Learner beliefs about vocabulary teaching in secondary ESL classrooms: a proficiency and grade-level analysis

5.1 Research context and participants

The study was conducted in a secondary school in Hong Kong that emphasised expanding students’ English vocabulary. Originally, the study included 572 participants, ranging from secondary one (S1) to secondary five (S5) students, with an age range of approximately 12–16 years old. However, 16 of them did not participate in the school examination, resulting in a final sample of 556 participants (320 female and 236 male). The final sample included 94 S1, 117 S2, 109 S3, 119 S4, and 117 S5 students. These students exhibited varying English proficiency levels, categorised as low (n = 96), mid (n = 378), and high proficiency (n = 82). For details on how proficiency levels were determined, see Section 5.4.

5.2 Research instrument

We developed an online questionnaire in Chinese and English and administered it to the student participants to gather both quantitative and qualitative data. The questionnaire was developed based on key issues identified in previous research on L2 vocabulary teaching in the works of Nation (2022), Schmitt and Schmitt (2020), and Thornbury (2002), as well as teachers’ opinions on vocabulary teaching and learners’ common beliefs. It had two sections. The first part consisted of 12 items, 10 four-point Likert-scale items (1−strongly disagree and 4−strongly agree) and two open-ended items that focused on students’ beliefs about vocabulary instruction. The Likert-scale items (items 1−10) included questions about the selection of target vocabulary, the suitability of using L1 translation, various instructional techniques and resources, and the importance of strategy-based vocabulary instruction (Cronbach’s α = 0.650). The two open-ended questions inquired about the best way for English teachers to facilitate the understanding of an unfamiliar word’s meaning and its long-term acquisition. The second part of the questionnaire examined students’ beliefs about eight types of classroom activities for vocabulary building (Items 11a−11h) (Cronbach’s α = 0.760) and students’ beliefs about the use of vocabulary activities (Items 12−16) (Cronbach’s α = 0.606) on a four-point Likert scale. Two open-ended questions were also included to gather students’ justifications for the activities they liked or disliked for vocabulary learning and identify other preferential activities. The questionnaire was piloted on a group of junior (n = 25) and senior secondary students (n = 26) at another school resulting in new vocabulary activities being identified which were added to the questionnaire; the category “no idea” was also added to avoid forcing a choice upon respondents when they had no opinion.

5.3 Procedures

One week before data collection, a briefing was conducted to introduce the guidelines for administering the questionnaire to all teachers whose students were to participate in the study. To ensure a high response rate and address any queries, teachers were instructed to bring their students to the school’s computer room during an English language lesson to complete the online questionnaire. Students were informed of their right not to participate in the study, and their consent was obtained electronically. Each student was given approximately 40  minutes to complete the questionnaire. The students’ English scores from the school examination were obtained from an English panel head to provide data on their English proficiency levels for statistical analysis.

5.4 Data analysis

The questionnaire responses and English proficiency data were analysed using SPSS 26.0. For each part of the questionnaire that covered students’ beliefs about vocabulary teaching and activities, descriptive statistics provided a general picture. Subsequently, a principal component analysis (PCA) was conducted on each section of the questionnaire to identify the different belief components. Based on the identified components, follow-up analyses were performed to examine how students with different proficiency levels (low, mid, high) and in different grades (S1 to S5) differed in their beliefs. The categorisation of proficiency levels was based on students’ English examination scores for each grade, as explained below. As the examination papers were different for each grade, the scores were not directly comparable across the five grades. Therefore, z-scores were calculated for individual students in their respective grades to provide a standardised examination score for each of them. Students with z-scores less than -1 were classified as having low proficiency, those between -1 and 1 as mid-proficiency, and those higher than 1 as high proficiency. Thus, a student who was considered to have high proficiency would be a high achiever in their cohort (see Chung and Fung 2023 for details). To determine whether there were significant differences in beliefs among students with different proficiency and grade levels, 3 × 5 mixed ANOVAs were conducted, followed by post-hoc Bonferroni tests. The component scores generated by the PCA were used as the dependent variables. Kruskal-Wallis tests followed by Mann-Whitney U tests were performed for components that were not normally distributed. For the open-ended questions, we identified and categorised themes, which largely aligned with the content of each question. The quantitative and qualitative data results were compared, allowing themes to emerge; representative excerpts were extracted from the open-ended responses and triangulated with the quantitative findings, resulting in a better understanding of the learners’ beliefs.

6.1 Learner beliefs about vocabulary instruction

Table 1 shows the descriptive statistics of the students’ beliefs about various aspects of vocabulary instruction. Over 80 % of the students showed agreement that teachers should encourage the guessing of word meanings (item 9, M = 3.17) and incorporate the teaching of VLS into lessons (item 10, M = 3.12). Around three-quarters of the students tended to agree with the importance of using word repetition (item 6, M = 3.00) and visual aids (item 8, M = 3.02) in vocabulary instruction. However, there were mixed beliefs regarding vocabulary selection and coverage. While over 80 % of students indicated teachers should cover as many unknown words as possible (item 1, M = 3.13), students held differing opinions on whether they should be given autonomy in selecting the words to be taught as part of their vocabulary instruction (item 5, M = 2.64). Approximately half of the students indicated they should always be given the chance to choose the words they want to learn, while about one-third held the opposite view. Whether to use the L1 in vocabulary instruction was also somewhat mixed (item 7, M = 2.31), with 36 % of students indicating L1 use should be avoided while 55 % indicated it should not. More than two-thirds of the students tended to disagree with item 4 (M = 2.08), signalling that they believed teachers should use a variety of resources alongside textbooks and storybooks when teaching vocabulary. Finally, over 70 % of the students disagreed with item 2 (M = 2.02) and item 3 (M = 2.05), indicating that addressing both the meaning and the usage of the target words was considered essential in vocabulary instruction.

To identify the underlying components of students’ beliefs about vocabulary instruction, we conducted a PCA with varimax rotation (Table 2). The analysis resulted in the extraction of three components with eigenvalues greater than one, explaining 56.10 % of the total variance. Component 1, labelled “learner agency,” highlighted the proactive role students indicated they should play in managing their own learning within the context of vocabulary instruction. Learner agency is linked to students’ awareness of their responsibility for their actions (Matsumoto 2021) and their ability to act independently and make their own choices (Manyukhina and Wyse 2019). Therefore, this component included items that minimised the teacher’s role in covering various aspects of vocabulary knowledge (items 2 and 3) and selecting vocabulary items outside the textbooks (item 4) while emphasising the students’ role by allowing them to choose the words they wanted to learn (item 5). Component 2, labelled “explicit vocabulary instruction,” promoted the intentional learning of lexical items (Schmitt and Schmitt 2020). It contrasted with Component 1 by embodying a teacher-centred approach that emphasised the direct teaching of vocabulary. Component 2 focused on whether teachers should teach as many vocabulary items as possible (item 1), use repetition as a memory aid (item 6), and employ visual aids to illustrate word meanings (item 8). This component underscored the need for teachers to deliver structured and visually supported instruction to enhance the effectiveness of vocabulary teaching. Component 3, “strategy use” encompassed items that advocated for avoiding direct translations (item 7), encouraging students to guess word meanings independently before receiving explanations (item 9), and equipping students with strategies for autonomous vocabulary learning (item 10). In essence, component 3 sought to uncover the participants’ beliefs about the importance of developing strategies that guide students to expand their vocabulary.

After identifying the three components, we performed two-way 3 × 5 ANOVAs/Kruskal-Wallis H Tests, followed by a series of post-hoc tests to examine whether students with different proficiency and grade levels held significantly different beliefs (Table 3). Results revealed no significant main effect for proficiency (F = 0.611, p = 0.543, partial ŋ² = 0.003) or grade level (F = 0.949, p = 0.436, partial ŋ² = 0.010) on component 1, “learner agency.” Additionally, no significant interaction effect was found (F = 1.635, p = 0.113, partial ŋ² = 0.035). For component 2, “explicit vocabulary instruction,” no significant effect was found for proficiency (F = 0.071, p = 0.931, partial ŋ² = 0.000). However, as the component scores were not normally distributed for grade level, we ran a Kruskal-Wallis H Test, which showed that students at different grade levels held significantly different beliefs (H(2) = 9.607, p = 0.048, partial ŋ² = 0.015). Post-hoc Mann-Whitney U tests revealed that S5 students held significantly more positive beliefs towards “explicit vocabulary instruction” in vocabulary lessons than their S1 (p = 0.006) and S3 (p = 0.040) peers. A similar trend was observed in S4 students compared to their S1 peers (p = 0.049).

Regarding component 3, “strategy use,” there was no significant main effect for grade level (F = 1.870, p = 0.115, partial ŋ² = 0.020). However, a main effect for proficiency was found (F = 7.949, p < 0.001, partial ŋ² = 0.042), indicating that students’ beliefs significantly differed between proficiency groups. Post-hoc analyses revealed that high-proficiency students viewed vocabulary teaching that focused on strategies more favourably than mid- (p = 0.013) and low- (p = 0.003) proficiency students. Additionally, there was a significant interaction effect (F = 2.367, p = 0.017, partial ŋ² = 0.050). Bonferroni analyses revealed that among S2 students, those who reached a high proficiency level held significantly more favourable beliefs than their mid-proficiency (p = 0.013) and low-proficiency (p = 0.009) peers.

In addition to providing responses to the Likert-scale items, students were also asked to share their perspectives on how teachers can effectively promote vocabulary instruction in the English language classroom. Several key themes emerged from our coding of the students’ responses. First, students emphasised the importance of visualisation in vocabulary instruction. They suggested that pictures, videos, and drawings can help them understand the meanings of unfamiliar words and remember them more easily. The use of visual support also made the learning process more engaging. One student wrote: “It is best to explain the words through pictures because it is boring to use chalk and talk only.” Besides visual aids, students stated that vocabulary is best taught through practical application. They commented that teachers should provide examples that illustrate how the words are used in context and give students opportunities to practise using the words themselves. For example, some students wrote that teachers should create stories with newly learned words or encourage them to make new sentences with the words.

Students also recognised the importance of VLS. They claimed that guessing at the meaning of unknown words and note-taking are effective strategies. For some students, vocabulary instruction was considered most effective when teachers explained word meanings using L1 translation (e.g., “Using Chinese is the best way to help us understand the meaning”) and plain language (e.g., “Teachers should explain it in simple words so that I can learn quickly”). Repetition in vocabulary instruction was also reported as a means to facilitate vocabulary acquisition (e.g., “Teachers should teach new words repeatedly to aid memory”). Nevertheless, some students maintained that teachers should incorporate various instructional practices and materials into their teaching, as exemplified by the comment that “There is no one best way to teach vocabulary.”

6.2 Learner beliefs about vocabulary activities

Regarding classroom activities that promote vocabulary building (Table 4), students favoured activities that are engaging and motivating; more than 80 % favoured movies (item 11e, M = 3.54), songs (item 11g, M = 3.39), and games (item 11c, M = 3.22). About 60 % of the students favoured group discussions (item 11d, M = 2.78), reading different types of texts (item 11f, M = 2.75), and vocabulary sharing (item 11h, M = 2.72). However, most of the students did not favour dictation, with approximately 60 % of them disagreeing with item 11a (M = 2.24). Regarding learning vocabulary through drama, the results were mixed (item 11b, M = 2.64), with half of the students preferring it, while 42 % held the opposite view. As for how classroom activities should be incorporated into vocabulary teaching (see Table 5), more than three-quarters of the students indicated that vocabulary activities should be conducted in small groups (item 15, M = 3.11) and that homework should always be assigned (item 16, M = 2.96). Two-thirds of the students appeared willing to have the same activities regularly (item 12, M = 2.83). However, beliefs about whether to conduct games (item 13) and to use only the activities provided in textbooks for vocabulary teaching (item 14) were mixed, as indicated by mean scores of 2.40 and 2.49, respectively.

A PCA was conducted to identify the underlying components of students’ beliefs about vocabulary activities. Table 6 reveals four components explaining 59.71 % of the variance. Components 1 and 2 reflected students’ receptivity to different types of vocabulary activities, while components 3 and 4 were related to beliefs about how teachers should incorporate those activities into lessons. Component 1, labelled “conventional activities,” was loaded with five items, including reading (items 11f), vocabulary sharing (item 11h), group discussion (item 11d), drama (item 11b), and dictation (11a). Component 2, labelled “engaging activities,” was loaded with three items, including movie appreciation (item 11e), song appreciation (11g), and games (11c). Component 3 had high loadings of three items reflecting indifference towards activities, including the belief that it is fine for teachers to use only activities from textbooks (item 14), not to conduct any games (item 13), and to repeat the same activities regularly (item 12). Component 4 contained two items that reflected a reliance on teachers, including the belief that teachers should always assign homework after vocabulary activities (item 16) and conduct vocabulary activities in small groups in class (item 15).

Two-way 3 × 5 ANOVAs were conducted to assess proficiency- and grade-level effects on student beliefs about vocabulary activities. Table 7 shows there were no significant main effects of proficiency (F = 2.839, p = 0.060, partial ŋ² = 0.015) or grade (F = 1.845, p = 0.120, partial ŋ² = 0.019) on “conventional activities,” and no significant interaction effect was found (F = 1.390, p = 0.199, partial ŋ² = 0.029). In contrast, a significant effect of proficiency was found for “engaging activities” (F = 3.765, p = 0.024, partial ŋ² = 0.019). Follow-up analyses revealed that low-proficiency students had a significantly stronger preference for engaging in activities than mid-proficiency students (p = 0.027), while the differences between other groups were not significant. No significant main effects of proficiency or grade, and no significant interaction effect were identified for both “indifference towards activities” and “reliance on teachers.”

Students were asked to explain their beliefs about various classroom activities designed to enhance vocabulary acquisition. Students who expressed a preference for games, movies, and songs explained that their choice was mainly due to the engaging and stimulating nature of these activities. Specifically, students perceived games, movies, and songs as interesting and enjoyable activities that could enhance their attention and focus, thus motivating them to engage in vocabulary learning (e.g., “Games can encourage me to learn vocabulary more”). Students also noted that these activities often created a relaxing classroom environment that was conducive to vocabulary learning (e.g., “I think such a relaxing environment is good for learning new vocabulary”). However, these activities were not always well-received. For example, one student considered games meaningless, stating, “Learning is more important than playing.” Vocabulary learning through movies presented challenges, such as difficulties in understanding the dialogues (e.g., “I cannot understand what the characters say”), and the value of learning vocabulary through songs was questioned, with one student noting, “Lyrics nowadays usually don’t contain many useful words.” Regarding drama, some students found that it helped them engage in vocabulary learning, while others perceived it as boring and time-consuming for vocabulary development. Regarding dictation and vocabulary sharing, students who strongly preferred these activities indicated that these activities facilitate vocabulary retention (e.g., “They help us remember the words better”) and growth (e.g., “Vocabulary sharing is easy to prepare. Both teachers and students can share their thoughts on different words, allowing everyone to learn from the sharing”). However, other students who did not prefer these activities often described them as boring, demotivating, and a waste of time. Specifically, dictation was perceived to facilitate only the retention of word form (e.g., “Dictation only focuses on the spelling of a word but not its meaning”), without facilitating acquisition (e.g., “After dictation, we don’t actually learn the words”). Dictation also appeared to exert pressure as there was a desire not to fail such assessments. Group discussions were seen as potentially making lessons more interactive, although they were also considered less effective than other activities and prone to causing distractions. Contrary to the general impression that interesting activities are favoured in class, the qualitative responses surprisingly revealed a preference for omitting them based on the belief that they compromise the teaching of the main syllabus for examinations.

7.1 Students’ beliefs about vocabulary instruction: a call for promoting autonomous vocabulary learning and the development of VLS

Our findings suggest that students predominantly believe that vocabulary instruction should include VLS, and cover as many vocabulary items as possible. Consistent with prior research (Amiryousefi 2015b), the use of visual aids was supported, and the belief in the effectiveness of word repetition aligns with the findings of previous investigations (Nation 2022; Schmitt and Schmitt 2020). Extending beyond these established principles, our study categorised students’ beliefs about vocabulary instruction into three components – “learner agency,” “explicit vocabulary instruction” and “strategy use.” Among these components, students who were more proficient in English showed a preference for strategy-focused instruction, which was less prevalent among students with mid and low proficiency. Students in S4 and S5 were more positive towards “explicit vocabulary instruction” than those in S1. This divergence underscores the nuanced role that proficiency and grade level play in students’ beliefs, echoing previous research findings (e.g., Brown 2009; Wan et al. 2011). However, our findings also revealed that senior secondary students (S4 and S5) tend to rely more on their teachers to develop their vocabulary skills.

Our analyses also revealed a low endorsement of “learner agency,” indicating a reluctance among students to engage in autonomous vocabulary learning. This may be attributed to the examination-oriented culture prevalent in Hong Kong’s senior secondary curriculum, as discussed by Yung (2019), which prioritises teacher-led instruction. This cultural context may inadvertently discourage self-directed learning activities such as voluntary extensive reading, leading students to rely more on their teachers for vocabulary learning. Given the importance of VLS on vocabulary learning (e.g., Prichard and Atkins 2021; Schmitt and Schmitt 2020), teachers should integrate it into their teaching to promote incidental learning (e.g., guessing meanings from context) and consolidate word learning. Employing VLS can also raise learners’ awareness of strategic learning, especially among low-proficiency learners who may not perceive its effectiveness.

7.2 Students’ beliefs about vocabulary activities: the need for a balance between conventional and engaging activities

Students, particularly those with lower English proficiency, expressed a preference for engaging activities like games, movies, and songs. The finding contrasts with earlier research indicating a preference for teacher-led activities, like drilling exercises, over student-centred communicative activities (Hawkey 2006; Peacock 1998). This disparity may stem from contextual differences in the educational settings or cultural preferences that influence student engagement and motivation. For example, Hawkey’s (2006) student participants came from a variety of schools, ranging from elementary to secondary levels in Italy, while Peacock (1998) focused on university students. Despite some resistance to conventional activities like dictation, students in our study still acknowledged their effectiveness for enhancing vocabulary retention and learning word forms (e.g., spelling). The complexity of student beliefs suggests a possible cognitive dissonance, where students recognise the benefits of less favoured activities yet prefer more engaging ones for vocabulary learning.

Although engaging activities can motivate students to learn new words, it is important not to overlook the effectiveness of conventional activities. As previous literature suggests, a variety of instructional techniques and resources is needed to promote vocabulary learning (e.g., Nation 2022). Thus, we propose a balanced approach, emphasising the importance of combining engaging and conventional activities to target different aspects of word knowledge effectively. For example, although games can be motivating, their educational value depends on their ability to target specific learning outcomes. This was evidenced by most of our participants who disliked engaging activities, preferring to focus on examination-related syllabus content, which highlights the need for purposeful implementation. Pedagogically, teachers should consider learner diversity by employing a wide range of vocabulary activities while making informed decisions to facilitate the learning of various aspects of word knowledge. Games can be used purposefully to create contexts where learners engage with words meaningfully, which enhances retention (Lee 2022). We would also emphasise the importance of having teachers demonstrate how different activities target various aspects of word knowledge, addressing the common learner practice of focusing predominantly on word form and meaning while neglecting other important aspects such as collocations and word associations (Chung and Fung 2023). It is these ancillary aspects of word knowledge that reinforce the value of diverse vocabulary teaching activities.

7.3 Consideration of grade- and proficiency-level differences to cater for learner diversity

Our study adds to the existing literature on learner beliefs (e.g., Brown 2009; Wan et al. 2011) by focusing on vocabulary teaching, revealing grade- and proficiency-level differences in students’ beliefs about vocabulary instruction and activities. Our focused exploration provides new insights into the differentiated approaches required for effective vocabulary teaching. Senior secondary students displayed a greater dependency on teacher-led instruction than their junior counterparts, suggesting a shift in learning strategies as students advance in their educational journey. This dependency may also reflect the increasing academic pressures that characterise senior secondary education in Hong Kong, emphasising the need for tailored instructional approaches that address these pressures while promoting effective vocabulary learning.

Teachers should consider these differences when planning curriculum and classroom activities, ensuring that they provide a rich mix of vocabulary learning opportunities that cater to the varying needs of students across different proficiency and grade levels. For example, incorporating multimedia resources like videos with captions has been found to enhance vocabulary acquisition, particularly for high-proficiency learners (Chen et al. 2018). The spacing of vocabulary activities across different lessons can also be considered as studies indicate that low-proficiency learners benefit more from spaced learning opportunities spread over several weeks, instead of being concentrated in one single session (e.g., Lee et al. 2021). Regarding curriculum design, educators should prioritise the development of VLS and promote reading for pleasure to foster broad vocabulary development at the junior secondary level. At the senior secondary level, where the curriculum may not provide sufficient time for vocabulary learning, teachers can implement more teacher-centred activities targeting specific words essential for understanding content in other subjects (see Fung and Chung 2024).