Prescriptive Research Articles

A Comparative Study on ERNIE Bot 4.0 Turbo and ChatGPT 4o’s Performance in Evaluating First-Year Undergraduate Persuasive Essays

This study compares the performance of two prominent AI language models, ERNIE Bot 4.0 Turbo and ChatGPT 4o, in evaluating first-year undergraduate persuasive essays within the social sciences domain. Drawing from the Louvain Corpus of Native English Essays, a comprehensive collection of academic writings by British and American university students, this study aims to examine the models’ capabilities in assessing the grammatical correctness, vocabulary usage, coherence, content depth, and writing style of the essays. This study adopts a structured evaluation framework based on IELTS writing criteria to assess the models’ performance. A 40 persuasive essays from the Louvain Corpus were evaluated by both AI models and compared with human raters’ evaluations to ensure validity. The findings reveal distinct differences in the assessment styles of the two models. ChatGPT 4o exhibits a more critical approach, pinpointing areas for improvement, such as lack of argument development, coherence issues, and grammatical errors. Conversely, ERNIE Bot 4.0 Turbo offers a more balanced assessment, acknowledging essays’ strengths and suggesting improvement areas. Notably, ERNIE Bot’s evaluation highlights potential biases in AI-based assessment systems, particularly in its unequal emphasis on viewpoints. This comparative examination offers valuable perspectives on the advantages and constraints of AI models in assessing scholarly compositions, underscoring the significance of amalgamating varied AI functionalities to establish more all-inclusive and efficient feedback systems for learners. By understanding these differences, researchers and educators can better utilize AI-assisted essay evaluation systems to enhance student learning experiences.

AI versus human effectiveness in essay evaluation

The evaluation of student essay corrections has become a focal point in understanding the evolving role of Artificial Intelligence (AI) in education. This study aims to assess the accuracy, efficiency, and cost-effectiveness of ChatGPT's essay correction compared to human correction, with a primary focus on identifying and rectifying grammatical errors, spelling, sentence structure, punctuation, coherence, relevance, essay structure, and clarity. The research involves collecting essays from 100 randomly selected university students, covering diverse themes, with anonymity maintained and no prior corrections by humans or AI. An analysis sheet, outlining linguistic and informational elements for evaluation, serves as a benchmark for assessing the quality of corrections made by ChatGPT and humans. The study reveals that ChatGPT excels in fundamental language mechanics, demonstrating superior performance in areas like grammar, spelling, sentence structure, relevance, and supporting evidence. However, thematic consistency remains an area where human evaluators outperform the AI. The findings emphasize the potential for a balanced approach, leveraging both human and AI strengths, for a comprehensive and effective essay correction process.

Grammatical Error Correction Detection of English Conversational Pronunciation Under a Deep Learning Algorithm

Grammar correction in spoken English can enhance proficiency. This paper briefly introduces the gate recurrent unit (GRU) algorithm and its appli- cation in English speech recognition and grammatical error correction of speech recognition results. The GRU algorithm was firstly used to recognize English speech, then transform it into a text, and finally correct the English grammar of the text. Additionally, the attention mechanism was incorporated to enhance the performance of grammatical error correction. Subsequently, simulation experiments were conducted. Firstly, speech recognition and grammatical error correction were independently verified. The performance of the proposed algorithm in correcting grammatical errors in spoken English was evaluated using a self-built speech database. The results demonstrated that the proposed GRU-based algorithm yielded the best performance in independent speech recognition, independent grammatical error correction, and the overall spoken grammatical error correction. The contribution of this study lies in using the GRU algorithm to convert speech into text and perform grammar correction on the text, providing an effective reference for grammar correction in English communication.

Error Analysis on the Written English of Students in Grade 8 at Orr's Hill Vivekananda College, Trincomalee, Sri Lanka

This study attempts to identify and analyze “The errors in the written sentences produced by the students in grade 8 of Orr’s Hill Vivekananda School, Trincomalee”. The study begins with a systematic review of the concepts and theories concerning Error Analysis, which sets the platform for the theoretical base. Subsequently, the Error Analysis focuses on finding the syntactical errors under the selected grammatical categories while searching for the possible reasons causing errors. Eventually, the researcher suggests some of the complications involved in language learning and language teaching process. Understanding grammatical rules and applying them in sentences without making mistakes is what the teachers expect from the students in the school. However, it does not happen in the way it is expected. In other words, making grammatical errors can be assumed to be part and parcel of the language learning process. There are many linguistic aspects governing this learning process. Therefore, this research examines the type of errors the students make and why they make such errors in their writing. This dissertation is an eye-opener for teachers, with as they should understand the root causes of errors while attempting to find treatment for them. It is paramount that with teachers approach error corrections based on the student’s linguistic competence and their affective factors embedded with different socio-ethnic backgrounds. This error analysis has been undertaken with to with raise awareness among the teaching staff and help students produce better writing with fewer or no grammatical mistakes.

The Grammatical Errors in Translating Indonesian into English of the English Study Program Students

Penelitian ini dilakukan untuk menjawab pertanyaan-pertanyaan berikut: (1) Apa saja kesalahan yang paling sering dilakukan oleh mahasiswa Program Studi Bahasa Inggris dalam menerjemahkan teks bahasa Indonesia ke dalam bahasa Inggris? (2) Bagaimana tingkat kompetensi mahasiswa dalam menerjemahkan teks bahasa Indonesia ke dalam bahasa Inggris? Berdasarkan pernyataan masalah yang dikemukakan di atas, maka tujuan dari penelitian ini adalah sebagai berikut: (1) Mengidentifikasi kesalahan-kesalahan yang paling sering dilakukan oleh mahasiswa Program Studi Bahasa Inggris dalam menerjemahkan teks bahasa Indonesia ke dalam bahasa Inggris, dan (2) Mendeskripsikan tingkat kompetensi mahasiswa dalam menerjemahkan teks bahasa Indonesia ke dalam bahasa Inggris. Subjek penelitian ini adalah 20 orang mahasiswa. Dalam pengumpulan data, penulis menggunakan metode deskriptif. Metode ini bertujuan untuk mendeskripsikan kesalahan-kesalahan yang dilakukan oleh siswa. Penulis memberikan sebuah teks bacaan berbahasa Indonesia kepada siswa dan siswa menerjemahkannya ke dalam bahasa Inggris. Hasil dari penelitian ini adalah: (1) Jenis kesalahan tata bahasa dalam menerjemahkan teks bahasa Indonesia ke dalam bahasa Inggris yang dibuat oleh mahasiswa Program Studi Bahasa Inggris berdasarkan penghilangan, penambahan, kesalahan urutan dan kesalahan pembentukan. Kesalahan tata bahasa yang paling banyak ditemukan dalam menerjemahkan teks bahasa Indonesia ke dalam bahasa Inggris yang dibuat oleh mahasiswa adalah penghilangan sebanyak 98 kalimat. (2) Tingkat kompetensi, berada pada tingkat kurang. _________________________________________________________________________________ The study aimed to address the following research questions: (1) What are the most common errors made by students of the English Study Program when translating Indonesian texts into English? (2) What is the level of competence of these students in translating Indonesian texts into English? Based on these questions, the objectives of the study were: (1) To identify the most common errors made by English Study Program students in translating Indonesian texts into English, and (2) To describe the students’ levels of competence in translating Indonesian texts into English. The subjects of this study were 20 students. A qualitative method was used for data collection, focusing on describing the errors made by the students. The students were given Indonesian texts to translate into English. The results of the study are as follows: (1) Types of grammatical errors in translating Indonesian texts into English, made by the students, included omission, addition, mis-ordering, and mis-formation. The most common grammatical error was omission, with 98 sentences exhibiting this issue. (2) The overall level of competence among the students was assessed as poor.

Reducing Prejudice Through Narrative Writing: The Effects of Contact Type and Linguistic Authenticity

This study investigates how the content and style of narrative writing about an outgroup reduce prejudice. The study compares cooperative contact, where writers describe working with an outgroup character, with parasocial contact, where writers describe the character alone. It further examines whether the contact effects vary with linguistic authenticity. In two experiments, I manipulated the types of contact and measured authenticity through language patterns. Study 1 ( N = 719) found that having contact with an outgroup character elicited more positive affect toward the character, which reduced the desire for distance from the outgroup. Cooperative and parasocial contact showed similar effects, but their impacts varied with linguistic authenticity. Authentic language was associated with stronger impacts in cooperative contact but not in parasocial contact. Study 2 ( N = 287) replicated that linguistic authenticity moderated the effect of cooperative contact on affect toward the character. These findings highlight linguistic dynamics in intergroup contact.

SỬ DỤNG GOOGLE DOCS VÀ MERCI APP TRONG LỚP HỌC VIẾT TIẾNG PHÁP

This study compares the effectiveness of two automated feedback tools, Google Documents, which will be further addressed as Google Docs, and Merci App, in improving writing ability in French as a foreign language. Forty second-year students used these two reviewer tools to correct their B2-level texts. The data includes 80 student drafts edited on two tools and a survey questionnaire after using these two applications. The results show that both tools help students correct many errors at the sentence, word, and fragment levels but cannot correct errors at the inter-sentence, idea, and coherence levels. Google Docs detects more grammatical errors than Merci App, but Merci App is superior when it comes to finding errors in spelling, vocabulary, expression, and punctuation. Students use the application's error correction suggestions to correct their papers at a very high rate, 92.6% with Google Docs and 81.7% with Merci App. The number of students who prefer Google Docs is also higher than the number of students who prefer Merci App. The study proposes to use both tools to correct students' writing errors and to help students become familiar with error explanations in Merci App to help them systematize language knowledge and develop auto-learning capacity.

The Variations in the Use of Sentence Connectors in Timed and Untimed Student Writing: A Comparative Corpus Study of British Students

This study investigates the use of sentence connectors in British students' timed and untimed writings, utilizing data from the British component of the International Corpus of English (ICE-GB). By comparing the frequency and patterns of 54 selected connectors between these two writing conditions, the research reveals that students tend to overuse certain basic connectors like and and or, particularly in timed examinations. The analysis also highlights the impact of time pressure on linguistic choices, showing a preference for simpler connectors in timed settings. The findings contribute to understanding native English speakers' connector usage and suggest implications for academic writing and teaching strategies.

Dynamic Assessment-Based Curriculum Learning Method for Chinese Grammatical Error Correction

Dynamic Assessment-Based Curriculum Learning Method for Chinese Grammatical Error Correction

The use of the semi-lexical word ‘while’ by non-native (Polish) and native users of English based on corpora studies

The use of the semi-lexical word ‘while’ by non-native (Polish) and native users of English based on corpora studies

Exploring identity building, language transmission and educational strategies for immigrant d/Deaf multilingual learners

ABSTRACT This article, grounded in sociolinguistics, examines the identity building, language transmission and educational strategies of immigrant d/Deaf multilingual learners (IDML). The scientific literature mainly focuses on a single pair of languages – the national spoken language and the national signed one. For example, Deaf Studies traditionally analyse the situation of the White Deaf user of English and American Sign Language (ASL). A major change has recently occurred with the emergence of multilingual, multicultural and inclusive Deaf Studies. At the same time, professionals and members of the school community regularly have to face the increasing diversity of d/Deaf students’ profiles. Qualitative data were collected on an IDML cohort residing in and attending primary or secondary schools in France from families who use a spoken or signed language other than French or French Sign Language (LSF) at home. The main results reveal how the building of d/Deaf identity depends on linguistic and cultural transmission within families and to what extent the diversity of IDML is considered in educational strategies at school. The goal of the present paper is to stress the urgent need for this new and crucial research area on IDML to be developed in the French context and beyond.

Students' grammatical errors in essay writing

Students' grammatical errors in essay writing

Writing a successful applied linguistics conference abstract: The relationship between stylistic and linguistic features and raters’ evaluations

The conference abstract (CA) is a promotional genre that is vital to academic success. However, composing a high-quality abstract can be challenging for both first (L1) and second language (L2) writers. Some researchers have performed contrastive analyses of accepted/rejected CAs, yet few studies have analyzed the extent to which different stylistic and linguistic features can predict reviewers’ scores. The current study furthers this line of inquiry, while also responding to scholars’ recent calls for more replication work in the fields of applied linguistics and second language acquisition. Using a corpus of 304 abstracts from an applied linguistics conference, the current study is an approximate replication that analyzes the extent to which 32 variables are predictive of CA raters’ evaluations. Data analyses consisted of multiple stages, including examining the relationships between CA scores and nine stylistic variables (e.g., rhetorical moves, study completeness) and 23 linguistic variables (e.g., grammatical errors, and lexical and syntactic complexity measures). Statistically significant variables were then entered into a regression model. Results suggest that seven variables accounted for approximately 25 percent of CA scores. The pedagogical implications of these findings are discussed for L1/L2 writers, along with future research directions.

English Grammar Auto-Correction Robot based on Grammatical Error Generation Model

The traditional English grammar error correction system has problems such as poor error recognition precision and error correction success rate to be improved. Therefore, in this study, data augmentation techniques were used to transform and process error correcting English texts, and a rule-based and shallow neural network-based English text grammar correction model was constructed. The experimental results showed that the grammar error generation (GEG), rule-based (RB), classification-based (CB), and recurrent neural network (RNN) models achieved accuracy rates of 93.92%, 82.17%, 79.41%, and 88.09% in correcting grammar errors on 1702641 test sentences in the One Billion word corpus, respectively. The experimental results showed that the English grammar error correction model designed in this study had a strong error correction ability, but the computational efficiency was low. The research results significantly improved the accuracy and generalization ability of English grammar correction, optimized learning costs, and brought positive impacts to educational applications, providing strong support for the development of intelligent English grammar correction.

Developing advanced citation skills: A mixed-methods approach to corpus technology training for novice researchers

In contrast with most existing corpus technology (CT) studies which focus primarily on general vocabulary and grammar learning, this study investigates the effectiveness of CT in addressing challenges encountered by novice researchers in improving an advanced linguistic skill in EAP writing: using reporting verbs to express rhetorical functions in citations in literature reviews. An eight-week CT training programme featuring a practical framework of the rhetorical functions of citations and utilising the Parallel EAP Corpora was implemented for 50 Chinese postgraduate students who were L2 English users. A mixed-methods approach was adopted. Paired-samples t-tests demonstrated statistically significant improvement with medium effect sizes in the frequency and diversity of reporting verb usage in citations, as well as in the complexity of expressing rhetorical functions of citations. These findings were further corroborated by corpus-aided descriptive statistical and textual analyses. Questionnaire data and interviews further revealed students' positive evaluations towards the training informed by the Parallel EAP Corpora, which facilitated participants' learning in various ways. These findings hold substantial value for EAP pedagogy, indicating that the incorporation of pedagogically sound CT training can significantly improve novice researchers’ development of academic discourse through nuanced citation practices and advanced rhetorical strategies for composing literature reviews.

Narrative Retelling Abilities of Kindergarteners Considering Early Literacy and Language Skills

Objectives: This study investigated the narrative retelling abilities of 5-year-old children at risk of narrative development, early literacy, and language skills focusing on story grammar, linguistic cohesive devices, and semantic-grammatical index.Methods: Narrative Retelling production using KONA was conducted on a total of fifty kindergarten children, 10 typically developing children (TD), 10 at risk of developmental dyslexia (DD), 10 at risk of narrative discourse (ND), 10 at risk of narrative discourse and early literacy (ND·DD), and 10 at risk of early literacy and language as well as narrative discourse to compare the production of story grammars, linguistic cohesive markers, and use of linguistic indicators total number of words (NTW), number of grammatical morphemes, mean length of T-units in morpheme (MLT-m), grammatical morpheme error rate (GME) between groups.Results: ND, ND·DD and ND·DD·SLI groups showed lower performance in producing story grammars, linguistic cohesive markers, NTW, and number of grammatical morphemes compared to the TD and DD groups. There were no significant differences between groups in the number of inappropriate uses of linguistic cohesive devices, MLT-m, and grammatical morpheme error rate.Conclusion: The ND·DD·SLI group with poor semantics and grammar showed significantly lower narrative retelling ability. This suggests that early intervention may be beneficial for children at risk for difficulties in language and narrative development. The ND group with good semantics and grammar was identified with their low narrative retelling abilities. In addition to typical development in semantics and grammar, there were children at risk of narratives with low abilities in early literacy. In addition to language skills, we need to screen and identity narrative abilities considering early literacy skills during the kindergarten period.

Adverbial confirm in colloquial Singapore English: insights from a text message corpus

ABSTRACT Despite its innovative, salient, and robust usage in Colloquial Singapore English (CSE), the adverbial use of confirm has largely gone unnoticed. In this paper, we provide morphosyntactic and semantic evidence to show that adverbial confirm behaves like a strongly subjective speaker-oriented adverb (SpOA). We argue that adverbial confirm developed as a result of contact with the Mandarin adverbial kending, likely due to English–Mandarin bilingualism amongst Chinese Singaporeans, although its use has proliferated throughout the general CSE-speaking population. Quantitatively, adverbial confirm emerges as the most frequently used SpOA when compared to surface-semantically equivalent strongly subjective SpOAs, such as definitely, sure, for sure and surely. The results suggest an increase in preference by CSE speakers to use confirm to express strong speaker certainty.

What can NLP do for linguistics? Towards using grammatical error analysis to document non-standard English features

What can NLP do for linguistics? Towards using grammatical error analysis to document non-standard English features

Email invitations to publish: Academically sound (such as SAFP) versus potentially predatory journals

Background: Researchers increasingly receive invitations by email to publish. We analysed email publication invitations received by staff members of the Department of Biostatistics, University of the Free State (UFS), comparing emails relating to accredited and non-accredited journals.Methods: This cross-sectional study included all publication invitations received via UFS email accounts by staff members from May 2023 to July 2023. The researchers independently completed the data form, then checked and resolved any discrepancies.Results: Of the 93 distinct emails received from 88 journals, only 15 (16%) were received from a journal appearing on the Department of Higher Education and Training (DHET) accredited journal lists. These included South African Family Practice (SAFP) and the African Journal of Primary Health Care and Family Medicine (PHCFM). Emails from non-accredited journals were significantly (p 0.01) less likely to refer to a journal with a health sciences-related title (37% vs. 86%), indicate the publisher (36% vs. 93%), provide a link to the journal website (59% vs. 100%), state a full physical address (24% vs. 80%), refer to author instructions (21% vs. 47%) or request the recipient to share the email with colleagues (5% vs. 47%). Emails from non-accredited journals were significantly (p 0.01) more likely to contain grammatical errors (63% vs. 0%) and flattering remarks regarding the recipient or his or her research work (49% vs. 0%), and to indicate the journal’s International Standard Serial Number (ISSN) (67% vs. 13%).Conclusion: Clear differences were found between email invitations from accredited versus non-accredited journals.Contribution: The findings provide insight into warning signals in email publication invitations.

Grammatical Errors in Students’ Recount Text Writing (A Case Study in The Third Grade of SMAN 22 Jakarta)

This study is aimed to find out grammatical errors and the sources of errors in students’ recount texts. The subject of this study is the third-grade students of SMAN 22 Jakarta in the academic year 2023/2024 which consists of 23 students. Specifically, their recount texts focused on recounting their vacation activities. From the entire third-grade class of 23 students, 16 students were selected as the study sample using the purposive sampling technique. This study used a qualitative approach to analyze and identify grammatical errors. The analysis was carried out by referring to Dulay, Burt & Krashen’s theory, a surface strategy taxonomy consisting of Omission, Addition, Misformation, and Misordering and the sources of errors based on Ellis’s theory are Interlingual Transfer and Intralingual Transfer. The result of this study showed that the total number of grammatical errors made by the students in their recount texts is 32 errors. Which consist of 4 omission errors, 7 addition errors, 8 misordering erros, and 13 misinformation errors. Accordingly, the most frequent error is misinformation error which corresponds to 13 (40,6%) of the total error. Furthermore, this study also revealed that most of the errors are caused by Intralingual Transfer (30 errors or 94% of the total error). Based on these findings, it concluded that the students had some difficulties in applying English grammar rules in their writings. Therefore, it is suggested to be aware and pay more attention to improve the English teaching and learning process.