Abstract
The present study discusses how a group of experienced raters use different types of scales to assess the development of oral proficiency in English as a second language (L2). Raters assigned rates to speech samples first using a holistic scale (CEFR, 2018) and then assigning rates for pronunciation, vocabulary, grammar and fluency performance using individual scales. The speech samples were recorded by five Brazilians. There were two data collection sessions, with the second one occurring 7-8 months after the first one. The results indicate high levels of agreement among raters for all scales. Furthermore, the raters detected changes in speakers’ performance in four out of five scales: L2 oral proficiency, vocabulary, grammar, and fluency, and these differences in rates across time were significant for oral proficiency, vocabulary, and fluency. Thus, the different types of scale allow detecting L2 oral proficiency development.
Highlights
Whether one is engaged in second language research or second language teaching, a pressing concern is to understand the language proficiency construct
One important point raised by Hulstijn (2012) is that language proficiency can be assessed in terms of basic language cognition and/or high language cognition
The present study investigates the assessment of L2 oral proficiency development by a group of raters using holistic descriptors from the Common European Framework of Reference for Languages (CEFR), as well as analytic scales to rate the speakers’ performance in terms of pronunciation, vocabulary, grammar, and fluency accuracy
Summary
Whether one is engaged in second language research or second language teaching, a pressing concern is to understand the language proficiency construct. In his review of research articles in the field of bilingual language cognition, Hulstijn (2012) highlights the fact that language proficiency is a variable that has been investigated either as an independent or a moderator variable, and that a number of instruments are used to assess this construct: vocabulary tests, such as the Peabody Picture Vocabulary Test or Nation’s vocabulary tests; translation tests; mean length of run measures; panels of judges; fluency measures; self-rating questionnaires; or even years or levels of language-program enrolment. One important point raised by Hulstijn (2012) is that language proficiency can be assessed in terms of basic language cognition and/or high language cognition. The core components consist of linguistic cognition “in the phonetic-phonological, morpho-phonological, morphosyntactic, and lexical/pragmatic domains” (Hulstijn, 2015, p. 42)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
