Learner Corpus Research Articles

A Corpus-Based Study: A Comparison between Native Speakers and Chinese EFL Learners in Using Adjectives in Argumentative Writings

The purpose of this study is to explore the differences in using adjectives between native speakers and Chinese EFL learners at beginner and intermediate levels, with the research data taken from three sub-written corpora of ICNALE, the written corpus of native speakers and the written corpus of Chinese EFL learners at A2 and B1 levels. The research data is coded by the UAM Corpus tool, version 3.2. The result is presented in two aspects, namely, the overall frequencies of the three sub-written corpora and the specific overused and underused wordings. The study reveals that compared with the native speakers, the Chinese EFL writers tend to employ adjectives that are vaguer and general in meaning. More importantly, the Chinese EFL learners’ lexical accuracy needs to be improved, as the result of the specific wordings shows that some adjectives are used inappropriately in the writing excerpts. However, in terms of lexical richness, the native speakers don’t show a significant variation in using adjectives, which differs from the findings of previous studies. The comparison between the beginner and intermediate Chinese EFL learners shows that the intermediate writers employ a wider range of higher leveled vocabulary, which is a major marker in many standard English writing tests. The study finally proposes two pedagogical implications for language teachers to improve EFL writers’ vocabulary variety as well as lexical accuracy, and the key point is to teach most essential academic adjectives under the syntactical and grammatical context.

言語間での転移学習のための事前学習モデルと多言語の学習者データを用いた文法誤り訂正

言語間での転移学習のための事前学習モデルと多言語の学習者データを用いた文法誤り訂正

The Varieties of English for Specific Purposes dAtabase (VESPA): Towards a multi-L1 and multi-register learner corpus of disciplinary writing

The Varieties of English for Specific Purposes dAtabase (VESPA first release) is the result of an international corpus compilation project that aims to address the lack of large-scale, open access, multi-L1, multi-discipline and multi-register learner corpora. This corpus report provides a detailed description of VESPA and illustrates possible uses of the corpus for register exploration of learner data. Specifically, it first offers an overview of the makeup of the corpus and the online interface that can be used to search and download the corpus. It then gives an illustrative example of a study where multi-dimensional analysis was used to investigate the relative importance of register vis-à-vis other factors in learner academic writing. In the concluding remarks, we identify priorities for future developments in the VESPA project, including the addition of more L1 components, more disciplines and more registers, as well as the compilation of a comparable corpus of native student writing.

The Methods of Annotating a Russian Learner Corpus

The Methods of Annotating a Russian Learner Corpus

Attribution in novice academic writing

Abstract Academic attribution, the direct acknowledgement of external sources, is investigated in two corpora of novice academic English, representing first and second language writing in linguistics. The forms and uses of attribution are analysed in a formal-functional framework. There is an overall underrepresentation of attribution in the learner corpus. However, the corpora have a similar proportional distribution of integral and non-integral attribution, but a difference in subtypes of these. Undated attributions are discussed as a special case. They occur in specific contexts, of which reference to course reading is peculiar to novice writing. Comparisons with expert corpora in Norwegian and English indicate that some, but not all, of the differences between the novice corpora may be linked to influence from the learners’ first language and culture.

A Comparative Corpus-based Study on the Use of Phrasal Verbs by Turkish EFL Learners and L1 English Speakers

This quantitative study aims to reveal the most frequently used phrasal verbs (PVs) by L1 speakers of English and Turkish EFL learners in written and spoken registers. With the purpose of spotting any overuse and/or underuse by Turkish EFL learners, it compares their usage to L1 English speakers’ through four corpora – two learner corpora and two native corpora. Additionally, the study investigates the most frequent adverbial verb particles (AVPs) and lexical verbs (LVs) in phrasal constructions comparing learners and native speakers. The results show that although LV types differ to a large extent, Turkish EFL learners display a similar profile to L1 English speakers in terms of types of PVs and AVPs. However, these verbs and particles are significantly underused, especially in spoken register – a result that contradicts previous research. Specifically, within the scope of this study, regardless of the register Turkish learners tend to favour few AVPs out of the mostly used ten while leaving out the others. The findings might be of use to EFL teachers regarding raising their students’ awareness on the contextual use of PVs in different registers with a combination of both implicit and explicit teaching in mind. In addition, learners can benefit from the ready-made PV lists to enhance their prospective usage in meaningful contexts.

텍스트 마이닝을 통한 한국어 학습자 사용 어휘 특성 연구 - 구어 말뭉치 분석을 중심으로 -

This study aims to examine a spoken corpus of Korean learners based on each learner's proficiency and nationality to identify the relational networks and main topics of core vocabulary and to confirm that Korean language education is being conducted appropriately in accordance with the grade goals suggested in the international Korean standard curriculum. To this end, core vocabulary was extracted using text mining techniques and data structures and common topics were found and then visualized and analyzed as relational networks. Text mining found that most Korean learners, other than advanced learners, use vocabulary according to proficiency and use vocabulary differently depending on nationality. These results demonstrate the need to provide vocabulary education to advanced learners according to their proficiency and show learners' purposes for learning Korean and values according to their nationality.

Accuracy order in L2 grammatical morphemes: Corpus evidence from different proficiency levels of Turkish learners of English

The present study empirically scrutinizes the fixed natural order of grammatical morphemes relying on a manual analysis of an EFL learner corpus. Specifically, we test whether the accuracy order of L2 grammatical morphemes in the case of L1 Turkish speakers of English deviates from Krashen’s (1977) natural order and whether proficiency levels play a role in the order of acquisition of these morphemes. With this in mind, we focus on the (in)accuracy of nine English grammatical morphemes with 2883 cases manually tagged by the UAM Corpus Tool in the written exam scripts of Turkish learners of English. The results based on target-like use scores provide evidence for deviation from what is widely believed to be a set order of acquisition of these grammatical morphemes by second language learners. In light of such findings, we challenge the view that the internally driven processes of mastering grammatical morphemes in English for interlanguage users are largely independent of their L1. Regardless of L2 grammar proficiency in our data, the observed accuracy of some morphemes ranked low in comparison with the so-called natural order. These grammatical morphemes were almost exclusively non-existent features in participants’ mother tongue (e.g., third person singular –s, articles and the irregular past tense forms), thus suggesting the influence of L1 in this respect.

Study on Epistemic Modality: Focusing on Learner Corpus and Native Speaker Corpus

The study aimed at investigating how Korean EFL writers employed epistemic modality in comparison with native speakers. For the study, the learner corpus (LC) and native speaker corpus (NSC) were created. The analysis focused on three categories: the overall frequency of epistemic modality, grammatical distributions, and semantic categories. The results of the study revealed that the overall frequency of epistemic modality in two corpora was similar, overlapping 6 top ten items. However, the close look at each frequency showed that Korean EFL writers employed a more restricted range of items, relying on the expressions of certainty and possibility. With regards to grammatical distributions, each group exhibited different ranges and frequencies of devices; while native speakers preferred adverbs, Korean EFL writers tended to favor modal verbs. Regarding semantic categories, Korean EFL writers preferred firmer assertion with the expression of certainty. Nevertheless, native speakers tended to favor more tentative expressions with the expressions of probability and doubt. Based on the results of the study, pedagogical implications were finally discussed.

Lexical Verbs in Verb-Noun Collocations: Empirical Evidence from a Malay ESL Learner Corpus

Previous studies have shown that verbs are indeed one of the essential lexical items that seem to dominate a text. At tertiary levels, learners are often presented with a list of verbs that are essential for English for Academic Purposes (EAP) courses such as academic writing. However, it is believed that the learning and teaching of these verbs goes far beyond the lists of EAP verbs. This is because knowledge of lexical verbs entails not only the knowledge of grammatical structures, but also the knowledge of semantic and syntactic structures. Therefore, this study aims to investigate how commonly used lexical verbs are used in learner writing in the perspectives of grammatical, semantic and syntactic features of verb-noun collocations. To achieve this, a corpus-based study using WordSmith Tools was employed on a learner corpus of Malay ESL learners to identify most-commonly used lexical verbs. These verbs were further given a linguistic analysis using a phraseological-based approach. The findings revealed that the use of transitive verbs led to several occurrences of verb-noun collocations and that the phraseological patterns of these verbs are bound by both lexical and grammatical patterns. This leads to the conclusion that the teaching of lexical verbs, specifically Verb-Noun collocations in academic writing should not only include the semantic elements, but also the syntactical elements of the verbs. It is then recommended that a lesson on lexical verbs should incorporate the teaching and learning of both lexis and grammar. Keywords: English for Academic Purpose; lexical verbs; phraseological-based approach; verb-noun collocations; learner corpus research

On Mean Dependency Distance as a Metric of Translation Quality Assessment

This paper has adopted a quantitative approach to carry out a linguistic study, within the theoretical framework of dependency grammar. Translation is a process where source language and target language interact with each other. The present study aims at exploring the feasibility of mean dependency distance as a metric for automated translation quality assessment. The current research hypothesized that different levels of translation are significantly different in the aspect of mean dependency distance. Data of this study were based on the written translation in Parallel Corpus of Chinese EFL Learners which was composed of translations from Chinese EFL learners in various topic. The translations were human-scored to determine the levels of translation, according to which the translations were categorized. Our results indicated that: (1) senior students perform better in translation than junior students, and mean dependency distance of translations from senior group is significantly shorter than the junior; (2) high quality translations yield shorter mean dependency distance than the low quality translations; (3) mean dependency distance of translations is moderately correlated with the human score. The resultant implication suggests the potential for mean dependency distance in differentiating translations of different quality.

Corpus analysis of the violations of gender coordination in the Russian speech of Kazakh students

The article is devoted to the description of violations of gender coordination between subject-nounsand different types of predicates found in Russian texts of Kazakh students. The purpose of the article isto identify typical deviations and establish the factors that cause them. The source of the material was theKazakh subcorpus of the Russian Learner Corpus (www.web-corpora.net/RLC). Examples of coordinationdisorders were extracted using the AgrGender tag, which marks various facts of violation of gendercoordination in the corpus, then, in accordance with the tasks set, these examples were classified, selectedand statistically processed. It is argued that violations of gender coordination are not accidental:they are explained by the complexity of mastering the strategy of coordination in effect in the Russianlanguage, and the formal nature of the category of grammatical gender. The article describes cases ofviolation of coordination of subjects with verb and nominal parts of predicates under the influence ofinanimate nouns, expansion of the strategy of semantic agreement, incorrect choice of the agreementcontroller, distant location of subjects and predicates, etc. In general, violations of coordination differfrom violations of agreement in attributive phrases. A corpus analysis of gender coordination disordersin Russian texts of Kazakh students can be interesting both from the point of bilingualism view researchand from the standpoint of teaching Russian as a non-native language.Key words: Russian, Russian Learner Corpus, gender coordination, Kazakh-Russian bilingualism,written text.

Lexical Bundles in Students' Argumentative Essays: A Study of Learner Corpus

This study investigates the structures of 3- to 5-word bundles used by EFL learners at the university level in writing argumentative essays. This qualitative corpus-based study focuses on answering the structural category and frequency of lexical bundles in students’ essays. The data of this study are sentences containing lexical bundles in the students’ essays, and the data were collected from students' essays compiled as a learner corpus. The lexical bundles were identified from the corpus with the assistance of a corpus tool, LancsBox, using the n-gram feature. This study used ten occurrences as the cutoff frequency and Gries’ DP as the dispersion threshold to identify the lexical bundles. The bundles were then classified into structural taxonomies, and the frequency of use of the lexical bundles was also investigated to complete the analysis. Academic Formulas List comprising bundles commonly used in the academic context was used to validate the bundles. The results show enormous structures of NP-based, VP-based, PP-based, and others identified in the learner corpus with NP-based bundles as the most frequent bundles and ‘the use of’ as the most frequent individual bundle. However, the variants of the bundles in the learner corpus are still dominated by fixed frames. In addition, apart from the shared bundles between the corpus and Academic Formulas List, there are some discrepancies related to the registers. It indicates that writing courses and writing materials should provide learners with more variants of lexical bundles and the appropriate context of use.

Evidence for the interdependence hypothesis: a longitudinal study of biliteracy development in a CLIL/bilingual setting

ABSTRACT However influential the interdependence hypothesis has become in bilingual research, it still lacks full empirical support. This longitudinal study explores the parallels in the biliteracy development (L1 Spanish and L2 English) of 20 students in a European immersion programme (i.e. CLIL) over a two-year period. A bilingual learner corpus of history narratives, based on history curriculum content, was collected during ninth and tenth grade. These essays were processed with MultiAzterTest, a state-of-the-art language analysis tool, and a Pearson correlation analysis was conducted to determine if any dimensions evolved in unison in both languages. The results show that some dimensions – length measures, nominalisation, subordination and lexical development – evolve in a similar fashion, thus supporting the interdependence and the common underlying proficiency hypotheses. Additionally, the results of a mixed-model analysis confirm that the fixed effect of time and language on such progress is significant, unlike the random effects introduced by the students.

Use of Linguistic Complexity in Writing Among Chinese EFL Learners in High-Stakes Tests: Insights From a Corpus of TOEFL iBT.

In studies on second language writing, linguistic complexity exhibited by learners has long been regarded as being indicative of writing proficiency. However, there are relatively scant studies focusing on the diversity and structural elaboration of complexity in L2 production data that are extracted from high-stakes tests [such as Test of English as a Foreign Language (TOEFL) and International English Language Testing System (IELTS)]. Using a large-scale learner corpus collected from a TOEFL (internet-based test (iBT), this study aims to explore the extent to which the three dimensions of linguistic complexity, syntactic, lexical, and morphological complexity, are associated with human scoring in high-stakes tests. In addition, we also tend to tap into within-genre topic effects on the production of complexity measures by learners. To this end, a total of 1,002 writing samples were collected from a TOEFL11 corpus, and six automated-coding instruments were used to investigate the variations of complexity among Chinese English as a Foreign Language (EFL) learners. The results from the correlation analysis, multiple linear regression, and independent sample t-tests indicated that there was not a linear correlation between the majority of linguistic complexity and human-rated score levels and that proficiency among Chinese EFL learners did not signal a discriminative power in their language production. In the meantime, strong within-proficiency topic effects were found on the majority of measures in the syntactic, lexical, and morphological dimensions.

A New Approach to Automatic Detection and Correction of Derivational Errors in L2 Russian

Learner corpora serve as one of the most valuable sources of statistical data on learners' errors. For instance, data from foreign-language learners’ corpora can be used for the Second Language Acquisition research. However, corpora representativity strongly depends on the quality of its error markup, which is most frequently carried out manually and thus presents a time-consuming and painstaking routine for the annotators. To make annotation process easier, additional tools, such as spellcheckers, are usually used. This paper focuses on developing a program for automatic correction of derivational errors made by learners of Russian as a foreign language. Derivational errors, which are not common for adult Russian native speakers (L1), but occur quite often in written texts or speech of Russian as foreign language learners (L2) [Chernigovskaya, Gor, 2000], were chosen as scope of our research because correction of such mistakes presents a formidable challenge for existing spellcheckers. Using the data from the Russian Learner Corpus (http://www.web-corpora.net/RLC/), we tested two already existing approaches to solve such kind of problems. The first one is based on a finite state automaton principle developed by Dickinson and Herring 2008, and it was test-ed as algorithm for derivational errors detection. The second one which relies on the Noisy Channel model by Brill and Moore, 2000, was used for studying errors correction. After we analyzed effectiveness of these tests, we developed our own system for autocorrection of derivational errors. In our program the algorithm of Dickinson and Herring was used as word-formation error detection module. The Noisy Channel model has been rejected, and we decided to use instead the Continuous Bag of Words FastText model, based on Harris distributional semantics theory [1954]. In addition, filtering rules have been developed for correcting frequent errors that the model is unable to handle. To restore automatically the correct grammatical word form, dictionary of word paradigms is used. Model results were validated on the data of Russian Learner Corpus.

Corpora and L2 acquisition: the L1 Portuguese – L2 Spanish subcorpus of CEDEL2

This paper presents the L1 Portuguese – L2 Spanish subcorpus of Corpus Escrito del Español L2 (CEDEL2), a new methodological resource for second language acquisition (SLA) research, which is freely searchable and downloadable (http://cedel2.learnercorpora.com). CEDEL2 is a large-scale, multi-L1 learner corpus of L2 Spanish which contains written productions from learners at all proficiency levels as well as 6 native control subcorpora (total size: over 1,100,000 words from over 4,000 participants). CEDEL2 follows strict corpus design criteria (Sinclair, 2005) and learner corpus design recommendations (Tracy-Ventura & Paquot, 2021a). In its current version (CEDEL2 v. 2), its Portuguese component includes an L1 Portuguese – L2 Spanish subcorpus, with 21,662 words written by 164 participants, and an L1 Portuguese native subcorpus, with 3,500 words from 16 L1 speakers of European Portuguese. Thanks to their design features (e.g., same design across subcorpora, inclusion of metadata about SLA-relevant variables, dual native control subcorpora) and freely available web interface, CEDEL2 and its Portuguese subcorpora allow researchers to investigate a wide range of topics in SLA.

CEDEL2: Design, compilation and web interface of an online corpus for L2 Spanish acquisition research

This article presents and reviews a new methodological resource for research in second language acquisition (SLA), CEDEL2 ( Corpus Escrito del Español L2 ‘L2 Spanish Written Corpus’), and its free online search-engine interface ( cedel2.learnercorpora.com ). CEDEL2 is a multi-first-language corpus (Spanish, English, German, Dutch, Portuguese, Italian, French, Greek, Russian, Japanese, Chinese, and Arabic) of L2 Spanish learners at all proficiency levels. It additionally contains several native control subcorpora (English, Portuguese, Greek, Japanese, and Arabic). Its latest release (version 2) holds material from around 4,400 speakers, which amounts to over 1,100,000 words. CEDEL2 follows strict corpus-design criteria (Sinclair, 2005) and L2 corpus-design recommendations (Tracy-Ventura and Paquot, 2021), and all subcorpora are equally designed to be fully contrastable, as recommended by Contrastive Interlanguage Analysis (Granger, 2015). Thanks to its design and web interface, CEDEL2 allows for complex searches which can be further narrowed down according to its SLA-motivated variables, e.g. first language (L1), proficiency level, self-reported proficiency level, age of onset to the L2, length of exposure to the L2, length of residence in a Spanish-speaking country, knowledge of other foreign languages, type of task, etc. These CEDEL2 features allow L2 researchers to address SLA questions and hypotheses.

The relationship between topic and metaphor in second-language learners’ essays

AbstractIn this article I investigate to what extent the use of metaphorical expressions in language learners’ texts vary according to the topic they have chosen to write about. The data come from the Norwegian learner corpus ASK, where the texts are from written assignments produced by adult second-language learners as part of an official Norwegian test and texts. Texts from two different prompts are selected, which are related to friendship and nature. Metaphors are defined according to conceptual metaphor theory and a triangulation of methods is used, alternating between a manual and an automatic extraction method.The results confirm the hypothesis that the two different prompts given to the learners in a language test not only triggers different metaphorical expressions but also influences the amount of metaphor used in the learners’ writing. This knowledge is important to researchers for comparing the use of metaphors between different groups, such as between different learners or between students in different stages of education. It is also important for test designers who decide on topics to be used in tests and teachers who help learners prepare for their tests. In addition, it is of interest for researchers, educators in general and the learners themselves who are interested in the effect the use of metaphors in texts have on raters’ evaluations in high-stake tests.

Review of Götz & Mukherjee (2019): Learner Corpora and Language Teaching

Review of Götz & Mukherjee (2019): Learner Corpora and Language Teaching