1. Introduction
Word lists have a long history. Research on word lists began to generate attention in 1953 when Michael West created the General Service List (GSL) and has flourished since 2000 when Averil Coxhead developed the Academic Word List (AWL). In the last ten years alone, a large number of word lists have been created and published. Word lists are a key theme in books (e.g., Nation, Reference Nation2016) and edited volumes (e.g., Webb, Reference Webb2020) about teaching and learning vocabulary. However, there are no comprehensive reviews of the applications of word lists to language learning and teaching. This is surprising because researchers typically justify the creation of new word lists by highlighting their value to learners. Moreover, the value and validity of lists may not be transparent to teachers, learners, and researchers, which hinders the chance for word lists to be implemented in pedagogical contexts.
This review is a timely response to these issues. Covering 50 publications from the last decade, it will identify current themes and issues in research on word lists for second language learning and teaching. It aims to raise awareness of (a) the value of word lists for language research and pedagogy and (b) the areas that need further attention in word list research. The review will focus on studies of word lists for learners of English as a second or foreign language. To provide a comprehensive overview of word list research, it will cover studies that are directly related to word lists (word list development and validation), as well as those using word lists to inform language learning, teaching, and assessment.
To offer readers a clear overview of current trends in word list research, we will only review works published in the last ten years (2013–2023). However, where necessary, we will link studies to seminal works such as West’s (Reference West1953) General Service List and Coxhead’s (Reference Coxhead2000) Academic Word List. Although a large number of word lists have been developed in the last decade, this review will focus on 50 word list studies that have been published and discussed in major international peer-reviewed Applied Linguistics and TESOL journals. This approach allows us to better identify the trends in word list studies that have generated interest over the last decade (2013–2023). References for these lists are presented in Appendix 1 (online).
2. What kinds of word lists have been published in the last decade?
Figure 1 presents the publication dates of the 50 word list studies published in the last ten years. After reaching a peak of eight studies published in 2018, slightly fewer studies have been published since then with only three published in 2022, and two lists published in 2019 and 2023. This suggests that there may now appear to be less of a need for new lists, or that word lists as a topic is not generating as much interest from researchers and publishers.
Number of word list studies published in the last decade (N = 50).
Word lists can be classified into subtypes according to their purpose (general, academic, and technical), modality (written, spoken, and digital), and lexical components (single word and multiword). In terms of purpose, word lists can be categorized into general service lists (aiming to help learners deal with the lexical challenges of different communicative situations in everyday language), academic vocabulary lists (aiming to help learners cope with the lexical challenges of communication in a wide range of academic disciplines), and technical vocabulary lists (aiming to help learners cope with the lexical challenges of communication in a specific discipline or professional area). Although general words and academic words still received attention from word list developers in the last decade, there was growing interest in creating lists of technical vocabulary (Fig. 2). In fact, 66% of the word list studies published in the last decade focused on technical vocabulary whereas academic word lists and general word lists only accounted for 20% and 14%, respectively. The increasing number of studies developing technical lists could be the result of widespread use of English as the medium of communication in academic and professional settings.
Word lists studies classified by list purpose (N = 50).
As for modality, written discourse is still dominant, being the focus of 58% of the word list studies (Fig. 3). Yet increasing attention has been given to spoken and digital discourse. Since 2013, almost every year, at least one study has created lists to help learners deal with the challenge of communication in either spoken discourse alone or both spoken and written discourse. Since 2022, word lists designed to support comprehension in digital discourse have appeared. In fact, a reasonable percentage of studies published in the last decade have developed lists from written plus spoken discourse (24%), spoken discourse alone (14%), and digital discourse (4%). The growing attention to spoken and digital discourse in word list research acknowledges the importance of comprehending spoken communication and the increasing need to understand digital discourse during and post the COVID pandemic period. This trend may be thanks to the findings of corpus-driven studies revealing that linguistic patterns in spoken discourse are different from those in written discourse to some degree (Biber, Reference Biber2006) and that knowledge of items from written word lists may not always be sufficient for learners to deal with the lexical demands of spoken discourse (Dang et al., Reference Dang, Coxhead and Webb2021). The growth in the number of spoken and digital word lists also reflects a growing number of second language acquisition (SLA) studies showing that vocabulary is learned from exposure to aural input and audiovisual input (see Reynolds, Reference Reynolds2023). It is also the result of an increase in spoken and multimodal corpora, which makes it less challenging to develop word lists that are representative of these modes of communication.
Word list studies classified by modality (N = 50).
In terms of lexical components, although most word lists published in the last decade (68%) examined single words, two major changes occurred. First, there was an increase in the number of lists of multiword sequences (24%) and lists made up of both multiword sequences and individual words (8%) (see Fig. 4). This reflects a relatively recent trend in SLA research, which has emphasized the importance of knowledge of multiword sequences for second language (L2) communication (see Boers & Webb, Reference Boers and Webb2018). Second, while a reasonable percentage of lists of single words (23.68%) still employed level 6 word families (e.g., Greene & Coxhead, Reference Greene and Coxhead2015), none of the lists developed after 2020 chose this lexical unit. In fact, word lists using lemmas/flemmas (e.g., Dang & Webb, Reference Dang, Webb and Nation2016) or word types (e.g., Bancroft-Billings, Reference Bancroft-Billings2020) made up a slightly higher percentage of published lists (28.95% each). The level 6 word family counts a base form (e.g., assure), its inflected forms (e.g., assures, assured, assuring), and its derived forms up to Bauer and Nation’s (Reference Bauer and Nation1993) affixes (e.g., assurance, assuredly, reassurance, reassuring, reassuringly) as one lexical unit. The lemma/flemma counts a base form (e.g., assure) and its inflected forms (e.g., assures, assured, assuring) as one lexical unit. The lemma counts word forms of different parts of speech as separate units but the flemma counts them as one unit. The word type counts unrepeated word forms as separate units. Other new lexical units were used in one study each: level 3 partial word families in Nation (Reference Nation2016) and nuclear word families in Cobb and Laufer (Reference Cobb and Laufer2021). The level 3 partial word family includes level 2 flemmas and four derivational affixes from level 3 (un-, -ly, -th, and -er). The nuclear word family counts the base form of a level 6 word family (e.g., assure) and the inflected forms and derived form made up of core affixes (e.g., assurance, assured, reassuring). Moreover, since Gardner and Davies’s (Reference Gardner and Davies2014) Academic Vocabulary List, a growing number of studies (13.16%) have developed parallel versions of word lists which employ different lexical units. Also, 6.52% of the word lists (e.g., Coxhead & Demecheleer, Reference Coxhead and Demecheleer2018) used the word type as the primary lexical unit and then added any level 6 family members of that word type which has technical meanings to the list. This variation in the lexical unit of word lists reflects a need to use different units of counting words for different target users (see Webb, Reference Webb2021).
Word list studies classified by lexical component (N = 50).
Another trend revealed by Fig. 1 is that most studies (92%) developed comprehension-oriented word lists (i.e., lists to enhance learners’ comprehension of target communicative situations). However, since 2016, a small number of studies have developed knowledge-based word lists (8%) to represent the vocabulary that learners may already know. This indicates a pathway which may broaden word list research. While knowledge-based word lists may have value in revealing the words that are more likely to be known, the number of these lists is limited, and all of them have appeared recently. As a result, the impact of these lists is not clear yet. Given that most word lists published in the last decade are comprehension-oriented, the rest of this article will focus on the development, validation, and applications of comprehension-oriented lists.
3. How has the development and validation of comprehension-oriented word lists changed over the last decade?
3.1. Who are the target learners?
Identifying the target learners of word lists is important because it will influence the decision on all steps of word list construction, validation, and dissemination (Nation, Reference Nation2016). Although all word lists developed in the last decade provided information about their targeted learners, this information is relatively general. Of the seven general service lists, six briefly mentioned their learners’ language proficiency, educational levels, morphological knowledge, and/or first language (L1) backgrounds, and one (Browne, Reference Browne2013) provided very vague information (L2 learners of English). Likewise, of the nine academic word lists, four (e.g., Chon & Shin, Reference Chon and Shin2013) did not explicitly mention target learners and five (e.g., Ackermann & Chen, Reference Ackermann and Chen2013) indicated that their learners were university English for Academic Purposes (EAP) students but did not provide further information. Similarly, although all 30 technical word lists stated the disciplines of their target users, relatively few studies provided additional information. Studies mentioned learners’ educational levels (60%), the country where learners took the language courses or used the language for specific purposes (33%), the kind of English courses (13%), and learners’ prior vocabulary knowledge (3.33%). The modest information provided about target learners makes it difficult for teachers, learners, and policymakers to judge whether these lists are relevant to their context or not.
Our review also shows that little attention has been paid in word list research to learners from less privileged backgrounds, especially those who would like to use English for professional purposes. All academic word lists and 60% of the technical word lists aimed to support university students and scholars. Although some technical word lists have been developed for professionals (23.33%) and pre-university students (13.33%), they either focused on highly skilled professionals (business, trade, aviation, rugby) or pupils at international schools or middle/secondary schools in Global North countries (US, UK, Singapore, Germany). Therefore, research on academic and technical word lists seems to be biased towards the needs of privileged groups. This tendency could be because in the past these groups were the most likely to use English for work and education. However, with globalization, a growing number of people from less privileged groups (e.g., street vendors at tourist attractions in the Global South) need to use English at work and could benefit from word lists developed to serve their professional purposes.
3.2. How many items do word lists include?
While word lists should be long enough to cover key items that need to be known for comprehension, they should be short enough to fit learners’ limited learning time. However, most general service lists and academic word lists published in the last decade are relatively long. A total of 71.43% of general service lists (e.g., Brezina & Gablasova, Reference Brezina and Gablasova2015) and 44.44% of academic word lists (e.g., Ackermann & Chen, Reference Ackermann and Chen2013) had 2,000 to 9,000 items. Although none of the technical word lists have more than 2,000 items, 20% of them still have from 1,000 to 2,000 items (e.g., Coxhead & Demecheleer, Reference Coxhead and Demecheleer2018; Green & Lambert, Reference Green and Lambert2019). Several attempts have been made to break sizable lists into smaller sub-lists (e.g., Coxhead et al., Reference Coxhead, McLaughlin and Reid2019; Dang & Webb, Reference Dang, Webb and Nation2016) or filter items in existing lists so that they have sizes that are more manageable to a specific learner population (e.g., Cobb & Laufer, Reference Cobb and Laufer2021). Yet the large sizes of most existing lists may cause difficulty in implementing them into pedagogy.
3.3. What corpora have been used to develop word lists?
Word lists published in the last decade have benefited from the advances in corpus linguistics. In fact, all 46 comprehension-oriented lists were corpus-based. The availability of mega corpora (e.g., Davies’s (Reference Davies2008) Corpus of Contemporary American English (COCA); Nesi and Thompson’s (Reference Nesi and Thompson2006) British Academic Spoken English Corpus) and open resources (e.g., Massive Open Online Courses) have made developing large corpora for word list research less challenging than it used to be. Corpora used to develop lists of general service vocabulary, academic vocabulary, and technical vocabulary have respectively reached 12 billion words (Brezina & Gablasova, Reference Brezina and Gablasova2015), 288 million words (Browne et al., Reference Browne, Culligan and Phillips2013), and 18 million words (Greene & Coxhead, Reference Greene and Coxhead2015). These corpora were much larger than those used to develop West’s (Reference West1953) GSL (5 million words) and Coxhead’s (Reference Coxhead2000) AWL (3.5 million words), and larger or comparable to the size needed for reliable lists of the most frequent 9,000-word types suggested by Sorell (Reference Sorell2013).
Large corpora provide researchers with rich data so that they can better identify the lexical items that occur frequently in the target discourse. However, larger corpora do not necessarily ensure better word lists if the corpora do not represent the language that target learners are likely to encounter. Corpus representativeness includes situational representativeness and linguistic representativeness (Biber, Reference Biber1993; Egbert et al., Reference Egbert, Biber and Gray2022, cited in Miller, Reference Miller2022). Situational representativeness refers to the extent to which corpora reflect the kinds of text and topics in the target domain whereas linguistic representativeness refers to the extent to which corpora reflect the distribution of lexical patterns in these domains.
Situational representativeness was considered by 95.65% of the reviewed studies. However, the situational representativeness of corpora in some studies is questionable. For example, the corpora used to create lists of general service vocabulary often underrepresents spoken language. Half of these lists (e.g., Brezina & Gablasova, Reference Brezina and Gablasova2015) were developed from corpora made up of largely written texts. Moreover, corpora for word lists to help learners deal with spoken communication in professional settings are very small in size, ranging from 4,157 words (Drayton & Coxhead, Reference Drayton and Coxhead2023) to 133,093 words (Coxhead & Demecheleer, Reference Coxhead and Demecheleer2018). The modest proportion of spoken texts in corpora for word lists, especially those in professional settings, might be due to the challenge of collecting and analyzing spoken data. Yet the imbalance between spoken and written texts means that word lists developed from these corpora are biased towards written language and may not represent the lexical items that learners are likely to encounter frequently in speech.
Miller (Reference Miller2022) pointed out that compared to situational representativeness, linguistic representativeness has rarely been taken into account by word list researchers. Our review supports Miller’s point to some extent. Only 21.74% of the reviewed studies took linguistic representativeness into consideration. Except for Brezina and Gablasova (Reference Brezina and Gablasova2015), these studies often compared items of a newly developed list with those of former lists of similar purposes. As these lists were developed based on different selection criteria, such comparison does not provide direct evidence of how well the newly developed list represents the target discourse domain.
While previous word list studies have considered corpus representativeness to some extent, further care could be taken to improve corpus representativeness to reduce differences across lists of similar purposes. Sorell’s (Reference Sorell2013) study with general words and Miller’s (Reference Miller2022) study with specialized words provide some useful directions. Although focusing on different kinds of vocabulary, these studies employed similar methods; that is, they created lists from multiple similar-sized corpora which represented the same target discourse domains and then compared the overlaps among these lists to determine their stability. Both Sorell (Reference Sorell2013) and Miller (Reference Miller2022) found that the stability of word lists increased according to corpus size. This means that larger corpora are essential to increase the reliability of word lists. Sorell (Reference Sorell2013) also suggested that in addition to size, careful sampling of texts is also important. Creating word lists from 50 million-word corpora which represented texts carefully sampled from four major text types, Sorell (Reference Sorell2013) found that the variation for the most frequent 1,000 words of any text type was around 2%, and for 3,000 words, it was less than 5%. Even with the most frequent 9,000 words, the variation was just 4% to 7%. These findings indicate that a considerable level of reliability of word lists is possible with a large and well-designed corpus (see Nation & Sorell, Reference Nation, Sorell and Nation2016 for further information of how to design corpora for word list studies).
3.4. What criteria have been used to select items for word lists?
3.4.1. Frequency, range, and dispersion as selection criteria for lists of single words
In the last decade, frequency, range, and dispersion continue to be the most popular selection criteria for lists of single words, but they have been measured using more sophisticated indices. Frequency has been used in 61.76% of the lists of single words (21 lists). Figure 5 shows that while absolute frequency was still employed by ten lists (e.g., Coxhead & Demecheleer, Reference Coxhead and Demecheleer2018), since 2015 relative frequency has been increasingly used (e.g., Drayton & Coxhead, Reference Drayton and Coxhead2023). Absolute frequency refers to the total number of occurrences of a word in a text or a corpus, and thus is a useful index if we only examine a single text or corpus. However, to examine the occurrence of a word across corpora or texts of different sizes, relative frequency is a more useful index. Relative frequency is calculated by dividing the absolute frequency of a word by the size of a text or a corpus and multiplying by the basis for normalization (e.g., 100 words or one million words). As relative frequency takes corpus sizes into consideration, it helps to minimize the bias toward corpus size in frequency counts. However, relative frequency may be misleading in small corpora because it can overestimate the occurrence of rare words in the corpus, and thus should be reported together with absolute frequency (Brezina, Reference Brezina2018).
Indices of frequency in lists of single words.
Frequency ensures the selected items represent the words that occur frequently in the corpus. However, relying solely on frequency may lead to the risk of including items in the list that are extremely frequent in a small number of texts, but are rare in most of the texts in the corpora. Therefore, range and dispersion are also important criteria, collectively employed in 52.94% of lists of single words. Range refers to the number of texts in the corpus in which a lexical item occurs while dispersion refers to the distribution of that item throughout the corpus. Using both range and dispersion ensures that selected items occur in a range of texts and the frequency of the items in each text are relatively similar. Range plus dispersion has been increasingly used as selection criteria in lists of single words (see Fig. 6).
Number of lists of single words using range and dispersion (N = 18).
3.4.2. Specialization in lists of single words
Another change in the last decade is related to the approaches toward distinguishing specialized words from general service words in the development of lists of academic and technical words. Specialized word lists developed before 2013 often followed Coxhead’s approach by assuming that L2 learners already know general service words, and thus excluded general service words, represented by West’s GSL, from specialized lists. This approach was still adopted by 21.43% of the specialized word lists published post 2013 (e.g., Bancroft-Billings, Reference Bancroft-Billings2020). Excluding general service words from specialized word lists takes learners’ prior knowledge of general service vocabulary into account and minimizes the chance of teaching and learning known items. However, as the GSL was used as the general service word baseline, specialized word lists developed with this approach have been inevitably affected by the limitations of the GSL.
Several approaches have been taken to address this limitation. The first approach, employed by 10.71% of the specialized word lists, was to use updated lists of general service words. Yet the quality of specialized word lists still depends on the quality of lists used as the general service vocabulary baseline. For example, items in Browne’s (Reference Browne2013) New Academic Word List (NAWL) are those outside Browne et al.’s (Reference Browne, Culligan and Phillips2013) New General Service List (NGSL). However, little validation evidence of the NGSL is provided, which makes the validity of the NAWL questionable.
A second approach, employed by 53.57% of the specialized word lists (e.g., Gardner & Davies, Reference Gardner and Davies2014), is to include general service words if they meet the selection criteria. This approach enables these lists to avoid the limitations of any general service vocabulary baseline. Yet it does not consider learners’ prior knowledge of general vocabulary, which may result in inefficient use of learning time on known items.
Expanding on both approaches, recently a small number of word lists (14.29%) (e.g., Dang et al., Reference Dang, Coxhead and Webb2017) included all general service words that met the selection criteria, but then classified them into levels based on their frequency in general English. This allows users to focus on learning academic and technical words from levels that are beyond their current knowledge of general vocabulary and enables specialized word lists to avoid limitations caused by general service baselines while still accommodating users’ prior vocabulary knowledge.
3.4.3. Selection criteria for lists of multiword sequences
Studies developing lists of multiword sequences have widely employed frequency, a common selection criterion of single word lists, as their selection criterion. Frequency was used by 81.25% of lists of multiword sequences (13 lists), and most of these lists (11 lists) had relative frequency as the index (e.g., Ackermann & Chen, Reference Ackermann and Chen2013). However, other criteria which were commonly used in lists of single words were less common in the development of lists of multiword sequences. Range was employed in 50% of the lists (e.g., Lei & Liu, Reference Lei and Liu2018) and dispersion in only 18.75% (Green & Lambert, Reference Green and Lambert2019). This modest application of range and dispersion could be because these criteria may not always be relevant to multiword sequences. Combining range and dispersion enables researchers to identify items that occur frequently in a large number of texts. However, multiword sequences do not occur as frequently. If we use range and dispersion as additional criteria, the corpora for the analysis should be much larger than those used to develop lists of single words. It may be challenging to develop such corpora, especially if the focus is on spoken and specialized contexts.
Apart from frequency, range, and dispersion, strength of association has been used in the selection of multiword sequences. Strength of association refers to the likelihood that two word forms reoccur together more than by chance, and thus is a useful criterion to select multiword sequences. Despite its value, strength of association was used as a criterion by only 37.5% of the studies. Mutual information (MI) and/or t-score were used as indices of strength of association (e.g., Rogers et al., Reference Rogers, Müller, Daulton, Dickinson, Florescu, Reid and Stoeckel2021) despite Gablasova et al.’s (Reference Gablasova, Brezina and McEnery2017) suggestions that apart from MI and t-scores, there are a range of indices of strength of association and the most appropriate index varies depending on research purposes.
Taken together, our review suggests that compared to lists of single words, lists of multiword sequences could benefit from more sophisticated selection criteria specifically developed for multiword sequences.
3.4.4. Supplements to corpus-driven criteria
Corpus-driven criteria (e.g., frequency, range, dispersion) offer word list research robust tools to identify the lexical items that students need to know for effective communication. They also enable researchers to identify items for their lists in an objective and quick way, which allows other researchers to replicate their studies. However, word lists derived solely from corpora have limitations. Due to practicability (e.g., financial constraints, copyright and ethical restrictions), researchers may not be able to access relevant texts. This can have negative impacts on the representativeness of a corpus and in turn, word lists themselves. Moreover, only using corpus-driven criteria often results in overly long lists, making it challenging to incorporate a list into curricula (Dang & Webb, Reference Dang, Webb and Nation2016). Practitioners (e.g., Stein, Reference Stein2017) raise the concern that word lists developed from corpora of language use at large may not represent the experience of a learner in a specific context, potentially discouraging implementation of word lists in pedagogy.
Recognizing this concern, 52.18% of the 46 word lists have used information from other sources (dictionaries, concordance lines, experts) (e.g., Dang, Reference Dang2020; Lei & Liu, Reference Lei and Liu2016) to further filter items from corpus-driven word lists. Despite these efforts, it is important to note that 32.61% (e.g., Brezina & Gablasova, Reference Brezina and Gablasova2015) relied solely on corpus-driven criteria. Meanwhile, 15.22% mentioned using supplements to corpora-driven criteria, but did not describe these criteria and/or procedure in detail making replication impossible (e.g., Browne, Reference Browne2013). This lack of information makes it difficult for researchers and practitioners to judge the lists’ quality.
3.5. How have word lists been validated?
Validation is an essential stage of word list development. However, 43.48% of the 46 word list studies (e.g., Chon & Shin, Reference Chon and Shin2013) did not conduct validation, while one study (Browne, Reference Browne2013) mentioned having validated their list but did not report how the validation was done. This is problematic because the validity of the word lists is questionable and thus discourages use in research, language learning, and teaching.
All remaining 25 studies, which included validation, used lexical coverage in corpora as a validation criterion (e.g., Brezina & Gablasova, Reference Brezina and Gablasova2015). Lexical coverage refers to the proportion of words in a corpus covered by items from a word list, and thus is a useful criterion to validate word lists. SLA studies (e.g., Schmitt et al., Reference Schmitt, Jiang and Grabe2011; van Zeeland & Schmitt, Reference van Zeeland and Schmitt2013) have found that L2 learners’ comprehension of texts increases according to the proportion of known words in the text. Therefore, the higher the lexical coverage provided by a comprehension-oriented word list, the better the list may be. However, lexical coverage is only one factor determining the value of a pedagogical word list. Other factors (e.g., target users, corpus quality, selection criteria) also matter. Therefore, two recent studies (Drayton & Coxhead, Reference Drayton and Coxhead2023; Nation, Reference Nation2016) have used Nation’s (Reference Nation2016) framework to evaluate their lists. This framework is presented as a series of aspects that need consideration to make a word list suitable to a specific pedagogical context (see Appendix 2). The first aspect (purpose) focuses on the learners and kinds of vocabulary knowledge that the list targets. The next three aspects (unit of counting, main word lists, and other lists) cover the lexical units of the list and whether it is relevant to the target learners. The fifth and sixth aspects (corpus and making the lists) respectively evaluate the quality of the corpora used to develop the list and criteria used to select items for the list. The last two aspects (self-criticism and availability) consider whether list developers explicitly acknowledged the limitation of the lists and made it available for further evaluation.
Using Nation’s framework makes the procedure of construction and validation of Nation’s (Reference Nation2016) and Drayton and Coxhead’s (Reference Drayton and Coxhead2023) lists transparent to researchers and practitioners. However, because these researchers used their own judgment to evaluate their lists against Nation’s framework, their evaluation may not always reflect what teachers, learners, and policymakers think. To make word lists relevant to learning and teaching, word list developers should involve these stakeholders in the validation. Yet none of the 46 comprehension-oriented word lists included this validation step.
He and Godfroid (Reference He and Godfroid2018) and Dang et al. (Reference Dang, Webb and Coxhead2022) have included teachers and learners in the validation of word lists. He and Godfroid (Reference He and Godfroid2018) asked English as a Second Language (ESL) teachers to rate the usefulness and difficulty of items from a word list used in an EAP course and used this information to sequence items in the list in terms of priority for learning. Dang et al. (Reference Dang, Webb and Coxhead2022) used ESL/English as a Foreign Language (EFL) teachers’ judgment of word usefulness and EFL learners’ knowledge to compare the BNC/COCA2000 (Nation, Reference Nation2016) and the New General Service List (Brezina & Gablasova, Reference Brezina and Gablasova2015), the two general service lists which provided the highest lexical coverage among four established lists of general service words. He and Godfroid’s (Reference He and Godfroid2018) and Dang et al.’s (Reference Dang, Webb and Coxhead2022) findings showcase the value of involving stakeholders in word list validation to make word lists better matched to pedagogical contexts. Yet the limited involvement of stakeholders in the validation of published lists in the last decade may reduce the face validity of word lists in pedagogical contexts.
4. To what extent have published lists been applied in research on vocabulary learning, teaching, and assessment?
4.1. Applications of word lists for learning and teaching from word list developers’ perspectives
Nation and Coxhead’s (Reference Nation and Coxhead2021) research on native speaker vocabulary size and text coverage based on the BNC/COCA 25,000 word lists indicated that adult, educated native speakers of English tend to have a vocabulary of about 20,000 general word families and depending on their academic discipline, they may have additional knowledge of a few hundred to several thousand technical words. Targeting this vocabulary size may be a daunting task for many EFL learners given that they may learn only 400 word families per year (Webb & Chang, Reference Webb and Chang2012). However, lexical profiling studies using corpus-based word lists such as the BNC/COCA lists and their earlier version (BNC lists) have indicated that knowledge of the most frequent 2,000 to 3,000 general word families might be sufficient for L2 learners to comprehend a range of discourse types and for a higher degree of comprehension, a knowledge of 8,000 to 9,000 word families is likely to be needed. These findings indicate that the lexical demands of the tasks faced by L2 learners are more manageable than that suggested by the vocabulary size of native speakers. This finding is important because it demonstrates the value of corpus-based word lists in setting realistic learning goals for L2 learners. When combining this information with factors affecting learners’ vocabulary growth (Keuleers et al., Reference Keuleers, Stevens, Mandera and Brysbaert2015), it can inform the decision on what vocabulary should be prioritized in L2 pedagogy. Therefore, Nation (Reference Nation2016) has pointed out that word lists have several applications in pedagogy, including setting learning goals and informing the selection of target words, the design of activities and materials across the Four Strands, and the design of tests to assess learners’ vocabulary knowledge.
Of the 46 studies which developed comprehension-oriented word lists, 93.47% explicitly suggested possible applications of these lists in research and pedagogy while 6.52% did not. Table 1 shows that setting learning goals is the most frequently mentioned application (81.40%). It was followed by indicating target words for deliberate vocabulary learning activities and informing material design (62.79% each). Next came reference tools for learners and teachers (37.21%) and informing vocabulary assessment (30.23%). These applications are aligned with Nation’s (Reference Nation2016) suggestions about the applications of word lists for language learning and teaching.
Applications mentioned by the developers of the 46 comprehension-oriented word lists
While word list researchers can see many potential applications of recently developed lists in research and pedagogy, few lists have been used in lexical profiling and vocabulary testing, the two major research lines of studies of word lists. Similarly, little evidence has been found about the applications of these lists in pedagogy.
4.2. Applications of word lists in lexical profiling research
Published word lists can be used in lexical profiling studies to indicate the number of words associated with unassisted comprehension of texts, to identify the words that are likely to be unknown to learners, and to determine the extent to which words that are useful for learners reoccur in the texts. Findings of such analyses indicate the extent to which useful words are likely to be learned from the materials and inform the selection and adaptation of the materials so that they can better facilitate learning. Results largely depend on the quality of the word lists used for analyses. However, most lexical profiling studies published in the last decade (see Appendix 3) used either the BNC/COCA 25,000 lists (65.63%), its earlier version (BNC 14,000) (18.75%), or the AWL (9.38%). Only 12.5% have employed other lists, all of which were published before 2013 – EAP Science List (Coxhead & Hirsh, Reference Coxhead and Hirsh2007), Academic Formulas List (Simpson-Vlach & Ellis, Reference Simpson-Vlach and Ellis2010), and Phrasal Expressions List (Martínez & Schmitt, Reference Martínez and Schmitt2012) – and those published from 2013 onward – Academic Collocation List (ACL) (Ackermann & Chen, Reference Ackermann and Chen2013), Academic Vocabulary List (AVL) (Gardner & Davies, Reference Gardner and Davies2014), Academic Spoken Word List (Dang et al., Reference Dang, Coxhead and Webb2017), and Medical Spoken Word List (Dang, Reference Dang2020). Importantly, most of these lists were used by no more than one study.
Another application of published lists in lexical profiling research is to examine L2 learners’ free productive vocabulary knowledge in writing and speaking. Free productive vocabulary knowledge is often reported as lexical sophistication; that is, the percentage of advanced words produced by learners in spoken and written texts. Depending on specific groups of learners, advanced vocabulary can be operationalized differently. For example, multiword sequences could be considered as advanced vocabulary for EFL learners because many EFL learners have insufficient knowledge of these lexical items (Nguyen & Webb, Reference Nguyen and Webb2017). Technical vocabulary could be regarded as advanced vocabulary for learners studying English for academic, specific, or professional purposes because it is different from vocabulary in general communication to some extent (Coxhead, Reference Coxhead2018). Moreover, advanced vocabulary in spoken discourse may be different from that in written discourse (Dang et al., Reference Dang, Coxhead and Webb2021). Despite the different ways of operationalizing advanced vocabulary, lexical profiling studies do not seem to make good use of available lists to expand on earlier studies. Studies in the last ten years (see Appendix 4) still operationalize advanced vocabulary as items appearing in academic written word lists or those that are not general service word lists. Moreover, 62.5% of studies still used the AWL and the GSL to represent academic written words and general service words.
4.3. Applications of word lists in vocabulary testing
An essential step in constructing vocabulary tests is to identify a representative sample of words to be tested. The quality of these tests largely depends on the quality of the word lists from which the test items were sampled. Published word lists developed from large and representative samples of language in the target discourse are valuable sampling pools. If these lists are well-designed, it increases the validity of tests that sample items from the lists. If not, these tests cannot accurately measure knowledge of the targeted vocabulary.
Despite the large number of word lists published over the last decade, current vocabulary tests (see Appendix 5) only sample items from a very small number of word lists. The BNC/COCA 25,000 (Nation, Reference Nation2016) is the most commonly sourced list, employed by 54.55% of studies. It is followed by the AVL (Gardner & Davies, Reference Gardner and Davies2014) (adopted by 18.18% of the studies). Browne’s (Reference Browne2013) NGSL, Ackermann & Chen’s (Reference Ackermann and Chen2013) ACL, and Liu’s (Reference Liu2011) phrasal verb list were used by 9.09% each. Although a large number of technical word lists have been published in the last ten years, none have been used to develop vocabulary tests. Except for the Updated Vocabulary Levels Test (Webb et al., Reference Webb, Sasao and Balance2017) and the Vocabulary Size Test (Coxhead et al., Reference Coxhead, Nation and Sim2015; Nation & Coxhead, Reference Nation and Coxhead2021), both of which were developed from the BNC/COCA lists, the use of other newly developed tests in vocabulary research is limited. Therefore, it is fair to say that the impact of word lists on vocabulary testing is limited to the application of the BNC/COCA lists.
Taken together, while some attempts have been made to use more recent lists, most research on lexical profiling and vocabulary testing is still based on a very small number of lists (the GSL, the AWL, and the BNC/COCA 25,000). Little evidence is seen in the applications of other recently published lists. This suggests that despite what word list developers have suggested about the potential of recently developed lists, most have had limited practical value while one list (BNC/COCA 25,000) has had great value. The predominant use of the BNC/COCA 25,000 could be because it is the only list that has a large number of words at different frequency levels, and it is freely available on Paul Nation’s website and open access software such as RANGE (Heatley et al., Reference Heatley, Nation and Coxhead2002), Lextutor (Cobb, Reference Cobbn.d.), or AntwordProfiler (Anthony, Reference Anthonyn.d.). Meanwhile, the popularity of the AWL and the GSL could be because these lists are popular among teachers and learners (see Section 5) and are also available via various open access software.
5. To what extent have published lists been applied in pedagogical contexts?
Exploring the extent to which published lists have been applied in pedagogical contexts is important. If key stakeholders cannot see the value of word lists, they may be hesitant to implement them in teaching and learning. Consequently, word list research is unlikely to have an impact beyond the research community. Despite its importance, studies in this line are limited in number.
5.1. Key stakeholders’ perception and practice
To the best of our knowledge, only five studies have investigated teachers’ applications of word lists in real pedagogical contexts. Two are small in scale and only examined the implementation of word lists briefly. Dang and Webb (Reference Dang, Webb, Griffiths and Tajeddin2020) delivered a questionnaire to 16 Vietnamese EFL teachers whose teaching experience ranged from two to 24 years. Part of their questionnaire asked teachers to indicate resources for their vocabulary instruction. Word lists and their related applications were the three least popular: word lists (31.25%), research-based vocabulary tests (31.25%), and lexical profilers (12.5%). In contrast, textbooks were the most popular (87.5%). Coxhead et al. (Reference Coxhead, McLaughlin and Reid2019) interviewed a fabrication tutor on how he used a research-informed technical word list in his course. The tutor reported that he used the list as a reference tool. When preparing materials, he looked through items in the lists, selected words that were relevant to the topic of each session, and developed a topic-based glossary for his students to use during the course. The list also helped him design activities to enable students to actively learn words (e.g., writing words on the board, drawing their attention to the words).
Three large-scale studies have been conducted to survey teachers’ applications of word lists in pedagogical contexts: Burkett (Reference Burkett2015), Banister (Reference Banister2016), and Thompson and Alzeer (Reference Thompson and Alzeer2019) (see Appendix 6 for an overview of these studies). The studies used questionnaires and follow-up individual interviews to explore the perceptions of teachers, learners, and material and test designers regarding published lists. While Banister (Reference Banister2016) focused specifically on the AWL, the other studies examined perceptions of the AWL as well as other published lists. Participants in these studies were mainly working at tertiary institutions in Global North countries (US, UK, and Canada) and focused on teaching academic English.
All three studies found that teachers, learners, and material and test designers had positive opinions about published lists. Of Banister’s (Reference Banister2016) participants, 88% thought that the AWL was useful. Banister (Reference Banister2016) reported that the five most common reasons for positive attitudes were: “the AWL contains relevant vocabulary” (rated as either strongly agreed or agreed by 94.29% of the teachers), “this type of general academic vocabulary will be useful for students” (91.43%), “the AWL is based on corpus research not teacher judgment” (85.71%), “the AWL set clear vocabulary learning goals” (65.71%), and “the AWL is easy to incorporate into my lesson” (62.86%). Likewise, a considerable percentage of participants in Burkett’s (Reference Burkett2015) study and Thompson and Alzeer’s (Reference Thompson and Alzeer2019) indicated that word lists were useful (84.21% and 60.81%, respectively). Unfortunately, Burkett’s and Thompson and Alzeer’s participants did not specify which word lists they were referring to. Consequently, it is unclear whether their positive attitude toward word lists is toward a specific list or a range of lists.
While most participants had a positive attitude toward word lists, only 43% of Banister’s participants reported using the AWL. Likewise, 50.53% of Burkett’s (Reference Burkett2015) participants and 79.72% of Thompson and Alzeer’s (Reference Thompson and Alzeer2019) participants reported using published lists. Burkett (Reference Burkett2015) reported that, among lists, the AWL was the most popular, used by 48% of participants, while Thompson and Alzeer (Reference Thompson and Alzeer2019) reported that 35.85% of their participants used the AWL. A very small number of other word lists were also reported to be used (GSL, Oxford3000, BNC/COCA, GSL, New General Service List, and NAWL), but the percentages of participants using these lists were small (2.11% to 15.09%). It is important to note that, despite the large number of technical word lists and lists of multiword units published in the last decade, all word lists reported as being used by these teachers were made up of general service or single academic words.
Regarding how word lists were applied in pedagogy, both Burkett (Reference Burkett2015) and Thompson and Alzeer (Reference Thompson and Alzeer2019) found that language teaching was the most popular application (53.68%, 40.54%, respectively), followed by language learning (46.32%, 32.43%), developing materials (25.26%, 21.62%), and developing tests (21.05%, 21.62%). In addition to these applications, Thompson and Alzeer’s participants also reported using word lists in course design (16.22%) and for research purposes (16.22%). The applications reported by these teachers were consistent with what word list developers have proposed. However, in their surveys, Burkett (Reference Burkett2015) and Thompson and Alzeer (Reference Thompson and Alzeer2019) did not distinguish between published lists and in-house lists. Therefore, it is unclear from these findings to what extent published lists have been applied.
Banister’s (Reference Banister2016) interviews showed that teachers had several ways of implementing the AWL into their teaching: (a) using the AWL Highlighter to focus students on words that are worth learning, (b) providing language-focused learning activities for students to use AWL words (e.g., gap filling, word building activities, pronunciation, collocation exercises), and (c) instructing students to look at the percentage of AWL words in passages to select relevant texts for reading activities. The way that Banister’s interviewees implemented the AWL in their teaching was aligned with what word list researchers have suggested. However, it does not mean that all teachers followed such a principled approach. Two-thirds of Banister’s participants reported that they only introduced the AWL briefly and then let students use the list as a self-study tool. Similarly, half of Burkett’s (Reference Burkett2015) participants who reported using wordlists for self-study indicated that they used lists discreetly rather than with any supplementary materials. While giving learners the freedom to learn items from published lists in their own time helps learners to develop autonomy, to ensure effectiveness, teachers should provide careful training to students. Yet it is unclear from the survey results if the teachers trained their students and whether the students followed their instruction.
Burkett (Reference Burkett2015), Banister (Reference Banister2016), and Thompson and Alzeer (Reference Thompson and Alzeer2019) also revealed a number of concerns from teachers, learners, and course and material developers about word lists. Answers to open-ended questions and interviews indicated that most of these concerns were related to features of the word lists themselves. Concerns included the following:
• Having lists which are separate from what was taught in prescribed textbooks makes it difficult to incorporate the lists in meaning-focused activities.
• Before implementing published lists, teachers and learners, and course and material developers need to know where they can access lists and how lists were developed and validated.
• Some word lists have not gone through a rigorous validation process, which may negatively impact learning.
• Teachers, learners, and course and material developers would welcome supplementary materials to come with published lists.
• Most published lists lack rich information about the words (e.g., context in which the words are used and their definition).
• Most lists are too long to fit in a language course.
In addition to concerns related to the lists themselves, several participants in Banister’s (Reference Banister2016) study reported that word lists may be misused due to a lack of knowledge about how to incorporate them into learning programs. In fact, Burkett (Reference Burkett2015), Banister (Reference Banister2016), and Thompson and Alzeer (Reference Thompson and Alzeer2019) also showed that other common reasons why word lists were not used is that potential users were unaware of the existence of published lists, and they lacked experience and training in using word lists. This suggests that for published lists to be implemented in pedagogical contexts, it is important to increase awareness of lists that are freely available and provide greater information about lists so that they can be effectively implemented in teaching and learning.
Taken together, the three studies provide useful insights into stakeholders’ perspectives on word lists. However, although results suggested generally positive attitudes towards word lists, these findings may be representative of teachers who were interested in word list studies. Most participants in Burkett’s (Reference Burkett2015) and Banister’s (Reference Banister2016) studies were working in the US and the UK while Thompson and Alzeer’s participants were researchers. Thus, they may be more aware of published word lists than teachers, learners, and course and material developers working in low-resourced contexts. Moreover, given that the AWL was the most frequently used list, most opinions may relate primarily to the AWL. Perspectives on other published lists are less transparent.
5.2. Impact of local prescribed word lists on learning, teaching, and assessment
While surveys with teachers show that published lists appear to be implemented in pedagogical contexts to some extent, in many EFL contexts, what is covered in textbooks and tests is often affected by word lists prescribed by ministries of education. Two studies have examined the pedagogical impact of prescribed lists.
Jin et al. (Reference Jin, Li and Li2016) examined the extent to which items from the List of Basic English cover the vocabulary in the reading texts of the senior high school entrance exam, a compulsory exam students need to take at the end of their basic education stage. The List of Basic English was prescribed by the Ministry of Education of China to guide the design of materials in the basic education curriculum in China. Results showed that items from this list plus marginal words and proper nouns accounted for 92.82% of the total number of words in the reading texts. Jin et al. pointed out that this coverage is below 95%, the coverage cutoff points suggested by vocabulary research for reasonable comprehension of written texts. Therefore, they noted that there are still gaps between the vocabulary covered in the curriculum and that assessed in the test.
Reynolds et al. (Reference Reynolds, Shih and Wu2018) examined the effect of learning items from a reference word list developed by the College Entrance Examination Center in Taiwan on vocabulary acquisition and retention of first- and second-year university students in this context. In Taiwan, before entering university, students studied General English courses at the secondary school level. The design of materials and tests at secondary schools in Taiwan is guided by this reference list. Analysis of the participants’ scores on items in the Vocabulary Size Test showed that the words being tested in the VST and also occurring in the reference word list were more likely to be learned than those that did not occur in the list. The finding indicated that this prescribed list may have a significant impact on L2 learners’ vocabulary learning and retention in Taiwan.
Together, Jin et al.’s (Reference Jin, Li and Li2016) and Reynolds et al.’s (Reference Reynolds, Shih and Wu2018) findings suggest that prescribed lists developed by local authorities have been incorporated into the curriculum more effectively and have greater impacts on learning, teaching, and assessment than word lists developed by researchers. However, little information is known about how these lists were developed and to what extent they reflect the items needed for communication. It is important to investigate this overlap. Given the significant impact of prescribed lists on learning, if items in these lists were not carefully selected, it means that teaching and learning time will be wasted on words that are not helpful for future language use.
6. How can the impact of word list research be maximized in foreign language learning and teaching?
In the last decade, a large number of word lists have been published that can potentially inform language learning, teaching, and assessment. However, Coxhead’s (Reference Coxhead2000) AWL appears to be the only list to date that is recognized beyond the research community and which has had a significant impact on language learning pedagogy. Therefore, we believe that instead of putting efforts into developing new lists, researchers should focus more on exploring ways to promote existing word list research findings among key stakeholders and develop greater support for the implementation of word list use inside and outside language learning classrooms.
6.1. Further exploring the perceptions of key stakeholders
As a first step, it is essential to collect more evidence on the extent to which teachers and learners are aware of available word lists and their related research findings. It would be particularly useful to know the extent to which different word lists have been used in various contexts. Although a small number of studies have touched on these areas to some degree, they are based on self-reported data from a sample of convenient stakeholders, most of whom are teachers working at universities in the Global North and/or are interested in word lists, which may bias results. Moreover, no studies include the opinions of policymakers (e.g., those working at ministries of education), which may have the greatest impact on what is covered in curriculum, textbooks, and tests. This in turn affects teachers and learners’ selection of vocabulary for teaching and learning. As revealed from Burkett’s (Reference Burkett2015) surveys, one reason why his participants were hesitant to implement published lists in their contexts was the inconsistency between what was presented in these lists and what was covered in the textbooks. Therefore, if policymakers can see the value of published lists for students, they may incorporate these lists to inform the design of materials and tests for students. This will then create a positive test washback to encourage teachers and learners to teach and learn items from published lists.
6.2. Addressing the concerns of key stakeholders
While waiting for further evidence to be collected, several actions should be taken to address issues that were reported in earlier studies (Banister, Reference Banister2016; Burkett, Reference Burkett2015; Thompson & Alzeer, Reference Thompson and Alzeer2019) that discourage the implementation of published lists in pedagogy.
6.2.1. Accessibility
Teachers and learners need to not only be aware of existing lists and how these lists were developed and validated, but they also need to know where they can access these lists. Of the 46 comprehension-oriented word lists, 4.35% were not available (e.g., Liu & Han, Reference Liu and Han2015) and 63.04% are only available as article appendices (e.g., Chon & Shin, Reference Chon and Shin2013), which makes access for teachers and learners challenging.
Word list developers could make their lists and relevant guidelines available in open platforms such as the OSF (www.osf.io) and IRIS (www.iris-database.org) to improve accessibility. Furthermore, researchers can make word list research findings more accessible to practitioners via Open Accessible Summaries in Language Studies (OASIS) – https://oasis-database.org/. To save key stakeholders time from searching for information, it is also worth having a uniformed online archive for teachers and learners where word lists can be uploaded together with their guidelines.
6.2.2. Developing ready-made materials and providing CPD training to teachers
One reason for the popularity of the AWL is the availability of a large number of ready-made materials for learning AWL items. Of the 46 comprehension-oriented word lists, 65.22% are only available as a list of word forms. To use these lists for text analysis, teachers will need to convert them into text files and use them with specialized software such as RANGE (Heatley et al., Reference Heatley, Nation and Coxhead2002) or Antword Profiler (Reference AnthonyAnthony, n.d.). This complex process may discourage teachers and learners from using lists. While certain efforts have been made to create user-friendly platforms to analyze texts (e.g., Cobb’s, Reference Cobbn.d. Lextutor; Smith’s, Reference Smithn.d. EAP Foundation) and check the frequency level of a BNC/COCA word and its family member (Anthony’s [Reference Anthony2013] Word Family Finder), there would be value in having researchers work closely with publishers to codesign ready-made learning materials that help students to learn word list items (e.g., textbooks). There is also a need to offer teachers Continuing Professional Development (CPD) training opportunities that follow research-informed principles for using word lists in text analysis software to set learning goals, assess learning progress, and design learning materials and activities. Teachers can then reflect on and explore how to implement these ideas in their own contexts and evaluate the success of such interventions.
6.2.3. Developing word lists to accommodate needs
One reason why word lists may not be used is that teachers and learners may not believe most lists are relevant to their context. Therefore, word list researchers should be mindful when thinking about developing new lists. Researchers could develop close partnerships with key stakeholders to inform all steps of designing and implementing word lists into pedagogy. Such coproduction can help lists match the needs of target users and thus maximize their impact. If there are no clear needs, it is not worth putting effort into developing new lists. A more sustainable way to tackle concerns about the relevance of published lists to specific contexts is to provide key stakeholders with training so that they can develop Do-It-Yourself word lists to serve their own needs. Nation’s (Reference Nation2016) book is among the very first attempts to do so, but more actions should be taken.
6.2.4. Promoting learners’ autonomy in vocabulary learning
Most published lists present word forms without any contextual information. This makes it difficult to implement the use of lists into pedagogical contexts, because knowing a word involves not only learning its form. One useful way to address this concern is to give learners more autonomy in exploring the features of items in published lists. This can be done by engaging them in data-driven learning (DDL) activities to gain information about the selected words (e.g., meaning, collocations, part of speech). For example, concordances can be effective for learning collocations and the different senses of words.
7. Directions for future research
Several strands of research require further investigation. The first is exploring the impact of published lists on pedagogy. Research investigating how word lists are implemented in pedagogical contexts is clearly warranted. Such studies should include the participation of teachers, learners, materials designers, and ministry of education officials, and should employ more sophisticated research methods than questionnaires and individual interviews. Moreover, interventions that examine the effects of implementation of word lists in language programs are needed. Such studies may provide evidence on whether the claims of researchers about the value of word lists for learning and teaching hold true. Another avenue for further research is to examine the effects of CPD training activities and DDL activities on the implementation of word lists in teaching and learning. This might highlight optimal learning outcomes of word list informed pedagogy. Together, findings of studies in this strand may provide evidence of the value of published lists for L2 learning and teaching, which in turn may increase awareness of the use of word lists among the language learning community.
In addition to promoting findings, replicating word list studies for learners in underrepresented contexts is another useful direction for future research. Most published lists, especially academic and technical word lists, are in English and based on analysis of language produced by privileged groups of people. Developing lists for learners from disadvantaged backgrounds and in other languages would make word list research more inclusive. Moreover, a great deal has been learned from the many studies of English word lists that could be applied to the development of lists in other languages. While there have been some initiatives (e.g., Coxhead et al., Reference Coxhead, Parkinson and Tu’amoheloa2020; Coxhead & Tu’amoheloa, Reference Coxhead and Tu’amoheloa2019a, Reference Coxhead and Tu’amoheloab, Reference Coxhead and Tu’amoheloac, Reference Coxhead and Tu’amoheload; Finlayson et al., Reference Finlayson, Marsden and Hawkes2024; Jakobsen et al., Reference Jakobsen, Coxhead and Henriksen2018), for more word lists in other languages to be developed, it is crucial for researchers to develop large and representative corpora of other foreign languages and relevant corpus tools for researchers to analyze the vocabulary in these corpora.
Another important goal for word list developers is to critically think of more sustainable, low-cost ways to create word lists, which allow teachers and learners to create word lists to serve their own needs. While corpus linguistics offers an innovative way of creating word lists, creating corpora is time- and resource-consuming and developing corpus-based word lists requires certain levels of corpus literacy. Generative Artificial Intelligence (GenAI) from Large Language Models replicates natural language use at large, and thus could be a potential way to overcome these challenges.
7.1. Questions arising
1. What are key stakeholders’ perceptions about published lists?
2. To what extent have published lists been implemented in pedagogical contexts?
3. To what extent do published lists overlap with lists prescribed by local authorities?
4. What are the effects of a principled vocabulary learning program which implements word lists on L2 learners’ vocabulary knowledge and language proficiency?
5. What are the effects of CPD vocabulary training activities on teachers’ cognition and practice related to implementing published lists in pedagogy?
6. What are the effects of DDL/CALL activities that incorporate published lists on L2 learners’ vocabulary knowledge and language proficiency?
7. How can the findings of English word lists be applied to research on other foreign languages and underrepresented contexts?
8. How well can GenAI-generated word lists serve the needs of a specific context?
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0261444825000059.
Acknowledgements
We would like to thank Graeme Porte and the six anonymous reviewers for their constructive and valuable feedback, which has significantly enhanced the quality of our manuscript. Our thanks to Judith Hanks for sharing her experience in preparing a proposal for a state-of-the-art review.
Thi Ngoc Yen Dang is an Associate Professor in Language Education at the University of Leeds. She has taught English and Applied Linguistics/TESOL in Vietnam, New Zealand, and the UK. Her research interests include vocabulary studies and corpus linguistics, and she supervises MA and PhD students in these areas. She has developed a range of word lists for second/foreign language learners and examined the applications of these lists in various aspects of language learning, teaching, and assessment. Her articles have appeared in various journals (e.g., Applied Linguistics, Language Learning, TESOL Quarterly, Language Teaching Research, Studies in Second Language Acquisition). Email: T.N.Y.Dang@leeds.ac.uk
Stuart Webb is a Professor of Applied Linguistics at the University of Western Ontario. He currently teaches on the Masters in TESOL program and supervises students at the MA, PhD, and postdoctorate levels. His research interests include vocabulary studies, second language acquisition, and extensive reading, listening, and viewing. His articles have been published in journals such as Applied Linguistics and Language Learning. His latest books are The Routledge handbook of vocabulary studies, and How Vocabulary is Learned (with Paul Nation). Email: swebb27@uwo.ca
Open access
