Video-conferencing speaking tests: an investigation of context validity related to test administration

Abstract

Face-to-face speaking assessment provides the benefit of eliciting a broad interactional construct, but at the cost of being logistically complex, resource-intensive and difficult to manage. Advances in video-conferencing (VC) technology now make it possible to engage in online interaction more successfully than previously, thus reducing dependence upon physical proximity between the examiner-interlocutor and the candidate(s). It is therefore not surprising that such technology is seen as a valuable assessment tool in geographically remote and politically unstable areas of the world, or indeed in contexts affected by the social distancing required during the recent Covid-19 pandemic. However, the administrative conditions under which the test takes place, one of the key contextual parameters of the VC-delivered test, , is often overlooked, despite its potentially significant influence on candidates’ performance and therefore overall test validity (Weir 2005). In this chapter, we report on investigations into