INTRODUCTION
Introduction
Mastering the English language involves more than just understanding its vocabulary and grammar; it also requires the ability to use idiomatic expressions effectively The correct application of collocations is crucial for EFL/ESL learners, as it enhances fluency, improves comprehensibility, and leads to more native-like speech This aspect of language acquisition plays a vital role in Second Language Acquisition (SLA), as highlighted by Sinclair (1991), Nation (2001), and Durrant & Schmitt (2010) For instance, English speakers commonly use the phrase "the quality of life" rather than "life quality," and the verb "do" is typically paired with "yoga" or "karate." Understanding collocations significantly contributes to speaking more naturally in English.
Learning multi-word expressions like "play" or "go" is essential for mastering idiomatic English However, foreign language learners often struggle to combine words in a way that sounds natural to native speakers, as acquiring collocations can be quite challenging Therefore, more research is needed to explore the underlying reasons why non-native students frequently produce incorrect collocations, how to address this issue, and what strategies can be implemented to enhance their ability to create natural-sounding phrases.
Non-native speakers frequently struggle with English collocations, a challenge even advanced learners face (Youmei & Yun, 2005) Stubbs (1999) highlighted collocations as a particularly difficult aspect of language learning Research by Thongvitit and Thumawongsa (2017) indicates that the understanding of collocations, or the instinctive knowledge of word combinations, is often intuitive for native speakers and can seem arbitrary to non-native speakers Consequently, numerous studies have explored the factors contributing to collocational errors, with first language interference being a primary cause, where individuals apply their native language knowledge to a second language (Zhang).
Non-native speakers often translate ideas from their mother tongue into English, which can lead to unnatural or non-existent expressions due to negative transfer Additionally, overgeneralization is another common reason for these language challenges.
Speakers often apply learned concepts to a single instance and subsequently generalize to similar situations Furthermore, a lack of awareness regarding rule restrictions and approximations in communication strategies significantly influences learners' ability to produce appropriate collocations in English.
Numerous corpus-based studies have focused on various aspects of collocations, including their frequency (Shitu, 2015; Thongvitit & Thumawongsa, 2017), the relationship between collocation production and students' English proficiency (Chen, 2004), the impact of the first language on second language collocation production (Youmei & Yun, 2005), the analysis of collocational patterns (Chen, 2004; Shitu, 2015), and the pedagogical implications for teaching English collocations (Shin and Chon, 2014) However, there is a lack of research on the distribution of different collocational subtypes, particularly regarding which patterns are most and least prevalent Additionally, the appropriateness of collocations in students' essays has not received sufficient attention, nor has the relationship between the use of idiomatic collocations and essay scores been thoroughly explored.
Previous studies have highlighted various aspects of collocational errors in student essays, yet they often overlook critical elements For instance, the 2004 study primarily examined the percentage of lexical collocation errors without detailing the distribution of collocational patterns in student writing Similarly, Thongvitit and Thumawongsa (2017) focused on the types and frequency of English collocations in research article abstracts but failed to assess their appropriateness Furthermore, while Youmei and Yun (2005) investigated the problematic combinations and the influence of learners' first languages on second language collocations, they did not explore how these collocational errors affect overall essay scores.
This study aims to address unresolved issues related to collocational patterns in EFL writing by analyzing English essays written by Vietnamese EFL learners It will examine the frequency of various collocational patterns and assess the appropriateness of these collocations in students' work Furthermore, the research will investigate the impact of collocation appropriateness on essay scores, providing insights into how these patterns affect overall writing quality.
Aims of the research
This study aims to explore the collocational usage of EFL learners in their argumentative essays and to examine how this usage may affect their writing performance in English as a Foreign Language (EFL).
This research aims to measure the frequency of various collocation subtypes in students' essays, assess the appropriateness of these collocational patterns, and explore the impact of collocations on the overall grading of the essays.
Research questions
To achieve the research aims, this study will be designed to seek the answers to the following questions:
1 What is the proportion of different subtypes of lexical collocations in the essays of EFL writers?
2 To what extent the lexical collocations found in the essays of EFL writers are considered appropriate?
3 Is there any correlation between the essay scores and the number of appropriate lexical collocations produced in those essays?
Research significance
The idea of analyzing the collocational patterns via a compiled collection of student‟s written academic assignments at a university in Central Vietnam would bring about some significant results
Learner corpus data of collocations provides valuable insights into students' collocational and non-collocational patterns By utilizing this data, students can become more aware of collocational inappropriateness, leading to greater care in their writing Additionally, understanding commonly occurring collocational patterns in their essays can help students focus on improving their weaknesses, particularly in addressing frequent erroneous collocations.
This study's findings provide valuable insights for tertiary-level teachers focusing on students' collocation usage By examining learners' collocation repertoire, educators can enhance their writing lessons, effectively correct academic writing, and provide constructive feedback Understanding the distribution of collocational subtypes allows teachers to identify popular and less common patterns, as well as common collocational errors made by students Consequently, both researchers and educators can gain a comprehensive understanding of students' collocational knowledge, enabling them to tailor teaching materials and lesson content to improve students' intuition and proficiency in using appropriate collocations.
Finally, the data collection of collocations extracted from students‟ essays would be a valuable and authentic source that could be utilized or recycled for other purposes of corpus-based research
Recent studies on collocational patterns indicate a need for further investigation in this area to expand the research landscape The findings could serve as credible evidence for related studies and provide valuable resources for material developers, textbook authors, and academic journal writers focusing on collocation Ultimately, this research aims to offer suggestions for enhancing the teaching and learning of English collocation.
Scope of the research
This study investigates the use of collocations by EFL students at a university in Central Vietnam, specifically analyzing compositions from second-year and third-year students during the 2018-2019 and 2019-2020 academic years The research is centered on lexical collocations and their seven subtypes, which will be detailed in Chapter 3.
Thesis structure
The paper consists of five chapters in total
Chapter 1 is the introduction which has set the scene for the thesis This part includes six sections, namely introduction of the research, aims of the research, research questions, research significance, the scope of the research, and thesis structure
This chapter consists of two key sections: the theoretical background of collocations and an examination of corpus-based studies on collocational errors The primary objective is to critically analyze relevant published works and identify gaps that have been overlooked in previous studies Subsequently, it will be demonstrated how the current research addresses these gaps.
This methodology outlines the processes for collecting, analyzing, and illustrating data to address three key research questions It introduces the study's subject, describes the research instrument, details the construction of a non-native corpus, and explains the data collection procedure.
This chapter addresses three key research questions, presenting the findings through tables, bar charts, and pie charts The results will be explained and compared to similar and differing studies.
The final chapter summarizes the thesis and outlines conclusions based on the study's findings It discusses the limitations of the research and offers recommendations for students, teachers, and future researchers Additionally, it presents suggestions for further studies.
LITERATURE REVIEW
Theoretical background of collocational studies
Collocations are defined by various researchers and linguists through different perspectives (Firth, 1957; Nesselhauf, 2003; Sinclair, 1991) They can be understood as the arrangement of words together (Nesselhauf, 2003; Shitu, 2015), rather than merely the co-occurrence of words within a specific range (Benson, 1985) Some scholars describe collocation as "close word combinations in a sentence" (Mongkolchai, n.d., as cited in Shitu, 2015) or as "items that occur physically together or have a strong chance of being mentioned together" (Sinclair, 1991) Additionally, collocations are often seen as "predictable combinations of words" (Hill, 2000, as cited in Shitu, 2015).
Collocation is defined as a "group of words that occur repeatedly" (Beson, 1985, as cited in Thongvitit and Thumawongsa, 2017) or as "the readily observable phenomenon whereby certain words co-occur in natural text with greater than random frequency" (Lewis, 1997, p.8) Researchers have examined collocation through the lens of semantic properties, identifying three types of collocational restrictions according to Palmer (1981): those based entirely on meaning, those based on range where a word pairs with others sharing semantic features, and strict collocational restrictions that do not involve meaning or range, such as "addled with eggs and brains" (as cited in Youmei & Yun, 2005).
There are six basic features of collocations that distinguish them from other expressions (Alsulayyi, 2015; Salman, 2017; and Mohammed, 2019)
Collocations are combinations of words that frequently occur together and can be categorized as either flexible or fixed Flexible collocations allow for the insertion of descriptive modifiers, such as transforming "make an effort" into "make a great effort." In contrast, some collocations are fixed and cannot be interrupted by additional words, maintaining their original form.
Certain collocations in English, such as "brain drain," "rancid butter," and "take pride in," allow for some flexibility in their structure, as noted by Salman (2017) However, other combinations, including verb + adverb, adjective + adverb, noun + noun, adjective + preposition, and noun + preposition, remain fixed and inflexible, demonstrating the complexity of language use.
Collocations in English are arbitrary and lack clear explanations for why certain words pair together while others do not For instance, phrases like "throw trash" are more commonly used than "throw rubbish," and "great happiness" feels more natural than "big happiness." This phenomenon arises from habitual usage rather than specific rules governing word combinations Consequently, the unpredictable nature of collocations can often perplex non-native speakers (Salman, 2017).
Collocations exhibit both predictable and unpredictable phenomena Predictability arises when the mention of one word prompts listeners to anticipate its common collocates; for example, the word "shrug" often leads to the immediate association with "shoulder," while "shake" may evoke "hand" or "head." This expectation highlights how certain word combinations are inherently linked (Salman, 2017) Conversely, unpredictability occurs when a word's affinity for a specific collocate defies logical expectations, as some synonymous words do not consistently pair with their expected counterparts (Mohammed, 2019) Salman noted that collocations can be unpredictable, emphasizing that logic does not always guide our assumptions about synonymous terms.
“dense population” but not “thick population” On the contrary, people often say
Collocations like "thick hair" cannot be substituted with synonyms like "dense hair," highlighting their unique components According to Salman (2017), the non-generalizability and unpredictability of collocations primarily dictate their usage This complexity makes collocations challenging for non-native speakers to master due to their predictable and unpredictable nature.
Collocations refer to fixed word combinations where the order of the components is largely irreversible, such as "ups and downs," "pros and cons," and "give and take."
Collocations often carry metaphorical meanings, with some phrases being straightforward, like "do yoga" or "be proud of," while others, such as "heated debate," "hit the gym," and "be a big deal," are more abstract According to Salman (2017), many familiar collocations do not adhere to logical rules, making them difficult to interpret literally.
Good collocational knowledge is essential for foreign language learners to become fluent and effective communicators A diverse repertoire of collocations enables individuals to communicate more naturally and fluently Using collocations encourages speakers to express ideas concisely, avoiding wordy expressions and complicated sentences Consequently, collocations offer economical yet precise ways to convey the same idea without lengthy explanations.
2.1.3 Challenges to Differentiate Collocations Versus Idioms and Free Combinations
Collocations are determined by linguistic convention rather than logic or frequency (Chen, 2004) Research has positioned collocations on a spectrum between free combinations and idioms (Ha, 2013; Youmei & Yun, 2005) The difference between collocations and idioms is straightforward: collocations consist of word groups commonly used together (e.g., heavy rain), while idioms are expressions whose meanings cannot be inferred from their individual words (e.g., have a sweet tooth) However, distinguishing between collocations and free combinations is more complex Word combinations can be categorized based on restricted and unrestricted senses (Nesselhauf, 2003; Cowie, 1994; Lewis, 2001) as cited in Ha (2013) For instance, free combinations like "find an entrance" can be freely formed, whereas collocations, where the verb's meaning is restricted, require specific nouns (e.g., make a decision, take advantage of) (Ha, 2013).
The working definition of collocations applied in the current study
The working definition of collocation applied in the present research is based on Youmei and Yun‟s (2005) principles for the classification of word combinations as follows:
If the meaning of a phrase cannot be inferred from the individual meanings of its words, it is classified as an idiom, such as "to rain cats and dogs." Conversely, if the meaning can be deduced, it is considered a free combination or collocation.
A combination of words is considered a collocation when we can infer its meaning from the individual words, especially if their usage is limited or if the verb and noun are commonly expected to occur together, such as in the phrase "break the record."
If the meanings of individual words in a combination can help us infer the overall meaning, and the usage of these words is flexible or the verb and noun do not have a predictable relationship, then the combination is considered a free combination, such as in the phrase "look at the picture." (Youmei and Yun, 2005, p 107)
Various types of collocations have been identified by linguists, focusing on elements such as the association between collocation constituents, their fixedness, restrictions, collocational patterns, and specific usage contexts This article provides a concise overview of common collocations that typically occur in pairs.
2.1.4.1 Lexical Collocations vs Grammatical Collocations
Corpus-based studies on collocations
A corpus is defined as a collection of text, both written and spoken, stored in a computer (O'Keeffe et al., 2007) It encompasses a large array of language production, enabling the investigation of lexis, syntax, text, and discourse across various genres, such as newspapers, journals, and books (Koo, 2006) For effective corpus analysis, the text must be in a machine-readable format, allowing for quick retrieval of examples of words or phrases in context from vast datasets (Koo, 2006) Essentially, a corpus serves as a collection of electronic texts accessible via text analysis software.
To conduct effective research in a corpus, it is essential for researchers to consider the language variety, whether the corpus is specialized or general, its size and update frequency, and its representativeness As noted by Dobrić (2009), this representativeness is a crucial characteristic of any corpus, as "there is no one corpus to suit all purposes" (O'Keeffe et al., 2007, p 3) Various types of corpora exist, including general versus specialized, static versus monitor, native versus learner, native versus translated, and monolingual versus multilingual (Dobrić, 2009) Each type of corpus possesses unique language characteristics that cater to the diverse needs of users.
Corpus-based studies leverage corpus data to examine, validate, refute, or refine theories and hypotheses (Tognini-Bonelli, 2001) This methodology typically employs a deductive approach, where a selected corpus serves as a catalyst for confirming or challenging established theoretical constructs Essentially, it tests the validity of a hypothesis against meticulously chosen corpus data (Biria, 2017).
Quantitative techniques play a crucial role in corpus-based studies To effectively compare the usage of the words "big" and "large," it is necessary to analyze their frequency in the corpus, identify the various words that co-occur with each adjective (known as collocations), and assess the commonality of those collocations These aspects rely on quantitative measurements.
A key aspect of the corpus-based approach involves moving past mere quantitative errors to offer functional interpretations that clarify the reasons behind these errors Consequently, significant effort in corpus-based research is focused on elucidating and illustrating these quantitative mistakes.
2.2.3 Corpus-based studies on collocational errors
When judging the appropriateness of collocation, researchers have used reliable collocation dictionaries such as Oxford Advanced Learner‟s Dictionary, the Collins COBUILD English Dictionary, The BBI Dictionary of English
Research on word combinations utilizes resources such as the Oxford Dictionary of English Idioms and corpora like the Corpus of Contemporary American English (COCA) and the British National Corpus (BNC) (Ha, 2013; Youmei and Yun, 2005; Thongvitit and Thumawongsa, 2017; Chen, 2005; Molavi et al., 2014) These studies first consult dictionaries to verify the existence and correct usage of collocations, particularly in identifying misused English collocations (Thongvitit and Thumawongsa, 2017; Ang et al., 2011) If a collocation is not found in the dictionary, researchers then refer to native speaker corpora to assess the frequency of the potential collocation's usage.
The raw frequency of collocation in a corpus is crucial for identifying idiomatic expressions, with various studies establishing different thresholds for determining naturalness Cheng (2016) suggested that collocations with fewer than 10 occurrences in the BNC are not accepted by native speakers, while Laufer and Waldman (2011) proposed a higher minimum of 20 occurrences In contrast, Nesselhauf (2003) accepted collocations with as few as five occurrences, a view supported by Wang and Shaw (2008) and Ang et al (2011) Siyanova and Schmitt (2008) categorized collocation frequency into five levels, justifying a threshold of six occurrences as it captures nearly half of the data These studies illustrate that researchers often base their cut-off points on previous work or the size of their datasets.
Lewis (1997) highlighted that miscollocation is a common source of error in language learning, requiring learners to rely on linguistic conventions for acceptable word combinations Corder (1967) categorized collocational errors into systematic, stemming from ignorance of grammatical rules, and non-systematic, which occur accidentally without the learner's awareness Lexical errors, particularly involving deviant collocations, arise from insufficient knowledge of word usage and their associations (Dukali, 2017, p.57) Research on mis-collocations varies, with some studies focusing on lexical collocational errors and others on grammatical mis-collocations, while many investigate specific combinations like verb-noun or adjective-noun collocations Notably, most studies in this field originate from non-English speaking countries, including Malaysia, Indonesia, Libya, Iraq, China, and Taiwan The following section will provide an overview of these studies.
2.2.4 Approaches to research of collocational errors
2.2.4.1 Corpus-based studies on lexical collocation errors
Research on lexical collocations primarily focuses on three key questions, with a significant emphasis on determining the total percentage of lexical collocation errors identified in the analyzed essays.
(2) What lexical collocation error subtypes are found in the essays analyzed?
Numerous studies have identified the sources and causes of lexical collocation errors, employing quantitative methods to analyze data from 30 to 450 essays (Farghal and Al-Hamly, 2007; Namvar and Ibrahim, 2014; Ridha and Al-Riyahi, 2011; Shitu, 2015; Dukali, 2017; Wells, 2013) Researchers utilized reliable collocation dictionaries, corpus checks, and evaluations by native speakers to assess collocation appropriateness A significant finding is that verb-noun collocation errors are the most prevalent, followed by adjective-noun and noun-noun combinations, while verb-adverb and adverb-adjective errors are less common Notably, Dukali (2017) enhanced her research by incorporating an acceptability-of-collocation survey with 100 native participants to further validate her findings The study identified grammatical errors, including missing determiners, incorrect word order, and inappropriate verb or noun choices, as common causes of collocation errors.
Research by Wells (2013) highlights that incorrect word choice is the primary cause of lexical collocation errors in essays by Ukrainian and Korean students Key factors contributing to these errors include L1 interference, synonym usage, overgeneralization, word coinage, and ignorance of rule restrictions The studies indicate that lexical collocation errors predominantly stem from issues with word selection Non-native students, due to limited exposure to English-speaking environments, often rely on their mother tongue, leading to intuitive word matching and challenges in finding appropriate synonyms in English Consequently, L1 influence on word choice is frequently identified as a significant factor in lexical collocation errors across various research studies.
2.2.4.2 Corpus-based studies on grammatical collocational errors
Grammatical collocation errors often receive less attention than lexical collocation errors (Rodríguez-Fernández et al., 2015) Research, such as that by Sari and Gul (2019), has classified grammatical collocation errors into eight major types, including noun + preposition and adjective + preposition Their study of fifteen essays by first-year students revealed that the most frequent errors were found in adjective + preposition collocations, followed by verb + preposition and noun + preposition Notably, no errors were identified for noun + that clause, adjective + to infinitive, or adjective + that clause Given the limited scope of this research, further studies are needed to comprehensively understand the prevalence of various grammatical collocation errors.
Regarding the classification of grammatical collocation errors, a group of Spanish researchers divided grammatical collocation errors into eight different types:
Determination errors (errors resulting from the omission of a determiner when it is required by the collocation, or from its use when the collocation does not accept it),
Number errors (errors produced when either the plural or the singular form of a lexical unit is required for a particular collocation, but the opposite is chosen),
Gender errors (errors resulting from the choice of the incorrect gender form of the base),
Government errors (errors produced when the governing preposition of the base or the collocate is missing or mistakenly chosen),
Governed errors (errors resulting from the wrong use or omission of a preposition that governs the whole collocation),
Specification errors (errors produced when a modifier of the base is missing),
Pronoun errors (errors resulting from the inappropriate use or the absence of the reflexive pronoun of a verbal collocate),
Order errors (errors produced when the base and the collocate appear in the wrong order) (Rodr guez-Fern andez et al., 2015)
Research on grammatical collocation errors highlights their significance in teaching writing Fadlilah (2016) examined various theses to analyze grammatical collocation patterns and proposed enhancements for teaching She recommended incorporating discussions of English grammatical collocations into Academic Writing courses and encouraging students to analyze their peers' theses to identify collocation errors This approach can effectively raise students' awareness of grammatical collocations and enhance their writing skills.
Studies indicate that errors in grammatical collocation primarily involve prepositions, which often confuse both native and non-native speakers The rationale behind the use of specific prepositions in certain combinations can be ambiguous, making it challenging to select the correct one to pair with nouns, verbs, or adjectives Researchers utilize native corpora to analyze the frequency of preposition collocations and consult collocation dictionaries to ensure contextual accuracy For grammatical collocations involving structures like verb and “to-infinitive” or “that clause,” the framework established by Benson et al (1997) is commonly referenced These structures tend to be easier for non-native speakers to identify as incorrect Interestingly, while lexical collocation errors appear more complex than grammatical ones, research on lexical collocation errors significantly surpasses that on grammatical mis-collocations (Rodríguez-Fernández et al., 2015).
2.2.4.3 Corpus-based studies on both grammatical and lexical collocation errors
Research has explored both lexical and grammatical collocation errors to compare their prevalence A corpus-based study involving 150 EFL students from Taiwanese high schools, conducted by Chen, sheds light on these error types.