How to Assess L2 Proficiency? An Overview of Proficiency Assessment Research

Pascale Leclercq and Amanda Edmonds

This volume addresses an area of great interest to both language teachers and researchers interested in pedagogy and second language acquisition (SLA), namely how to assess proficiency in a second language (L2). For teachers, it is important to be able to assess accurately and reliably the L2 proficiency of their learners, whether for formative evaluation, summative evaluation or within the context of language certification. For researchers, valid proficiency assessment measures are crucial, as without them meaningful interpretation of research results remains elusive (Norris & Ortega, 2003: 717; Pallotti, 2009). In her thorough review of the assessment practices in more than 150 articles published in four major SLA journals, Thomas forcefully argues this point, concluding that 'there is evidence that L2 proficiency is sometimes inadequately assessed in current second language acquisition research. In certain cases, this compromises empirical claims; in many cases, it limits the generalizability of research results' (Thomas, 1994: 330). According to Thomas, and many others, it is important that proficiency assessment measures be valid (i.e. testing what they purport to test), reliable (i.e. providing trustworthy information) and practical (i.e. fitting the needs of their users in terms of ease of implementation). However, two decades after Thomas's call for a better control of the proficiency variable in the field of SLA, authors continue to highlight the difficulties associated with proficiency assessment. Hulstijn et al., for example, point to the lack of reliable level assignment as a general problem in SLA research:

SLA [...] has frequently simply taken groups of learners at supposedly different levels of ability, conducted cross-sectional research and claimed that the results show development. Yet the levels have been woefully undefined, often crudely labelled 'intermediate' or 'advanced', or 'first and second year university students' – which means little if anything in developmental terms – and which cannot therefore be interpreted in any meaningful way. (Hulstijn et al., 2010: 16)

Carlsen (2012: 2) deplores the fact that proficiency level remains a 'fuzzy variable' in digital learner corpora: although she admits that most learner corpora developers report on the different levels of proficiency included in their material, she claims, like Hulstijn et al., that 'the levels of proficiency are not always carefully defined, and the claims about proficiency levels are seldom supported by empirical evidence.' (For discussions of proficiency levels, see the chapters in this volume by Tracy-Ventura, McManus, Norris & Ortega, and Callies, Díez-Bedmar & Zaytseva.) Carlsen moreover argues that the reliability of corpus-based research is jeopardized by the fuzziness of the proficiency variable. Thus, despite the clear need for valid, reliable and practical methods of assessing L2 proficiency, it seems that research agendas do not always grant sufficient attention to this issue.

Our volume attempts to respond to this need for additional attention with a presentation of recent research concerning L2 proficiency assessment in L2 English and L2 French. Our aim is to provide a variety of perspectives on L2 proficiency assessment, reflecting the multiple approaches that contribute to the field of SLA research. Some of the chapters present measures of receptive proficiency (those by Prince, Zoghlami, Ankerstein and Bairstow et al., this volume), while others focus on the assessment of productive proficiency (those by Hilton, Osborne, Callies et al., Tracy Ventura et al. & McManus et al., this volume). Although most of the studies presented in this volume deal with L2 learners, Gertken et al. and Bairstow et al. take a slightly different perspective and present assessment tools for bilingual populations. This volume also includes testing instruments designed to be used in an educative setting (Prince, Zoghlami, Osborne, Gertken et al. & Callies et al., this volume), and a range of assessment instruments more specifically designed for SLA researchers: coefficient of variation (Ankerstein, this volume), translation recognition task (Bairstow et al., this volume), elicited imitation (Tracy-Ventura et al., this volume), measurement of oral fluency (Hilton, this volume), and cluster analysis (Callies et al., this volume).

This first chapter is intended to serve as a general introduction to proficiency assessment research within the larger field of SLA and to situate the different contributions to the volume. Our presentation is necessarily brief, and we refer interested readers to Norris and Ortega (2003), Bachman and Palmer (1996), McNamara (2000), Housen and Kuiken (2009) and Hulstijn (2010a, 2010b, 2011) for more complete treatments of the topic. Our own presentation will begin with a contextualization of the issue of L2 proficiency, followed by a discussion of the key concepts used in the fields of SLA and L2 proficiency research. We will then provide a preview of the different contributions to the present volume.

L2 Proficiency: A Moving Target

One of the challenges facing teachers, researchers, language testers and students interested in L2 proficiency is the defining of what it actually means to be...