Buchanan, William J (2004) Correlation between academic and skills-based tests in computer networks. British Journal of Education Technology (BJET), 37 (1). pp. 69-78. ISSN 0007-1013

PDF Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial. Download (223kB) | Request a copy

Abstract/Description

Computing-related programmes and modules have many problems, especially related to large class sizes, large-scale plagiarism, module franchising, and an increased requirement from students for increased amounts of hands-on, practical work. This paper presents a practical computer networks module which uses a mixture of online examinations and a practical skills-based test to assess student performance. For widespread adoption of practical assessments, there must be a level of checking that the practical assessments are set at a level that examinations are set at. This paper shows that it is possible to set practical tests so that there can be a strong correlation between practical skills-based tests and examination-type assessments, but only if the practical assessment are set at a challenging level. This tends to go against the proposition that students who are good academically are not so good in a practice test, and vice versa. The paper shows results which bands students in A, B, C, and FAIL groups based on two online, multiple-choice tests, and then analyses the average time these students took to complete a practical online test. It shows that there is an increasing average time to complete the test for weaker students. Along with this, the paper shows that female students in the practical test outperform male students by a factor of 25%.

Actions (login required)