Introduction

Improving the vocabulary knowledge of second language learners is an important goal for teachers, in view of the close relationship between vocabulary knowledge and all language skills. One consequence is an increasingly wide range of interventions designed to consider how such knowledge can be developed alongside the skills themselves (Clenton & Booth, 2020Clenton, J., & Booth, P. (2020) Vocabulary and the four skills: Pedagogy, practice, and implications for teaching vocabulary. Routledge. ). Evidence suggests that pedagogical activities used alongside a focus on meaning to enhance the salience of items (Sharwood Smith, 1991Sharwood Smith, M. (1991) Speaking to many minds: On the relevance of different types of language information for the L2 learner. Second Language Research, 7 (2), 118–132. ) and hence noticing (Schmidt, 1990Schmidt, R. W. (1990) The role of consciousness in second language learning. Applied Linguistics, 11 (2), 129–158. ) can maximise vocabulary learning from written and aural input that is used for broader, communicative learning purposes. The focus of such research has however tended to be on quantitative outcome measures, which, while important, give few if any insights into why different interventions have the impact that they do. In this paper we argue that exploring the strategies learners report employing during the intervention can make an important contribution to gaining those insights. The latter might be especially important for understanding why the outcomes from vocabulary learning interventions vary according to learner individual differences. In other words, we need to understand who benefits from what kind of intervention and why.

This article draws on data collected as part of a wider study (Zhang, 2018Zhang, P. (2018) Comparing different types of EFL vocabulary instruction for Chinese senior secondary school learners of English. Unpublished PhD thesis, University of Reading.), that explored the extent to which learners of different L2 proficiency levels benefited from different types of attention-enhancing pedagogical activities for vocabulary learning. That study established that levels of listening proficiency and vocabulary knowledge influenced how much learners benefited from different kinds of vocabulary instruction (Zhang & Graham, 2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. , bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ). A fuller understanding of those findings, however, can come from an exploration of the strategies learners of varying proficiency levels reported employing during the pedagogical activities. A range of strategic behaviour may be elicited by different pedagogical tasks (Macaro, 2017Macaro, E. (2017) Students’ strategies in response to teachers’ second language explanations of lexical items, Language Learning Journal, 45 (3), 352–367. ) and may vary across proficiency levels (Graham et al., 2010Graham, S., Santos, D., & Vanderplank, R. (2010) Strategy clusters and sources of knowledge in French L2 listening comprehension. International Journal of Innovation in Language Learning and Teaching, 4 (1), 1–20. ), a phenomenon which analyses of test data are less able to illuminate. This study therefore extends Zhang and Graham (2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. , bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ) and explores the strategies used by L2 learners of different levels of prior vocabulary knowledge (PVK) and listening proficiency in response to three types of vocabulary explanations after listening to oral input: explanations in the L2, codeswitched explanations, and those that give contrastive focus-on-form information. In doing so it provides new knowledge by illuminating why and in what way the interventions implemented had the impacts they did on learners of different proficiency levels.

Vocabulary learning through listening and proficiency

In formal learning contexts such as schools, listening to the teacher is likely one of the most frequently accessed sources of aural input. Furthermore, such input may be ‘enhanced’ by the teacher in some way, in order to focus learners’ attention on certain aspects of language. For example, they may emphasise orally certain words, repeat them, or explain their meaning. Different forms of input enhancement may vary in the degrees of ‘involvement load’ (Hulstijn & Laufer, 2001Hulstijn, J. H., & Laufer, B. (2001) Some empirical evidence for the involvement load hypothesis in vocabulary acquisition. Language Learning, 51 (3), 539–558. ) they prompt, however, in terms of whether they stimulate ‘need’, ‘search’, and ‘evaluation’ and at what level. In other words, the extent to which learners have to understand a given item for task completion (need), seek out the item’s meaning (search), or consider whether a given meaning or use for a word is the most appropriate one for a certain context (evaluation) matters. The level of ‘evaluation’ is considered to be especially important (Hulstijn & Laufer, 2001Hulstijn, J. H., & Laufer, B. (2001) Some empirical evidence for the involvement load hypothesis in vocabulary acquisition. Language Learning, 51 (3), 539–558. ), as more evaluation usually brings with it deeper processing and therefore is more likely to lead to better learning outcomes. Interestingly, and of relevance for this study, ‘involvement load’ seems to have only been investigated quantitatively (see for example Yanagisawa & Webb, 2022Yanagisawa, A., & Webb, S. (2022) To what extent does the Involvement Load Hypothesis predict incidental L2 vocabulary learning? A meta-analysis. Language Learning, 71 , 487–536. ) without considering whether activities designed to have different levels of ‘load’ really do prompt learners to engage in different kinds of mental processing. Likewise, very few if any studies have explored ‘involvement load’ in relation to vocabulary learning through listening.

Several previous studies have reported that more vocabulary learning occurs through listening among higher proficiency learners (e.g. Suárez & Gesa, 2019Suárez, M. D. M., & Gesa, F. (2019) Learning vocabulary with the support of sustained exposure to captioned video: Do proficiency and aptitude make a difference? The Language Learning Journal, 47 (4), 497–517. ; Vidal, 2011Vidal, K. (2011) A comparison of the effects of reading and listening on incidental vocabulary acquisition. Language Learning, 61 (1), 219–258. ), potentially because such learners, and especially those with a larger vocabulary size, understand more of the input and hence are in a stronger position to learn from it. By contrast, no proficiency effect was found in a study that explored vocabulary learning from aural input with post-listening activities among adolescent learners (Hennebry et al., 2013Hennebry, M., Rogers, V., Macaro, E., & Murphy, V. (2013) Direct teaching of vocabulary after listening: Is it worth the effort and what method is best? Language Learning Journal, 45 (3), 282–300. ). The study was followed up by a qualitative investigation (Macaro, 2017Macaro, E. (2017) Students’ strategies in response to teachers’ second language explanations of lexical items, Language Learning Journal, 45 (3), 352–367. ) that used stimulated recall interviews to explore the strategies used by 24 learners to understand their teacher’s L2 vocabulary explanations. Learners reported employing a limited range of strategies, largely searching for L1 cognates, regardless of whether the target word had an English cognate or not. The study acknowledged, however, that although the 24 interviewees represented a wide range of proficiency levels, it did not explore whether learners of different proficiency behaved differently in strategy use and whether such differences resulted in different vocabulary learning progress.

It is relevant to explore the strategies learners use to understand the input in which items are embedded as well as those they use to remember items, as the selection, combination, and pattern of using different strategies may result in different vocabulary learning outcomes from spoken input. Furthermore, previous research, summarised in Graham et al. (2008)Graham, S., Santos, D., & Vanderplank, R. (2008) Listening comprehension and strategy use: A longitudinal exploration. System, 36 (1), 52–68. , suggests that strategy use varies across learners of different listening proficiency levels, with metacognitive strategy use increasing with higher listening proficiency. In a later study, Graham et al. (2010)Graham, S., Santos, D., & Vanderplank, R. (2010) Strategy clusters and sources of knowledge in French L2 listening comprehension. International Journal of Innovation in Language Learning and Teaching, 4 (1), 1–20. also illustrated the complexity of the relationship between linguistic knowledge and strategy use, with lower levels of linguistic knowledge restricting some learners in the study in how effectively they could employ strategies in combinations. Such an interaction between linguistic knowledge and strategy use may have implications for how much vocabulary is learnt through listening, particularly in relation to how learners process the vocabulary to which they are exposed. This is an issue which, to our knowledge, has not yet been investigated.

In summary, the above discussion highlights that proficiency level does often, but not always, influence how much vocabulary is learnt from aural input, but that insights into why that is the case are limited. Additionally, previous studies have taken a largely quantitative approach. As argued by Hennebry et al. (2013)Hennebry, M., Rogers, V., Macaro, E., & Murphy, V. (2013) Direct teaching of vocabulary after listening: Is it worth the effort and what method is best? Language Learning Journal, 45 (3), 282–300. , exploring qualitatively how learners try to understand the vocabulary explanations they receive may provide additional evidence regarding any differences in learning related to proficiency level and may shed light on whether different teaching approaches induce different kinds of processing of language by listeners. This study therefore seeks to address that question through a qualitative analysis of strategy use in the context of a vocabulary teaching intervention.

The present study

We explored the impact of both listening proficiency and prior vocabulary knowledge (PVK) on vocabulary learning within different forms of instruction (Zhang & Graham, 2020bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ). We also explored how listening proficiency developed during the different interventions (Zhang & Graham, 2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. ). Four classes (137 learners), all first-year high school EFL learners (aged 15–16) from one school in China, were randomly assigned to a No Explanation (NE) group, L2 group, Codeswitching (CS) group, or Contrastive Focus on Form (CFoF) group, and participated in a six-week classroom intervention in which 60 lexical items (18 nouns, 13 verbs, 12 adjectives and 17 collocations) were taught. The approaches used in the ‘explanations’ groups were selected based on those implemented in other similar studies of vocabulary instruction but not previously all compared: L2, codeswitched explanations (as in Tian & Macaro, 2012Tian, L., & Macaro, E. (2012) Comparing the effect of teacher codeswitching with English-only explanations on the vocabulary acquisition of Chinese university students: A Lexical Focus-on-Form study. Language Teaching Research, 16 , 367–91. ), and CFoF as developed by Laufer and Girsai (2008)Laufer, B., & Girsai, N. (2008) Form-focused instruction in second language vocabulary learning: A case for contrastive analysis and translation. Applied Linguistics, 29 , 694–716. . These authors contrasted CFOF with approaches like codeswitching that “simply state the meaning of L2 words “, targeting instead “learners’ understanding of the similarities and differences between their L1 and L2 in terms of individual words and the overall lexical system”, through the explicit highlighting of such differences (p. 696). All three types of explanations targeted meaning, form, and use (Nation, 2022Nation, I. S. P. (2022) Learning vocabulary in another language (3rd ed.). Cambridge University Press.) by not only giving learners an explanation of what items meant but also an illustration of their use and a written version of the item displayed on a PowerPoint slide. For the L2 group, explanations were completely in the language being learnt, namely English and an additional example sentence was given; in the CS group, teacher explanations used both Chinese (L1) and English, followed by an additional example sentence; in the CFoF group, after giving the meaning of the word in the L1, the teacher explained in the L2 the way in which the word functioned differently in the two languages. See Figure 1.

As reported fully in Zhang and Graham (2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. , bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ), participants first completed a listening pre-test and an aural General English Vocabulary Test (GEVT) which also included the items taught in the intervention by way of a pre-test. Each group then had six intervention sessions (one per week, 10 vocabulary items taught per session). In each session they listened to a passage, followed by explanations of the items by their teacher (unless they were in the NE group, where they just listened to the passage and answered comprehension questions). All learners were tested on the 10 items at the end of each session, in an aural test of meaning recall, and then on all items again in combined delayed posttest two weeks later. They also completed a listening post-test. Findings reported in Zhang and Graham (2020b)Zhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. indicated that listening proficiency (LP) was a more important factor influencing vocabulary learning through aural input than PVK was, with most notable gains for learners with high listening proficiency and low PVK. High LP / low PVK learners learnt more under the CS than under the L2 condition. The CFoF approach was the most helpful for learning regardless of learners’ PVK and listening proficiency, whereas the NE approach was the least helpful. Listening proficiency, however (Zhang & Graham, 2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. ) gave rather different results: only the NE group showed significant improvement, as well as making significantly greater pre- to post-test improvement than the CFoF and the L2 groups did. Additionally, the CFoF group, who had the greatest vocabulary gains, improved the least in listening. Finally, learners from all groups with the lowest vocabulary scores at the start made the most progress in listening, suggesting that vocabulary instruction helped their listening.

Both studies (Zhang & Graham, 2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. , bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ) in their conclusions point to the importance of activities that encouraged deeper processing of the language to which they were exposed, with reference to ‘Involvement Load’ (Hulstijn & Laufer, 2001Hulstijn, J. H., & Laufer, B. (2001) Some empirical evidence for the involvement load hypothesis in vocabulary acquisition. Language Learning, 51 (3), 539–558. ) as discussed above. Enhanced ‘Involvement’ may have come from listening input and activities that required some degree of strategic behaviour, as long as they were not too far beyond the learners’ level of vocabulary knowledge and listening proficiency. The current paper explores that question by investigating whether any differences in vocabulary learning gains were related to learners’ strategic behaviour in response to the different types of teacher’s explanations, and the extent to which strategy use also varied by proficiency level.

The following questions were therefore proposed:

1. When learners listen to different kinds of teacher explanations of target vocabulary items, does their strategy use vary according to the type of vocabulary explanation they have experienced?

2. Are any differences in strategy use related to learners’ listening proficiency and how much they benefit from the intervention they experienced?

Overall, we were interested in looking specifically at how learners used strategies and whether patterns of use emerged by intervention group, proficiency level, and how much was learnt in the intervention.

Method

Participants and procedures

Twelve learners were selected from the original 137 in Zhang and Graham (2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. , bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ), two high and two low proficiency level learners from each intervention group. Proficiency levels were based on scores from the General English Vocabulary Test (GEVT) and listening test that learners completed before the intervention (Table 1). A composite score from these two tests was created by converting scores into z scores and then combining them. Next, learners were rearranged in ascending order of their composite test scores and were divided into ‘Low’ and ‘High’ proficiency levels. In other words, a median split approach was used. However, mindful of the potential shortcomings of using a median split (Iacobucci et al., 2015Iacobucci, D., Posavac, S., Kardes, F. R., Schneider, M. J., & Popovich, D. L. (2015) Toward a more nuanced understanding of the statistical properties of a median split. Journal of Consumer Psychology, 25 , 652–665 ), extra attention was paid to learners at the borderline, making sure that those with very similar scores were in the same proficiency group. An independent sample t-test showed that the High-level learners significantly outperformed the Low-level learners in the composite test scores (p < .001). Once the two proficiency groups were established in this way, the four learners from each intervention group were selected on the basis that they had consented to take part in the interview as well as the tests and intervention, and that the combined scores for each of the four selected students were comparable across the intervention groups (see Table 1).

Table 1.Overview of interviewees’ test scores

Student code	Group	Proficiency level	Combined GEVT/listening	Vocabulary pre‑test	Vocabulary post‑test	Vocabulary delayed post‑test
L2H1	L2	High	100.00 (60 + 40)	11.00	33.00	21.00
L2H2			90.00 (55 + 35)	8.00	24.00	14.00
L2L1		Low	65.00 (45 + 20)	8.00	24.00	10.00
L2L2			47.00 (37 + 10)	4.00	7.00	6.00
CSH1	CS	High	101.00 (66 + 35)	21.00	56.00	29.00
CSH2			79.00 (59 + 20)	14.00	50.00	25.00
CSL1		Low	63.00 (48 + 15)	4.00	39.00	5.00
CSL2			43.00 (43 + 0)	15.00	36.00	7.00
CFoFH1	CFoF	High	110.00 (60 + 50)	7.00	59.00	32.00
CFoFH2			78.00 (53 + 25)	6.00	53.00	34.00
CFoFL1		Low	62.00 (47 + 15)	6.00	49.00	34.00
CFoFL2			53.00 (43 + 10)	6.00	52.00	17.00

Note. The students’ intervention group is shown first (L2, CS, CFoF) followed by proficiency level (L or H) and then student number (1 to 2).

Initial insights came from an observation of the vocabulary learning gains of this sub-sample of learners. Proficiency level did seem to make some difference to how much learners benefited from each intervention type, although the size of the sub-sample needs to be remembered. The greatest benefits for both short-term learning and long-term retention seemed to be for the two high proficiency level CFoF learners. Their gains contrast with those of the L2 group high proficiency learners. Furthermore, the low proficiency learners in that group made the least progress in vocabulary. Yet there was an indication of short-term learning for the low proficiency learners in the CS group, even though this largely disappeared by the time of the delayed post-test. Finally, also worth noting is that in the CS and CFoF groups, the two high proficiency learners differed in terms of listening scores.

Stimulated recall interviews

These learners completed individual 20-minute stimulated recall interviews immediately after the delayed post-test, in order to gain insights into the strategic behaviour they had employed in response to the vocabulary explanations experienced in intervention sessions shortly before.

Stimuli for the interview were based on six of the taught target lexical items (one from each intervention session) so that each type or word class was represented and reflected around 10% of the items taught: two nouns ( association , tourism ), two collocations ( figure out , at ease ), one verb ( threaten ), and one adjective ( convincing ). The length allocated to the SRI limited the number of stimuli presented. Learners first watched a video clip from a lesson showing the teacher explaining a target lexical item and were then asked about what they were thinking during the explanation and how they tried to understand it. There was also discussion around whether they tried to learn the target items and if so, how. These two sets of questions were posed in order to observe the distinction drawn by Nyikos and Fan (2007)Nyikos, M., & Fan, M. (2007) A review of vocabulary learning strategies: Focus on language proficiency and learner voice. In A. Cohen & E. Macaro (Eds.), Language learning strategies: 30 years of research and practice (pp. 251–274). Oxford University Press. between vocabulary comprehension strategies (VCS) and vocabulary learning strategies (VLS), a procedure also adopted by Macaro (2017)Macaro, E. (2017) Students’ strategies in response to teachers’ second language explanations of lexical items, Language Learning Journal, 45 (3), 352–367. .

Data analysis

The 12 stimulated recall interviews were transcribed and translated from Chinese into English by the second author. A coding scheme, combining both inductive and deductive approaches (Gu, 2014Gu, Y. (2014) To code or not to code: Dilemmas in analysing think-aloud protocols in learning strategies research. System, 43 , 74–81. ), was applied to identify learners’ strategy use, employing NVivo 11. The second author then drew up an initial code book comprising meaningful instances of strategy use (with examples and definitions) emerging from an overview of all transcripts, with reference to existing strategy taxonomies for vocabulary (Schmitt, 1997Schmitt, N. (1997) Vocabulary learning strategies. In N. Schmitt & M. McCarthy (Eds.), Vocabulary: Description, acquisition and pedagogy (pp. 199–227). Cambridge University Press.) and listening comprehension (Vandergrift & Goh, 2012Vandergrift, L., & Goh, C. C. M. (2012) Teaching and learning second language listening: Metacognition in action. Routledge. ). The initial code book was then sent to the first author for comments, after which it was further modified to generate a second code book. Two researchers then used this to independently code three randomly selected transcripts (one from each intervention condition), resulting in an inter-rater reliability rate of 83.33%. Differences in coding were resolved through discussion, which then led to further modification of the code book to generate a final code book (see Appendix). This was then applied to the coding of all 12 transcripts. The number of instances of different types of strategies reported by interviewees in each intervention group and by proficiency level were firstly obtained. In recognition, however that such an analysis ‘cannot capture how a given strategy is used or the particular combinations of strategies used to build meaning [nor] the effective use of a strategy, such as the accuracy of an inference, an appropriate connection to prior knowledge or the depth of summarization’ (Vandergrift, 2003Vandergrift, L. (2003) Orchestrating strategy use: Toward a model of the skilled second language listener. Language Learning, 53 (3), 463–496. , p. 477), the quantitative analysis was supplemented by a qualitative analysis looking specifically at how learners used strategies and whether patterns of use emerged by intervention group, proficiency level, and how much vocabulary was learnt by students.

Findings

Research Question 1: Strategy use in response to different kinds of vocabulary explanations

Our research question explored what strategies learners used in response to the explanations for the target items in each of the three experimental conditions. An initial, quantitative overview was obtained of the number of instances of different types of strategies reported by learners from the three groups. The type of strategy used differed between the three groups (Figure 2), suggesting that the interventions prompted different types of focal attention, either a more global focus on the listening input as a whole, or one largely focused more narrowly on the target vocabulary items. Learners from the L2 group reported using translation, selective attention to known words and linguistic inferencing more frequently than learners from the other two groups did, as well as more strategies in total. They were also the only group to use summarisation. These strategies came from both the Listening Comprehension and Vocabulary Learning sections of the coding framework, suggesting a more global focus. The number of instances of using repetition, making connections, and force were reported most frequently by the participants from the CS group, who were also the only group to use the keyword strategy and searching. All of these strategies came from the Vocabulary Learning section of the coding framework, indicating a narrower focus. The same was true of learners from the CFoF group, who tended to use the vocabulary learning strategy waiting (for the teacher’s explanation) more, alongside relatively high instances of translation and making judgement.

Figure 2.Quantitative overview of strategy use by intervention group

Qualitative analysis: Strategy patterns by intervention group

The quantitative overview did not, however, reveal how exactly the strategies were used, i.e., whether there were any particular patterns of strategy use within one particular intervention group. Further qualitative analysis with the 12 interview transcripts found that there were two patterns of strategy use in the L2 group, and one type of combination of strategies in the CS and one in the CFoF group.

The L2 group

In this group one pattern was the combination of matching with familiar words and translation, while the other was combining selective attention to known words, translation and making judgement. Selective attention and matching both involved familiar or known words but the former involved a degree of intentionality or forward-planning absent from the latter. The first of these patterns is shown in the following example from L2S1HS, a more skilled learner in the L2 group, in relation to the word tourism:

R:So when you first heard the sentence including the word “tourism”, did you catch the word?

L2H1:Yes.

R:What did you do then?

L2H1:I guessed it would have something related to “旅游”(tour).

R:So what you did was to link tourism with the familiar word tour, is that correct?

L2H1:Yes. I was quite sure that the word should have something to do with “旅游”(tour).

When L2H1 first heard the sentence, she realized that there was an unknown vocabulary item. She then matched the unfamiliar item with a familiar item ‘tour’. At the same time, there was an on-going process of translating the English word ‘tour’ to its L1 meaning in her mind. She continued:

R:So how did you figure out it was actually “旅游业”(tourism).

L2H1:With the help of your English explanation. I noticed that in your explanation, there is “industry”, and I realised that it was “工业”(industry). So I supposed the meaning of the word could be “旅游业”(tourism).

R:You actually translated the meaning of the word into Chinese?

L2H1:Yes, I only translated the familiar part of your English explanation into Chinese.

L2H1’s attention was firstly focused on the word ‘industry’ which was familiar to her which she then translated into the L1. Instead of understanding the remaining part of the explanation, she linked ‘industry’ with ‘tour’ which she had guessed previously and successfully worked out the L1 meaning of tourism .

She was, however, still struggling with the L2 explanation provided and moved to making judgement on that, having discovered that the direct translation of that explanation did not seem quite right to her (“does not match the word itself”, which she had initially interpreted as ‘tour industry’ (旅游工业). Indeed she seemed content to arrive at a global understanding, aided by further application of the matching with familiar words strategy only rather than consciously striving to commit the word to memory, as the following suggests:

L2H1:I still think your explanation does not match the word itself.

R:Since you were not allowed to write down the meaning of the word, what did you do to remember the word?

L2H1:I actually did not make any efforts to remember it. Firstly, I knew the word tourist, so I could link tourism with tourist. Second, I was very confident that tourism means ‘旅游业’ (tourism). So I think I have remembered that word.

Similar patterns were found for L2L1, a less skilled participant, this time illustrated through the phrase figure out.

R:Did you recognise figure out when you heard this sentence in my class?

L2L1:Yes, it seems that the pronunciation is very familiar to me…maybe I have heard about this collocation before.

R:What did you do when I was explaining this collocation to you?

L2L1:I tried to translate the English meaning that you provided into Chinese…but I failed. I knew that in your explanation, ‘understand’ should be a familiar word to me, but I couldn’t remember the exact Chinese meaning for ‘understand’…your explanation is too long…very difficult to me.

The student initially tried to match figure out with a potential familiar vocabulary item which has similar pronunciation (matching with familiar words). When provided with the L2 explanation, she tried to translate the L2 explanation word by word into Chinese (translation). She failed, however, as she forgot the L1 translation of the word ‘understand’ used in the L2 explanation, although she claimed that ‘understand’ was a familiar word to her (selective attention to known words). At the same time, she also felt that the L2 explanation provided by the teacher was too long and difficult, most likely because of her low level of existing vocabulary knowledge. Therefore, she treated the vocabulary explanation in a fairly passive way and was likely to give up at the end.

Thus, while the higher proficiency L2 learner was more active and successful in their strategy use than the lower proficiency L2 learner in terms of understanding the teacher’s explanation, in neither student was there clear evidence of successfully committing the target item to memory.

The CS group

One type of strategy combination was found in this group, where largely the opposite of what was noted for the L2 group emerged. That is to say, their focus was less on understanding the listening segment and more on committing the explained vocabulary item to memory. All four CS participants tended to combine waiting, force and repetition, with only CSH1 going beyond these strategies (see Strategy use by proficiency level section). The following extracts relate to the word threaten.

The identified combination of strategies and focus on memorisation can be illustrated through an excerpt from the interview with higher proficiency learner CSH2, who did however have lower listening proficiency:

R:What did you do when I played the listening segment including the word threaten ?

CSH2:I remember that I actually did not clearly recognise what was in the listening segment. Therefore, what I did was to wait for you to explain.…

R:What were you thinking after I explained this word to you?

CSH2:I recognised that this word was unfamiliar to me and I forced myself to remember it.

R:How exactly did you force yourself to remember?

CSH2:I repeated both the pronunciation of the word and its Chinese meaning many times.

R:You repeated them? In your mind or orally?

CSH2:Orally in a very low voice. In my opinion, reading out the word and its meaning can help me remember, especially when they were repeated orally for many times.

After the listening material was played by the teacher, the student failed to catch any information and did not understand the sentence. Therefore, she used the strategy waiting and waited for the teacher to provide the translation rather than arriving at one herself, perhaps because she had worked out from earlier intervention sessions that the teacher gave L1 explanations in due course. The student therefore seemed to intentionally pay attention largely to the L1 explanation, with less focus on understanding the L2 listening passage. The student then forced herself to remember the word by reading it out and its corresponding Chinese meaning several times (repetition).

The above combination of strategy use was also reported by lower proficiency student CSL2 in relation to the phrase at ease that occurred in the fourth intervention session.

R:Did you recognise the collocation at ease when you heard it for the first time?

CSL2:Yes, I did.

R:What did you do then?

CSL2:Nothing.

R:Just recognised it and did nothing?

CSL2:Yes, and waited for you to explain.…

R:So what did you do to cope with my explanation?

CSL2:Forced myself to remember.

R:How exactly did you force yourself to remember?

CSL2:I linked the Chinese meaning and the English word and read them out for a few times.

While this students’ focus on vocabulary learning rather than understanding and the combination of strategies here is very similar to that seen in CSL2, the higher proficiency learner, commenting that “reading out the word and its meaning can help me remember, especially when they were repeated orally for many times”, arguably displayed a higher degree of evaluation of self and strategy use. Notable also is that both of these CS learners had relatively low listening proficiency, even CSH2 who was in the higher proficiency group overall, perhaps further explaining their focus on vocabulary learning rather than on understanding of the passage.

CFoF group

In the CFoF group, one strategy use pattern was reported by the four interviewees: selective attention to known words, waiting and translation. These individual strategies were also used by the other two intervention groups, but combined with different strategies. In particular, the L2 group used selective attention to known words and translation in conjunction with making judgement rather than waiting.

The first example given relates to the item threaten.

R:What did you do when you heard the sentence including threaten?

CFoFH1:I recognised that there was an unfamiliar word to me and I linked this word ( threaten ) with ‘angry’ in the same sentence. …

R:What was on your mind at the time when I was explaining the word to you?

CFoFH1:You gave us another word threat which means ‘威胁’. I followed your instruction and thought about threaten, which means ‘恐吓’. The two Chinese meanings are connected … The two words have a similar Chinese meaning but are different in forms.

When listening to the sentence including the target word, CFoFH1, a more skilled learner, firstly linked it with ‘angry, a familiar word for her (selective attention on known words). She then waited for and listened to the teacher’s explanation (waiting). During this period, she reflected on the individual vocabulary items, linking threaten with ‘threat’, thus evaluating and processing individual lexical items at quite a deep level. She also seemed to try to go beyond the teacher’s explanation by using the translation strategy when she tried to compare and contrast both the meaning and form of the words herself.

Furthermore, the strategy the CFoF learners used most frequently was making judgement (for an example see CFoFH1 as discussed in relation to Research Question 2). This involved them evaluating the usefulness of the teacher’s explanation and as such might be considered a metacognitive strategy. Thus while the CFoF group seemed to firstly rely more on receiving information about the item’s meaning and use rather than trying to work this out themselves through their frequent use of waiting, they then followed this up with strategies for focusing on the word itself and evaluating its meaning. Overall, one might call their approach a focused one, concentrating on vocabulary learning rather than on overall understanding of the listening passage, in contrast most clearly with the L2 group.

Research Question 2: Strategy use by proficiency level

In addition to differences by intervention group, there were also instances of differences in strategy use by proficiency level. On a quantitative level, high proficiency learners reported a greater number of instances of different types of strategies than low proficiency learners.

More qualitatively, high proficiency learners in the CFoF and L2 group seemed to be prompted by the teacher’s explanations to adopt fairly complex processing. For example, CFoFH1 combined selective attention to known words, translation and summarisation, as outlined above. By contrast, the CFoF and L2 conditions seemed to provide information of a complexity that was sometimes too great for the low proficiency learners, which was then reflected in their strategy use. This can be illustrated by comparing how two learners from the CFoF group with different proficiency levels reacted quite differently to not fully understanding the teacher’s explanation:

Both students felt that they knew the target word [matching with familiar words] but failed to recall its Chinese meaning [translation]. Yet while CFoFH1 actively turned to the context to guess the meaning [linguistic inferencing], CFoFL1 gave up at this stage, responding with silence. In addition, while CFoFL1 said she was waiting for the teacher to explain, CFoFH1 made a judgement about the teacher’s explanation and engaged in monitoring, commenting that the second part of the explanation was not relevant for understanding the target item. To learn the word she remembered it as a collocation [remember as a collocation], how it had featured in the listening passage, while CFoFL1 just gave up [give up], commenting that the word was too long.

By contrast, in the CFoF group there were instances of learners with lower levels of listening proficiency being helped by the teacher’s explanation, as in the following example from CFoFH2, who had high PVK but low listening proficiency. Even though the latter did not permit her to recognise the collocation at ease within the listening passage itself, she applied strategies to help her remember it that seemed prompted by the teacher’s explanation:

R:Did you recognise the collocation “at ease” when you heard it for the first time?

CFoFH2:No, I didn’t.

R:So what was on your mind when I showed the sentence to you?

CFoFH2:I thought about “ease” and “easy”. [matching with familiar words]

R:Well, did you actually form this link before I explained this word to you?

CFoFH2:Yes. When I first saw “ease”, I can think about “easy”. [matching with familiar words]

R:So what was on your mind when I was explaining this word to you?

CFoFH2:You also mentioned that “ease” came from “easy”, which gave me a deep impression because this confirmed my hypothesis. [making judgement] Then, I used the Chinese meaning of “easy” to help me remember the meaning of this collocation. [translation]

Rather differently, and as outlined earlier, low proficiency learners in the L2 group seemed to have benefited the least from the intervention. Their ability to use a range of strategies flexibly and in effective combinations seemed often to be hampered by their failure to understand enough of the input in the first place, either from the passage or the teacher’s explanation, as in the following, which contrasts responses from a low and a high proficiency learner:

The use of a combination of different strategies by the more proficient L2H1 — identifying an unfamiliar item; using linguistic inferencing to work out its meaning from the context; matching the target word with a familiar word; translating that word into Chinese, and finally evaluating and monitoring her interpretation — contrasts with the much more limited strategy use of learner L2L2. She reportedly did nothing after listening to the sentence, except waiting for the teacher’s explanation [give up + waiting], presumably because her lower vocabulary size and listening proficiency meant that she understood little of it and hence had little with which to be ‘strategic’. Although she tried to translate the English meaning of the word provided into Chinese, she failed to do so and decided just to ignore that part.

Finally, for the CS group, there were differences between learners of higher and lower listening proficiency in terms of whether they attended just to the teacher’s explanation or also used strategies to make sense of target items within the listening passage itself, or to relate the explanation to what was heard, as indicated below. Here, CSH2 was overall considered to be high proficiency, but her listening proficiency was relatively low:

Here, CSH2 used few strategies apart from force and repetition, especially noticeable in comparison with the variety used by CSH1.

Discussion

Our research questions explored learners’ strategy use when listening to different kinds of teacher explanations of target vocabulary items, whether those strategies varied according to the type of vocabulary explanation and learner proficiency, and whether any differences in strategy use related to gains in vocabulary and listening proficiency. Our qualitative analysis showed that the attentional direction (Ellis, 1999Ellis, R. (1999) Factors in the incidental acquisition of second language vocabulary from oral input. In R. Ellis (Ed.), Learning a second language through interaction (pp. 35–62). John Benjamins. ) of learners from the CFoF group in general seemed to be on the target words, made more salient by the cross-linguistic information provided and hence encouraging greater ‘noticing’ (Schmidt, 1990Schmidt, R. W. (1990) The role of consciousness in second language learning. Applied Linguistics, 11 (2), 129–158. ) and deeper thinking about them. In addition, making judgement was used frequently by the CFoF group, implying greater evaluation of lexical items, and hence greater involvement load as per Hulstijn and Laufer (2001)Hulstijn, J. H., & Laufer, B. (2001) Some empirical evidence for the involvement load hypothesis in vocabulary acquisition. Language Learning, 51 (3), 539–558. and thence better learning, especially when learners had higher levels of PVK. The L2 group, as well as using more strategies in total than learners in other two groups, also focused more on global understanding than committing vocabulary to memory. The opposite was the case for the CS group who focused on memorisation and reported a narrower range of arguably more passive strategies.

Turning to pre-existing proficiency, having a larger vocabulary size is recognised as being related to better comprehension (e.g. Webb, 2021Webb, S. (2021) Research investigating lexical coverage and lexical profiling: What we know, what we don’t know, and what needs to be examined. Reading in a Foreign Language, 33 , 278–293. http://​hdl​.handle​.net​/10125​/67407). Perhaps unsurprisingly therefore, more proficient learners who knew more words to start with likely had greater understanding of the listening input and the teacher’s oral explanations as a whole. That then may have enabled them to engage in more advanced strategic behaviour, using a combination of different strategies, compared with the less proficient learners, especially within the L2 and CFoF groups. Additionally, the findings in Zhang and Graham (2020b)Zhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. indicated that, regardless of the intervention condition, the higher learners’ listening proficiency was before the intervention, the greater short-term vocabulary gains they made; also, that learners with a lower level of listening proficiency were helped more by the CFoF approach. The latter may have occurred because they gained information from the cross-linguistic explanations information that they were less able to extract from the listening passage alone, as in the example given from CFoFH2. In turn, however, the information provided by the teacher may have obviated the need for the learners in the CFoF group to try to gain an overall understanding of the listening passage, which may explain why they saw the smallest gains in listening proficiency (Zhang & Graham, 2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. ). Finally, Zhang and Graham (2020b)Zhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. also found that in the CS group, more proficient listeners benefited the most for vocabulary learning, perhaps because they were able to supplement the basic information given in the teacher’s L1 translation of the word with information gained from the listening passage regarding its use, which learners with a lower level of listening proficiency were unable to do. This was illustrated in the contrasting examples of strategy use from CSH1 and CSH2. In other words, taken together these findings indicate that the strategies learners use depends on a combination of both learner factors (such as PVK) and the teaching they experience. The more support offered by the teacher (higher in the case of the CFoF group, and lower for the L2 group), the more learners need to apply strategic behaviour.

Our analysis of strategy use thus enabled us to better understand the quantitative findings of Zhang and Graham (2020aZhang, P., & Graham, S. (2020a) Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on form and incidental learning. Language Teaching Research, 24 , 765–784. , bZhang, P., & Graham, S. (2020b) Learning vocabulary through listening: The role of vocabulary knowledge and listening proficiency. Language Learning, 70 , 1017–1053. ), giving insights into the impact of PVK, listening proficiency and intervention conditions on vocabulary learning as revealed by the strategies learners used as they responded to the teacher’s vocabulary explanations. While overall the CFoF approach led to the greatest vocabulary gains, it was much less helpful for developing listening proficiency. The analysis of strategy use by the CFoF group indicated that the intervention encouraged learners to focus on the target lexical items much more than trying to understand the original listening passage. Perhaps for that reason, CFoF also emerged as potentially helpful for learners with weaker listening skills, who were able to gain rich information about lexical items without needing to understand the original passage so fully. Somewhat differently, the learners who made the smallest vocabulary gains were the low proficiency L2 learners, whose strategy use indicated that they struggled to make sense of either the passage or the teacher’s explanations. The higher proficiency learners in the L2 group, who made the smallest gains of all the higher proficiency learners who were interviewed, also seemed to focus less on vocabulary items and more on trying to make sense of the original passage.

There are some limitations to our study, largely in relation to the challenges of using stimulated recall interviews with adolescent, relatively low proficiency learners, whose responses to the interviewer tended to be short, with an unwillingness or low capacity to elaborate further. Future studies with such learners might want to consider incorporating stimulated recall with such approaches as focus group interviews with such learners to gain further insights. In addition, future research might usefully explore the use of stimulated recall to gain insights into listening strategy use in different learning contexts, such as content-focused settings. While there is important new emerging evidence of the central role of listening strategy use for processing content knowledge (Zhou et al., 2023Zhou, S., Fung, D., & Thomas, N. (2023) Towards deeper learning in EMI lectures: The role of English proficiency and motivation in students’ deep processing of content knowledge. Journal of Multilingual and Multicultural Development, 1–16. ), using stimulated recall rather than broader tools such as retrospective interviews would provide clearer insights into strategy use in relation to specific aspects of the input learners encounter, as studies are beginning to show (Fung & Lo, 2023Fung, D., & Lo, Y. Y. (2023) Listening strategies in the English Medium Instruction (EMI) classroom: How students comprehend the teacher input. System, 113 , ; Zhou & Thompson, 2023Zhou, S., & Thompson, G. (2023) Exploring the role of English proficiency, self-efficacy, and motivation in listening for learners transitioning to an English-medium transnational university in China. System, 113 , ). In spite of the above limitations, our study, does, however, advance the field by showing the merits of using a qualitative analysis of strategy use to help better understand the impact of classroom-based interventions, and also, likely for the first time, to provide qualitative insights into the role of ‘involvement load’ (Hustijn & Laufer, 2001Hulstijn, J. H., & Laufer, B. (2001) Some empirical evidence for the involvement load hypothesis in vocabulary acquisition. Language Learning, 51 (3), 539–558. ) for vocabulary learning. Finally, it also highlights the need for teachers to consider strategy-focused interventions to help learners to develop the most appropriate strategies to use in response to the teacher’s vocabulary explanations.