Easy ways to raise dropout rate

九月 10, 1999

It is good to find a member of the Quality Assurance Agency board questioning the QAA approach to judging teaching quality, and Roger Williams is absolutely right in questioning the use of numerical scores ("Why the numbers don't add up", THES, August ).

Since the QAA judges quality against the objectives laid down by each institution, there is no way to compare the judgements made in different institutions. Point scores that encourage such comparisons are worse than pointless; they lead to bad practices, such as the Higher Education Funding Council for England basing its Fund for the Development of Teaching and Learning funding on teaching quality assessment results, and pernicious practices such as league tables. These arguments against point scores are far more important than the ones Williams presents.

Incidentally, his arguments apply equally to student degree assessments.

But Williams does not go far enough. If the QAA is really serious about wanting to foster self-assessment of quality, then its business is not to verify or possibly challenge the self-assessment, but to assure itself that the self-assessment has been carried out properly. That is not an assessment task but an audit task. The sooner the QAA realises that and goes over to audit practices, the better. There it can learn much from the earlier Academic Audit Unit and the Higher Education Quality Council, where Peter Williams, who is now a member of the QAA, pioneered such an approach. Where the earlier audits fell short was that they lacked teeth and at times were too polite, but the QAA now has more than enough teeth.

External assessment will lead to a deprofessionalisation of the academic profession and a dumbing down of what is offered to students; properly carried out external audit ought to have the opposite effect.

Lewis Elton

Professor of higher education

University College London

请先注册再继续

为何要注册?

  • 注册是免费的,而且十分便捷
  • 注册成功后,您每月可免费阅读3篇文章
  • 订阅我们的邮件
注册
Please 登录 or 注册 to read this article.