Assessing interactional competence:
features, scoring, and practicality
The University of Melbourne
Work on L2 Interactional Competence (IC) has grown rapidly in recent years (Lee & Hellermann, 2014; Pekarek Doehler, 2017; Pekarek Doehler & Pochon-Berger, 2015), and there is now some research on assessment of IC (Galaczi, 2014; Grabowski, 2013; Youn, 2013, 2015). However, this research area is still small and beset by thorny problems, such as: what markers of IC distinguish ability levels and can thus be assessed? How can test taker performances in a joint achievement (such as spoken interaction) be assessed individually? How can assessment of IC be done in a way that is practical and not overly resource intensive?
This talk will discuss these problems and sketch out research approaches for tackling them. It will first describe some of the candidate features for IC assessment as well as methods for assessing them that allow valid (in Kane’s, 2006, sense) conclusions to be drawn as to test takers’ IC. It will also discuss the problem of scoring an individual performance within the context of a joint achievement, especially with regard to examiner effects (Brown, 2003; Ross, 1992; Young & He, 1998). Furthermore, the problem of practicality of assessing IC will be considered, which presents one of the central real-world barriers for a broader integration of IC in language tests. Lastly, future research directions for the assessment of IC will be outlined.