Session 1C

While phrasal verbs (e.g., lay off and make up) are prevalent in English and used ubiquitously by native speakers, second language (L2) learners of English have great difficulty acquiring them in a targetlike manner (Garnier & Schmitt, 2016). In this line of research, the current study aims to examine the extent to which L2 learners benefit from form-focused instruction that draws their attention to phrasal verbs through various instructional activities proposed by Ranta and Lyster (2018). Given that L2 learners' autonomy is known to significantly affect L2 learning (Zhou, 2016), the present study also investigates the degree to which L2 learners' autonomy in selecting their own instructional materials affects their acquisition of phrasal verbs and their overall learning experience. 
To answer these research questions, a mixed methods study was conducted in collaboration with a major Canadian airline. The participants were 36 employees of the airline who were L2 learners of English. They were each assigned to either a treatment or control condition (i.e., 18 participants per condition). After completing a pretest that measured their linguistic knowledge of phrasal verbs, the participants in the treatment condition individually partook in six 30-minute online instructional sessions delivered via COLLIE, a computer-assisted language learning platform that employs artificial intelligence technologies. During each session, the participants were asked to select one of three instructional material options (i.e., a work-related pseudo-email; a press release article from the airline; a news article from major news outlets). They then received form-focused instruction built around the materials they selected, including noticing, awareness, and practice activities that drew their attention to a total of 24 target English phrasal verbs over the six sessions. After the last session, the participants completed an immediate posttest and a semi-structured interview, the latter of which documented their perceptions of the instructional sessions and their overall learning experience. The interview also included a stimulated recall session focused on the selection of materials, during which participants reflected on their own decision-making processes and their impact on L2 learning. A delayed posttest was administered six weeks later. The participants in the control condition also completed all the tests but did not receive any instruction on phrasal verbs.
Results reveal the significant effects of form-focused instruction on the acquisition of phrasal verbs and the role of L2 learners' autonomy in material selection. The presentation concludes by highlighting the current study as a useful model for researcher-practitioner collaboration from which researchers, practitioners, and industry can benefit.
References:
Garnier, M., & Schmitt, N. (2016). Picking up polysemous phrasal verbs: How many do learners know and what facilitates this knowledge? System, 59, 29–44.
Ranta, L., & Lyster, R. (2018). Form-focused instruction. In P. Garrett & J. Cots (Eds.), The Routledge handbook of language awareness (pp. 40–56). Routledge.
Zhou, M. (2016). The roles of social anxiety, autonomy, and learning orientation in second language learning: A structural equation modeling analysis. System, 63, 89–100.

The rise of corpus linguistics as a field has led to increasingly interdisciplinary studies with learner corpora being utilized in second language acquisition (SLA) research (McEnery et al., 2019). Learner corpora allow SLA researchers to examine learner language for focal linguistic structures that would have been more infrequently found in the traditionally smaller sample of language from SLA studies (Gass et al. 2020). Using the Gachon Korean EFL Corpus, this study examines the relationship between learners' collocational competence and their proficiency level. The importance of collocational competence has been emphasized to a great degree in SLA (Altenberg & Granger, 2001; Ellis, 2002; Hill, 2000; Howarth, 1998; Laufer & Waldman, 2011; Lewis, 2000; McEnery et al., 2019). Collocations that are frequently used among native speakers can help learners, as ready-made chunks of language, produce more native-like language with greater ease (Hill, 2000; McEnery et al., 2019). Although a massive amount of data accumulated in corpora has propelled the corpus-based research on learners' formulaic knowledge, few studies have incorporated corpus linguistics as a tool to analyze collocational competence (McEnery et al., 2019). 
With this gap in mind, this study analyzed the appropriateness of Korean learners' construction of English formulaic units, specifically light verb constructions (LVCs). LVCs, generally formed with a commonly-used verb and a noun phrase in its direct object position (e.g., make a deal) (Nagy et al., 2020), can trigger L2 learners to produce awkward expressions (Altenberg & Granger, 2001; Neff-van Aertselaer & Bunce, 2008; Shirato & Stapleton, 2007; Wang, 2016) as the verb has lost some of its original meaning when used in LVCs . In particular, the study zeroed in on LVCs with the English verb make, given that other English language verbs, such as do and have, are also used as auxiliary verbs.
The Gachon Korean EFL Learner Corpus on which the analysis was based compiles the written data of learners from a wide range of proficiency levels as measured by their reported TOEIC scores. To investigate the relationship between learner proficiency level and collocational competence, the corpora was divided into three learner sub-corpora based on learners' TOEIC score ranges. All LVCs containing make were identified and coded for appropriateness using the key word in context (KWIC). A chi-square test comparing the two variables showed that there was a positive relationship between learners' collocational competence and their proficiency level. 
Our findings support the notion that L2 learners have more collocational competence in an L2 at higher proficiency levels. This study provides a better understanding of the relationship between collocational competence and L2 learner language proficiency. Findings will be discussed with regard to the utility of learner corpora for SLA research as well as pedagogical implications on how to learn and teach formulaic units, such as LVCs, taking learners' proficiency levels into consideration.

Lemmas and word families are different word counting units used in corpus analysis, word list compilation, dictionaries, testing, pedagogy, and research (Brezina & Gablasova, 2013; Coxhead, 2000; Dang & Webb, 2016; Nation, 2006; Nation & Beglar, 2007). Preference for word families in L2 vocabulary tests and lexical profiling of texts rests on the assumption that learners' knowledge of base words extends to understanding derived forms. Hence, they can understand related derived words with relative ease (Bauer & Nation, 1993). Preference for lemmas, on the other hand, is based on studies that showed that learners had limited morphological knowledge (McLean, 2018). However, in these studies, L2 derived words were tested in isolation, or in semantically opaque sentences, and most learners were not advanced.
Our study investigates the effects of contextual clues and learner proficiency on comprehension of derived words. Using a within-subject design, we compared comprehension in three conditions: 1. absence of clues, 2. availability of syntactic clues, 3. availability of semantic clues.
Ninety EFL learners at three proficiency levels participated in three successive tests in which they had to supply the meaning of 22 derived pseudowords constructed with non-word stems and 22 frequent affixes, e.g., stacement, gummful. Meanings of the non-word stems were provided. In each test, learners were given identical questions, e.g., "If stace means 'to participate,' what does stacement mean?" In Test 1, absence of clues condition, the derived items occurred in isolation; in Test 2, syntactic clues condition, the sentence structure was a clue to the part of speech, e.g., "I am asking for your stacement"; in Test 3, semantic clues condition, they occurred in sentences with semantic clues, e.g., "Full and active stacement in school activities is required of all students."
Test scores were compared by 3 (proficiency level) x 3 (clue condition) ANOVA with Repeated Measures. Results showed effects of both variables. The largest increase in comprehension scores occurred with the addition of syntactic clues, especially in the lowest proficiency group. We relate the results to morphological knowledge as such, to comprehension of derived words in texts, and to choices of word counting units in tests and text profiles.


References
Bauer, L., & Nation, I.S.P. (1993). Word families. International Journal of Lexicography, 6(4), 253-279
Brezina, V., & Gablasova, D. (2015). Is there a general core vocabulary? Introducing the New General Service List. Applied Linguistics, 36(1), 1-22. 
Coxhead, A. (2000). A new academic word list. TESOL Quarterly, 34(2), 213-238. 
Dang, T.N.Y., & Webb, S. (2016). Evaluating lists of high frequency vocabulary. In I.S.P. Nation (Ed.), Making and using word lists for language learning and teaching (pp. 153-167). John Benjamins. 
McLean, S. (2018). Evidence for the adoption of the flemma as an appropriate word counting unit. Applied Linguistics, 39, 823–845.
Nation, I.S.P. (2006). How large a vocabulary is needed for reading and listening? TheCanadian Modern Language Review, 63(1), 59-81. 
Nation, I.S.P., & Beglar, D. (2007). A vocabulary size test. The Language Teacher, 31(7), 9-13.