with 80,000 permutations built in to allow them to be made available for students to use and practise on at all times without compromising security.
In fact, there are no serious security concerns because of the random factors, and so the same tests are used every year without the need to be hidden or rewritten.
Similar advantages have been found on other projects using random factors; see for example Thoennessen and Harrison (1996). The project has been producing random-based computerized tests for modules for several years now.
It began with a single statistics test, the success of which led to the use of the technology for the production of others.
Our criteria for accepting a module for a computerized test are that it must have a large number of students and that the tests should be able to be used for a number of years.
We have produced five random-based PC tests so far; two of which have a number of versions for different modules and situations.
Three of these replace written tests in maths, stats and IT, while the other two are diagnostic tests for the numeracy and computing skills of new students. 49 Mike Thehvall A unique style of computer-assisted assessment The main features common to all the tests are as follows. • Detailed feedback is given at the end of each t....






Related documents