Teach and Learn (TAL) is a Computer-Aided Assessment (CAA) resource that has been developed in Bristol, but is available to any higher education (HE) institution in the UK. In addition to being web-based, the beauty of TAL is the use of an industry standard database to store the questions and associate each with a full set of metrics that together both describe and classify the question. TAL is used, principally by the departments of Maths/ Engineering and Chemistry, to deliver tests for both formative and summative assessment. The TAL system is extremely versatile; many of its operational features will be demonstrated in the presentation.

For TAL to be useful within biomedical sciences, however, there must be a means to categorise questions that reflects the breadth of subject material. Until recently, TAL currently lacked the means to accurately classify biomedical questions, which partly explained why TAL was not more extensively used in biomedical departments. To address this we have adopted the system of Medical Subject Headings (MeSH), a biomedical thesaurus, developed and maintained by the American Library of Medicine, which is used in academic databases such as PubMed and Web of Science.

The TAL project is used by several UK HE institutions and by several departments within Bristol. With the implementation of MeSH as a means to categorise questions, the way is open for TAL to be used more widely by biomedical departments in the faculties of Science and Medicine.

Several departments use the multiple choice question (MCQ) format in paper-based tests. Whilst simple in principal, there are significant logistical problems associated with maintaining a large bank of MCQ questions. These logistical problems become compounded if staff are required to track details such as:

ß Question author

ß Target group: Programme and Level

ß Date last used in examination

ß Average values for correct/incorrect responses for each

examination

These logistical problems are minimised using a database to associate the questions and associated fields. The database that provides this dimension to TAL is Oracle, an industry standard database, which is widely used within Information Services in Bristol.

In practice, TAL is used most often to provide self-assessment tests to students. Access to objective tests has been shown to improve deep learning and, if adequate feedback is provided, such tests exemplify a student-centred learning activity. Moreover, CAA has the potential to support the learning of large groups of students, and offers economies of scale; it requires no more effort to make web-based CAA available to one student compared with one thousand students.

It should be clear that properly supported self-direct learning and computer-aided self-assessment offers our best chance of providing a stimulating and challenging learning experience for students, particularly as staff/student ratios worsen. In addition, CAA can be used to test the attainment of benchmark levels of competence. For example, additional help in mathematics could be targeted specifically at those students who score poorly in a CAA test designed to test student’s ability to solve basic mathematical problems. This ‘benchmarking’ approach can help target resources to support the learning of an increasingly diverse student population.

We thank the University of Bristol for a Learning and Teaching Grant.atics could be targeted specifically at those students who score poorly in a CAA test designed to test student’s ability to solve basic mathematical problems. This ‘benchmarking’ approach can help target resources to support the learning of an increasingly diverse student population.

We thank the University of Bristol for a Learning and Teaching Grant.