Turkish EFL Learners Use of English Collocations
* Author to whom correspondence should be addressed.
Yildiz Journal of Educational Research 2020, Vol. 5, Issue 2, pp. 1-30; doi.org/10.51280/yjer.2020.006
Abstract
Keywords: Grammatical collocation; lexical collocation; language transfer; level of L2 proficiency
Introduction
Research in second/foreign language learning has shown a resurgence of interest in collocations. On the one hand, some studies have discussed collocations theoretically and defined them based on lexical, syntactic, and semantic features (e.g., Gorgis & Al- Kharabsheh, 2009; Howart, 1998; Nation, 2001). On the other hand, several studies extensively investigated collocations from pedagogical aspects endeavoring to identify the development of collocational knowledge at different proficiency levels, and L2 learners’ collocational mistakes along with certain strategies of teaching collocations (e.g., Balcı & Çakır, 2012; Lewis, 2000; Lindstromberg & Boers, 2008; Rahimi & Momeni; 2012; Vasiljevic, 2008). The research literature also entails corpus studies that have analyzed comprehensive collocation data with regard to frequency of occurrence and come up with corpora of spoken and written English such as British National Corpus (BNC) and Corpus of Contemporary American English (COCA) that are available online to researchers, educators, and teachers. These corpora provide language teachers with a large collection of samples of written and spoken language from an extensive range of sources in the English language and help the development of authentic language teaching materials and resources in English (Vasiljevic, 2014).
Collocations, simply defined as word combinations, play a crucial role in second/foreign language teaching because they constitute one of the essential portions of native speaker competence (Men, 2018; Nesselhauf, 2003). According to Kennedy (2000), “the strategy of acquiring formulaic speech is central to the learning of language” (p.110). Alexander (1984) points out that three Cs, namely collocation, context, and connotation should be the primary focus of teaching process. Likewise, to Nattinger (1988), learning collocation is essential because these associations help learners store words in the memory. They also aid in defining the semantic area of a word, and let people know the kind of words that can come together. In line with these researchers, Taiwo (2004) states that collocations deserve to be one of the central parts of vocabulary learning as effective performance of ESL learners depends on their storage of collocations. In addition, Shin and Nation (2007) argue that learning collocations provides learners with an effective way to enhance language fluency and native-like selection of language use. Fan (2009) also emphasizes the significance of collocation competence for effective communication and how it helps L2 learners write or sound more native-like. Emphasizing the fundamental role of collocations in
vocabulary description and pedagogy Carter (2006) pointed out that “for the learner
of any second or foreign language, learning the collocations of that language is not a luxury if anything more than a survival level mastery of the language is desired because collocation permeates even the most basic, frequent words” (p.2). Apparently, multiword strings offer L2 learners alternative ways of saying or writing something and improve their style in language skills including reading, listening, speaking, and writing. Despite its significance in L2 learning, few studies have come up with suggestive tips for teaching collocations (e.g., Taiwo, 2004; Vasiljevic, 2014). That is, language teachers are still uncertain about which collocations to teach and how to teach them in language classes. Meanwhile, Carter and McCarthy (1988) argue that collocation is an important facet of English lexicon, and because foreign-learner texts involve deviant forms of collocations, there is certainly a need for an understanding of and a concern with collocations by teachers and students. Accordingly, this present study attempts to find out the types of collocations that Turkish pre-intermediate and advanced L2 learners at tertiary level produce in their in-class compositions. In parallel 3
with this, this study also intends to show whether there are any manifestations of similarities or differences among learners of different proficiency in the written production of L2 collocations. Some previous studies also show that although collocation knowledge is an essential part of communicative competence, it is usually troublesome to L2 learners of English, even at very advanced stages of language proficiency (e.g., Bahns & Eldaw, 1993; Howarth, 1998; Mahmoud, 2005; Koosha & Jafarpour, 2006). As Taiwo (2004) emphasizes “learners’ lack of knowledge of collocational patterns of lexical items makes them prone to all sorts of collocational errors, which can be more disruptive in communication than grammatical errors” (p.1). Therefore, L2 teachers need to know the factors affecting L2 learners’ collocation attainment. The current study, thus, aims to investigate whether L1 interference plays a role in Turkish pre-intermediate and advanced L2 learners’ difficulties with word combinations in essays. Defining Collocations The origin of the term, collocation, in the literature goes back to Palmer (1933, as cited in Kennedy, 2003) who defined a collocation as “a succession of two or more words that must be learnt as an integral whole and not pieced together from its component parts” (p. i). In Dictionary of Language Teaching and Applied Linguistics, collocations are defined as the ways in which words are combined together. They refer to restrictions on how words can be used together, for example, which prepositions and verbs are used together, or which nouns appear with particular verbs (Richards, Schmidt, Platt & Schmidt, 2002). Accordingly, collocations, by definition, constitute expressions that exhibit words in company or conventional, recurring and non-idiomatic expressions that are made up of basically two lexical items such as constant/sharp (but not continuous/strong) pain (Adj + N) (Gorgis & Al- Kharabsheh, 2009). However, to Leśniewska (2006), although they have received considerable attention in applied linguistics over the last few decades, collocations remain “notoriously difficult to define and different (often slightly contradictory) definitions proliferate in the literature” (p.95). 4
Types of Collocations Different researchers categorize collocations from different
because the issue of collocations has been studied in different fields such as corpus linguistics, semantics, and phraseology. Emphasizing the importance of collocations in developing L2 proficiency, Howarth (1998) suggested a continuum of word combinations from the most free combinations to the most fixed idioms, instead of discrete classes. Table 1 presents Howarth’s examples of word combinations drawn from the native speaker corpus along with their categorization. Table 1. Howarth’s categorization of collocations (p.35) Combination
Introduce
Benson, Benson and Ilson (1986a) categorized collocations into two groups, namely lexical collocations, which consist of verbs, adjectives, nouns, and adverbs (e.g., relieve pressure, heavy rain, follow closely), and grammatical collocations, which are phrases consisting of a dominant word, such as a noun, adjective, or a verb, and a preposition or a grammatical structure like an infinitive or a clause such as “feel sorry to” and “listen to music”. To Benson and colleagues, lexical collocations may appear as 5
combinations of verb+noun (e.g., compose music), adjective+noun (e.g., strong tea), noun+verb (e.g., bombs explode), adverb+adjective (e.g., deeply absorbed), and verb+adverb (e.g., argue heatedly). However, grammatical collocations have various types of patterns like noun+preposition (e.g., blockade against), adjective+preposition (e.g., angry at smb.), verb+object+preposition+object (e.g., We invited them to the meeting), and adjective+that (e.g., We were afraid (that) we were going to capsize the boat). Furthermore, depending on their cohesiveness, Benson, Benson and Ilson (1986b) classify lexical combinations into five groups: compounds (completely frozen, no variation is possible), idioms (relatively frozen expressions whose meaning do not reflect the meanings of their component words), transitional combinations (more frozen but less variable than collocations, their meanings are close to the meanings suggested by their component parts), collocations (loosely fixed, arbitrary, recurrent word combinations), free combinations (the least cohesive of all, the component parts can combine with other lexemes freely). Nesselhauf (2003) categorizes word combinations into three major classes: free combination, collocation, and idiom. For instance, in the combination, read a book, the occurrence of the verb read does not necessarily expect the company of a book, as read can be followed by any work in written form such as a letter, a newspaper or a report provided that it is semantically and syntactically acceptable (e.g., write a book, buy a book). Here, the senses in which read and a book are used are both unrestricted, producing a free combination. On the other hand, in the combination deliver a letter, the sense in which deliver is used is much more restricted than that of read in read a book. The verb, deliver, according to the Oxford Advance Learner’s Dictionary (OALD), means, “to take goods, letters, etc. to the person or people they have been sent to”. Therefore, deliver, in this sense, can only come together with a limited number of nouns such as goods and letters. The occurrence of the verb, deliver, demands the co-occurrence of a letter to a certain degree, forming a collocation. Following these explanations, some previous studies (e.g., Nesselhauf, 2003; Youmei & Yun, 2005) classify word combinations into three major classes: Idioms: The meaning of a combination cannot be deduced from the meanings of its member words (e.g., to rain cats and dogs).
Collocations: The meaning of a combination can be deduced from the meanings of its member words. But, the senses in which member words are used are restricted to some extent or the verb and noun in the combination are mutually expected (e.g., make a decision). Free Combinations: The meaning of a combination can be guessed from the meanings of its member words. Also, the senses in which the member words are used are unrestricted or the verb and noun in the combination are not mutually expected. (e.g., learn a language). Collocational categorizations proposed by Nesselhauf (2003) and Benson et al. (1986a) were adopted for the analysis of the data in this study. The Role of L1 Transfer and Proficiency Level in Collocation Performance Collocations are generally considered to be challenging in L2 teaching and learning due to various factors such as negative transfer, learners’ proficiency level, degree of restriction of a combination, or overgeneralization. L1 interference and learners’ proficiency level were examined within the scope of this study. Many research studies provide evidence which supports the idea that L1 interference plays a major role in L2 learners’ collocation performance. Among such studies is Fan’s (2008) investigation of a native corpus and a non-native corpus. Fan (2008) concludes that the performance of the Hong Kong students in collocational use might be adversely influenced by their L1, plus their inadequacy in the lexis and grammar of the target language. In addition, Taiwo (2004) suggests that L2 teachers should focus on areas of differences in the collocational patterns of L1 and the target language because collocations in L1 are often translated directly into English. To illustrate, Taiwo gives (2004) Yoruba learners’ expressions of English: house money for rent, their eyes were black then for they were not civilized then. Nesselhauf (2003) studied the use of collocations by German advanced learners of English, and analyzed the use of verb-noun collocations such as take a break or shake one’s head in free written production and concluded that the learners’ L1 had a much stronger effect than earlier studies had predicted. The researcher remarked that downplaying and disregarding the influence of L1 in teaching L2 collocations is misleading. Nesselhauf (2003) suggested that non-congruence between L1 and L2 collocations should receive 7
particular attention in language teaching. Furthermore, the findings of Mahmoud’s (2005) study support the claim that EFL students are dependent on interlingual and intralingual strategies for producing English collocations. That is, Mahmoud (2005) presented that Arabic-speaking students made errors when producing collocations, particularly the lexical combinations in English due to interlingual transfer from Arabic to English. They used the interlingual transfer strategy, and altered the Arabic words with English ones. As a result, some collocations were produced accurately due either to positive interlingual transfer or direct acquisition from the L2 language input (e.g., reach an agreement, a white lie) whereas other collocations were produced inaccurately due to negative L1 transfer (e.g., * by this way instead of in this way), *in the phone instead of on the phone, * ends with instead of ends in). As a result of the study with Taiwanese L2 learners of English, Chen (2008) also reported that V+N collocations were the most demanding type of lexical collocation whereas N+Prep collocations were the most difficult kind of grammatical collocation for the participants. It was found that V+N collocations presented the most frequent lexical collocational error types whereas N/V+Prep displayed the most frequent types of grammatical collocational errors. The researcher claimed that L1 transfer was one of the main sources of these collocation errors. Bahn (1993) found very similar results to that of Chen (2008) and concluded that the participants had difficulty in V+N lexical collocations and N+Prep grammatical collocations. The
results
participants’ opinions and their scores on the collocation test. Bahn (1993) suggested that L1 transfer was among the most important contributing factors in collocation errors. Likewise, Kasuya (2008) confirmed that Japanese students tend to guess English word partners from Japanese collocational restrictions. For example, when students wanted to express that their friend is having “a serious illness”, they attempted to say “a heavy illness”, because in Japanese when illness is serious a speaker says “omoi (heavy) + byouki (illness)”. Moreover, a few studies revealed a positive correlation between learners’ general English proficiency level and collocation performance in tests, suggesting that collocation competence is an important aspect of language proficiency (e.g., Al-zahrani (1998); Bahn (1993); Chen, 2008). In the aforementioned study that Chen (2008) did 8
with university students, he found a significant positive correlation between the participants’ achievement test in collocation and their English subject scores, signifying that collocation competence is an essential part of language proficiency. On the other hand, in a study among L2 learners with different English proficiency levels, Youmei and Yung (2005) concluded that there was an apparent similarity across different-level learners in the production of L2 collocations. They had similar difficulties in the choice of words. The researchers proposed that L2 learners with different proficiency levels might be challenged by choosing the right verbs that collocate with the nouns. Keshavarz and Salimi (2007), however, reported a strong correlation between the participants’ collocation performance on open-ended and multiple-choice cloze tests and overall language proficiency measure. The researchers suggested that learners’ collocational competence and proficiency level are positively connected, and that proficient language users are prone to know a large number of collocational patterns. Based on the literature, the following research questions were investigated in this paper:
Methodology
Research Design This paper is about a cross-sectional small-scale research study which involves three groups: pre-intermediate L2 users of English, advanced L2 users of English, and native speakers of English. A cross-sectional research design was adopted as it allows researchers to collect data from different sub-groups at one specific point in time (Levin, 2006). The three groups included in the present study also helped the researcher explore whether collocation competence and language transfer in collocation 9
performance was related to foreign language proficiency. That is, the pre-intermediate and advanced groups were compared in terms of the amount and the types of collocations they produced as well as the amount of L1 transfer they performed in the current study. The types of the collocations native speaker participants made use of were also compared to those produced by pre-intermediate and advanced groups. Setting The study was conducted in two prestigious state universities in Istanbul. English Preparatory Program and Department of Foreign Language Education of a state university were the settings where the data for pre-intermediate and advanced groups was gathered. The native speaker data was collected from the American students in the Lifelong Learning Centre of another state university. Participants 30 participants were included in this study. There were 12 non-native users of English in each group, and 6 native English speakers. The pre-intermediate participants were attending the intensive English program in the School of Foreign Languages at a state university in Turkey. The advanced-level participants were freshmen, studying in the Department of Foreign Language Education at the same state university. Their ages ranged from eighteen to twenty two. The native speaker participants were all American and studying at another Turkish state university as exchange students taking Turkish courses in the Lifelong Learning Center when the data was collected. These native speakers were chosen as they were convenient and ready to make a voluntary contribution during data collection. They were between the ages of nineteen and twenty-three. They were intermediate students in Turkish and had been living in Istanbul for eight months. Table 2 presents the number and gender distribution of each group.
Data Collection Instruments and Procedure The following tasks were undertaken to gather data as a part of the proposed research: Task 1: All the participants performed an in-class writing in which they were asked to compare and contrast two cities on one page in forthy minutes. Task 2: The participants took a multiple-choice collocation test which involves 21 test items. The first kind of instrument, in-class writings, was mainly conducted to identify the number and type of collocations utilized by the participants. Such writing also provided the kinds of collocational mistakes the participants made. In this writing task, the participants were expected to write one-page essay about the two cities they had been or lived within forty minutes. The students were not allowed to use a dictionary. The second instrument, multiple-choice collocation test was administered to investigate whether the research participants transfer from L1 to L2 in relation to V+N, Adj+N, and PP collocations. The participants were permitted to submit the test when they finished answering all test items. Each test item had been chosen cautiously and its proof reading was completed by the native speaker who was teaching English in Turkey for 11
twelve years. A Turkish native speaker who was an expert in language transfer also assessed the appropriateness of test items. Due to the time constraint, the test involves 21 items. It composes of 7 V+N items, 7 Adj+N items, 5 Adj+Prep and 2 N+Prep items. The items related to prepositions were merged under PP during the data analysis. The tasks were administered at the most convenient time for the participants and the lecturers whose permission had been obtained previously. Data Analysis Oxford Advance Learner’s Dictionary (OALD), Oxford
Dictionary2, and Longman Active Study Dictionary (LACD) were used in order to specify whether a word group used by the participants in their writings was a collocation, and whether a particular collocation was used correctly. These three dictionaries were chosen as they are recognized among the most comprehensive learner dictionaries. In addition, Nesselhauf’s criteria (2003) and Benson’s et al. classification (1986a; 1986b) were adopted to identify word groups as collocations and to categorize the collocation types used by the participants. According to Nesselhauf (2003), in order for a word group to be recognized as a collocation, dictionaries should represent clear indications that the use of a particular word (e.g., verb) is restricted to a few other words (e.g., noun). This means that the combination has a restricted sense in a specific context. Accordingly, the target collocations were judged accurate if they were found in the same form in two dictionaries. Besides, a native speaker also checked the collocation data and helped the researcher with regard to accurate use
categorization of the collocations in the essays. The native speaker was asked to judge whether the use of a word combination was correct or wrong in the given context. She was also asked to provide the correct form of the target collocation if it was used incorrectly by the participants. After the target word group was identified as a true collocation, its type was specified by means of Benson et al.’s (1986a; 1986b) categorization of collocations. In addition, Kruskal Wallis Test was run to explore whether the participants’ level of proficiency in English would make statistically significant differences in
This is an online dictionary: https://www.freecollocation.com/ 12
productive collocation proficiency. As follow-up tests, Mann-Whitney U tests were computed to evaluate pairwise differences among the three groups. As for multiple-choice collocation test, the total number of accurate answers for each section (V+N, Adj+N, and PP) for all groups was worked out. Then, the percentages of accurate answers were calculated for each group. Next, the number and the percentages of L1 transfer for each section (V+N, Adj+N, and PP) in each level was calculated and illustrated by means of tables and graphs. Here, the participant’s response was counted as L1 transfer error provided that he/she chose the option that reflected L1 transfer. For example, in the collocation test, the participant is presented with the test item, “Late hours and……………… (a. strong b. dark c. dense d. heavy) tea are no good for me” to check his/her collocational knowledge for Adj+N category. If the participant chooses the distractor, dark, rather than the correct answer, strong, the answer is counted as L1 transfer error. In English, the adjective, strong, is the collocate of the noun, tea, in contrast to the Turkish language in which dark is used to describe the density of the tea.
Results
Productive Collocation Proficiency Students’ collocation performances in their essays were examined thoroughly in order to investigate whether the proficiency in producing collocations changes in accordance with general proficiency level in English. It was assumed that advancedlevel L2 learners would have a better command of L2 collocations and produce more collocations, and make fewer collocation mistakes than pre-intermediate L2 learners. It was also presumed that the native group would present the best performance in the use of word combinations. Table 3 illustrates the number and the percentage of the collocations that each group used correctly in the writings.
As Table 3 indicates, the total number of collocations that advanced learners (n= 12) used accurately (132 correct collocations out of 147) in their writings outnumbered that of pre-intermediate learners (29 correct collocations out of 36). That is, advanced-level L2 learners tended to produce more collocations, and made fewer collocation mistakes compared to pre-intermediate L2 learners. Furthermore, because the participants in the advanced group would normally be expected to have a better command of English and be more fluent writers, they attempted to produce longer writings with more collocations within the allocated time. On the other hand, it was observed that the pre-intermediate group wrote shorter writings with fewer collocations, spending much time forming grammatical sentences, finding appropriate word combinations, and organizing their ideas. In addition, it was found that the native speaker participants (n= 6) utilized 85 collocations in their writings. This shows that they frequently use collocations. Accordingly, Kruskal-Wallis Test was conducted in order to explore whether these observed differences on the use of collocations in essays were statistically meaningful for the groups with varying degrees of proficiency levels. Statistically significant differences were found among the three groups of participants (H(2)= 24.108, p= .000) with a mean rank of 6.5 for the pre-intermediate, 19.1 for the advanced, and
26.2. for the native group. Following this, a series of Mann-Whitney U tests was
computed to compare the performances of the three groups. As Table 4 displays, the results indicated that the native speakers’ productive collocation performance (Mdn=14) was significantly better than the performances of the advanced group (Mdn=11.5), U= 14
7.500, p<.01 and of the pre-intermediate group (Mdn= 2), U= 0.000, p<.001. In addition, the results also revealed that the advanced group (Mdn= 11.5) performed significantly better than the pre-intermediate group (Mdn= 2), U= 0.00, p<.001. Table 4. Participants’ Productive Collocation Performances
Types of Collocations This study also deals with the type/s of the collocations that the preintermediate and advanced L2 learners produced in their writings. It was predicted that the advanced learners would use more various types of collocations than the preintermediate learners. The result of the Task 1 almost partly contradicted this prediction and revealed that the advanced group attempted to generate V+N (e.g., lose confidence), V+Adv (e.g., pass swiftly), and Adj+N (e.g., dense population) types of lexical collocations, and V+Prep (e.g., contribute to), Adj+Prep (e.g., glad to), Prep+N (e.g., by bike) grammatical collocations. Similarly, the pre-intermediate group created V+N (e.g., spend time), and Adj+N (e.g., a warm climate) lexical collocations, and V+Prep (e.g., live in), Adj+Prep (e.g., different from), and Prep+N (e.g., at night) 15
grammatical collocations. It seems that both groups used the same types of collocations except for V+Adv lexical collocation that the advanced group utilized. However, it should be stressed that the number of the collocations that each group used for each type differs. For instance, the advanced participants tended to produce 45 V+Prep collocations whereas the pre-intermediates generated 13 V+Prep collocations. Figure 1 and Figure 2 show the percentage of each collocation type that pre-intermediate and advanced groups produced in their essays.
Figure 1. The Percentages of Collocation Types Used by the Pre-Intermediate Learners in Essays
Figure 2. The Percentages of Collocation Types Used by the Advanced Learners in Essays This study was also concerned with if there were any differences between English native speakers and non-native English speakers in terms of collocation production. The native speaker participants produced 5 types of lexical collocations, namely V+N (e.g., hold festivals), V+Adj (e.g., feel liberated), V+Adv (e.g., research thoroughly), Adj+N (e.g., heavy rain), Adv+Adj (e.g., densely populated), and 3 types of grammatical collocations; V+Prep (e.g., spread out), Adj+Prep (different from), and 16
Prep+N (at the bus stop). Figure 3 illustrates the percentages of the types of collocations that the native speaker group produced. V+N V+ADJ 9% 13%
Figure 3. The Percentages of Collocation Types Used by the Native Speakers in Essays These findings demonstrated that there were similarities and differences in the types of collocations that native and non-native speakers of English utilized in their writings. That is, both the native and non-native English speakers made use of similar types of collocations except for V+Adj and Adv+Adj collocations that the native English speakers applied in their essays. Furthermore, all groups preferred V+Prep, Adj+N, Adj+Prep, and Prep+N word combinations more than other types of collocation. The Role of L1 Transfer in L2 Collocation The current study also investigated if the L2 learners’ L1 influences their production of collocations. It was presumed that the influence of learners’ L1 (Turkish) on the production of L2 (English) collocations would diminish but still exist among the advanced-level L2 learners. To investigate this, a multiple-choice collocation task which involves 21 items was conducted. The total numbers of correct and incorrect answers were calculated, and the questions that the participants left blank were extracted. Table 5 below presents the pre-intermediate and advanced learners’ accuracy scores for each category included in the collocation test.
Table 5. Accuracy Scores Based on the Collocation Test Types of Collocations
As Table 5 presents, the advanced group is more successful than the preintermediate group in all collocation categories. Among the three types of collocations, the pre-intermediate group presented the best performance in V+N (53%), which was followed by PP and Adj+N categories. The advanced group was the most successful in PP collocations (70%). When the performances of the two groups were compared, regarding V+N category, the total number of the advanced group’s correct answers is more than that of the pre-intermediate group (54 vs 45). On the other hand, concerning PP collocation, there is a considerable difference between the two groups: The advanced group surpassed the pre-intermediate group by giving 59 correct answers out of 84. Both groups are underperformed in Adj+ N category. Whereas the preintermediate group gave 25 accurate answers (29%), the advanced group came up with 41 correct answers (48%), which is lower compared to other collocation categories.
Table 6 indicates that both pre-intermediate and advanced participants made errors due to interference from L1 in all categories. The highest percentage of L1 transfer appeared in Adj+N category in both groups (48% for the pre-intermediate group; 43% for the advanced group). As for V+N type of collocations, the amount of language transfer errors in the advanced group is lower than that of the pre-intermediate group (25% vs 30%).The impact of L1 on the use of PP collocations seemed to be high in the pre-intermediate group (46%). The effect, however, diminished noticeably in the advanced level (24%).
Discussion and Conclusion This research study focuses on the use of L2 collocations among Turkish learners of English with different proficiency levels. In particular, the four research questions addressed in this paper intended to investigate the relation of collocation competence to L2 learners’ English proficiency levels, the type/s of the collocations produced by L2 learners with different L2 proficiencies, and the degree of L1 interference in the learners’ collocation performances. The collocation performances of two groups of L2 learners (i.e., pre-intermediate and advanced students) and a group of native speakers in a writing task and a collocation test were examined. In the current study, the advanced participants comfortably surpassed the preintermediate group with respect to the correct collocations they provided in the writing task within the allocated time. The results also displayed that the native speakers’ 19
productive collocation performance was significantly better than the performances of the advanced and of the pre-intermediate group. These results are in line with AlZahrani (1998) and Keshavarz and Salimi (2007), who found a positive correlation between L2 learners’ overall language proficiency and collocation performance. In addition,parallel to Fan (2009), these results imply that L2 learners need support in EFL classes to enhance their collocation competence which helps them achieve more effective communication, and improve their style in different language skills. Regarding the types of collocations identified in the participants’ essays, both pre-intermediate and advanced groups tended to use similar categories, namely V+N and Adj+N lexical collocations, and V+Prep, Adj+Prep and Prep+N grammatical collocations. However, it should be noted that in addition to these collocation types, the advanced group also produced V+Adv lexical collocations. Further, different form preintermediate and advanced L2 learners, the native speakers included V+Adj and Adv+Adj in their essays. These findings indicated that as language users’ level of proficiency increases, their collocation repertoire expands and includes variety. The results also showed that whereas V+Prep became the most frequently-preferred collocation type, V+N, V+Adv, V+Adj, and Adv+Adj became the least preferred types of collocations for the EFL learners. This result partly gets along with Chen’s (2008), Bahn’s (1993) and Youmei and Yung’s (2005) conclusion that V+N collocations are the most demanding lexical collocations. A possible explanation is that learners might avoid using the most demanding collocation types such as V+N that go beyond their existing proficiency level. Also, the learners made use of V+Prep collocations probably because they are more frequently exposed to this type of collocation in textbooks, extra materials, and teacher talks. They may also be tested on V+Prep type of collocations in the quizzes and exams. Thus, the learners tended to practice this type of collocations more than other collocation types. In addition, the participants in the pre-intermediate group did not use V+Adv lexical collocation in their writings and there were fewer instances of V+Adv word combinations in advanced group data in comparison with other collocation types. In addition, neither pre-intermediate nor advanced participants utilized V+Adj and Adv+Adj types of collocations in their writings. These findings imply that L2 learners should come across such types of collocations in language 20
materials and teacher talks more frequently and need more practice with various collocation types in different contexts. Furthermore, L2 adult learners may need explicit teaching of certain word combinations rather than being exposed to them implicitly. L2 teachers should allocate more time for collocation practices in language classrooms. In addition, the result of the collocation test reveals that there is an obvious similarity across the two EFL learner groups in the production of Adj+N collocations. That is, rather than V+N and PP categories, the participants in both pre-intermediate and advanced groups had the most difficulty in finding appropriate adjectives that would collocate with the nouns in the collocation test. Overall, the pre-intermediate group came up with fewer correct answers than the advanced group in the collocation test. This, in turn, displays that collocation performance is related to proficiency level. However, the accuracy score of the advanced group is not very sufficient (154 correct answers out of 252). Here, language teachers may play a crucial role in developing L2 learners’ collocational knowledge. They can model various kinds of collocations via L2 materials and teacher talk, and provide L2 learners with more exposure and practice by means of different language activities on word combinations. Furthermore, the results of this study reveal that the impact of learners’ L1 on the production of L2 collocations exists among L2 learners regardless of their proficiency levels. However, depending on the type of the collocation, its impact either declines or remains the same in the advanced-level. This finding confirms Nesselhauf’s (2003), Taiwo’s (2004), and Kasuya’s (2008) argument that EFL/ESL learners do transfer collocations from their first language improperly to the second language context. Here, language teachers may emphasize the collocational differences between the learners’ mother tongue and the second language, and help L2 learners increase their awareness of multiple word strings in the L2. All in all, this small-scale research suggests that comprehensive instructional programs in EFL learning should incorporate classroom tasks that systematically promote L2 learners’ awareness of word combinations, help them practice collocations, and increase their collocation threshold. As Taiwo (2004) points out, collocations are indispensable components of lexicon. They are as equally significant as grammatical errors and indeed, more disruptive in communication. Thus, it might also be concluded that some collocations should specifically be selected for teaching with reference to L1 21
in L2 classrooms. In other words, comparing and contrasting similarities and differences between L1 and L2 collocations might help EFL learners attain higher collocation proficiency. Here again, L2 teachers should increase learners’ awareness of the concept of L2 collocation and its characteristics by exposing learners to various types of collocations in L2 classroom settings. In this way, as Shin and Nation (2007) emphasize, L2 learners’ English may sound more natural. Learners may also express themselves in a variety of ways instead of being stuck in independent house* (detached house), my grow country* (home/native country), hard cold* (bitterly cold) or enter an exam* (take an exam). Ultimately, learning collocations will help them write and speak English more naturally, fluently and accurately. However, further research is called for to explain the role of L1 in L2 collocation performance and to propose teaching strategies that incorporate collocation learning into English
programs. Limitations Because this is a small-scale research, its conclusion must be seen as limited and suggestive. This study examined the collocation performance of a limited sample of university students, whose English language proficiency levels were either preintermediate or advanced. It would therefore be interesting to replicate the study by extending learner variables, such as language proficiency level, age and different language backgrounds. Further, in this study, six native speakers were reached. A better comparison between native and nonnative speaker collocation data could have been done if the numbers of the participants had been larger and kept equal. In addition, due to time-constraint, the collocation test used in this study involved only three types of collocations. Different conclusions could have been drawn if the test had involved more various collocation types, and if the number of items for each type had been more than seven.
References Alexander, R. L. (1984). Fixed expressions in English: reference books and the teacher. ELT Journal, 38(2), 127-134. Al-Zahrani, M. S. (1998). Knowledge of English lexical collocations among male Saudi college students majoring in English at a Saudi university. [Unpublished doctoral dissertation]. Indiana University of Pennsylvania. Balcı, Ö. & Çakır, A. (2012). Teaching vocabulary through collocations in EFL Classes: The case of Turkey. International Journal of Research Studies in Language Learning, 1(1), 21-32. Bahn, J. (1993). Lexical collocations: A contrastive view. ELT Journal, 47(1), 56-63. Bahns, J. & Eldaw, M. (1993). Should we teach EFL students collocations?. System 21(1), 101-114. Benson, M., Benson, E., Ilson, R. (1986a). The BBI combinatory dictionary of English: A guide to word combinations. Amsterdam: John Benjamins. Benson, M., Benson, E., Ilson, R. (1986b). Lexicographical description of English. Amsterdam: John Benjamins. Carter, R. (2006). What is advanced-level vocabulary? The case of chunks and clusters. In C. Coombe (Ed.), Words Matter: The importance of vocabulary in English language teaching and learning (pp. 23-41). Alexandria, Va: TESOL.
Carter, R. & McCarthy, M. (1988). Vocabulary and language teaching. Essex: Longman. Chen, M. (2008). A study of the English collocation competence of college students in Taiwan. [Unpublished master’s thesis]. Taiwan University of Science and Technology. 23
Fan, M. (2009). An exploratory study of collocational use by ESL students –A task based approach. System, 37, 110-123. Gorgis, D. T. & Al-Kharabsheh, A. (2009). The translation of Arabic collocations into English: Dictionary-based vs. dictionary-free measured knowledge. Linguistik Online, 37, 21-33. Howarth, P.(1998). Phraseology and second language proficiency. Applied Linguistics, 19(1), 24-44. Jobbins, A. C. & Evett, L. J. (1998) Text segmentation using reiteration and collocation.
https://www.aclweb.org/anthology/C98-1097.pdf. Kasuya, M. (2008).Teaching collocation in Japanese classrooms. 1-16. Retrieved December 21, 2020 from https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.138.8395&rep=rep 1&type=pdf. Kennedy, G. (2000). An introduction to corpus linguistics. Beijing: Foreign Language Teaching and Research Press. Kenndey, G. (2003). Amplifier collocations in the British national corpus: Implications for English language teaching. TESOL QUARTERLY, 37(3), 467-487. Keshavarz, M. H. & Salimi, H. (2007). Collocational competence and cloze test performance: A study of Iranian EFL learners. International Journal of Applied Linguistics,17(1), 81-92. Koosha, M. & Jafarpour, A. A. (2006). Data-driven learning and teaching collocation of prepositions: The case of Iranian EFL adult learners. Asian EFL Journal, 8, 192-210. 24
Levin, K. A. (2006). Study design III: Cross-sectional studies. Evidence-Based Dentistry, 7, 24-25. Leśniewska, J. (2006). Collocations and second language use. Studia Linguistica, 95105. Retrieved February 8, 2021 from https://filg.uj.edu.pl/documents/41616/4333138/12306-Lesniewska.pdf. Lewis, M. (Ed.). (2000). Teaching collocation: Further developments in the lexical approach. London: Language Teaching Publications. Lindstromberg, S. & Boers, F. (2008). Teaching chunks of language. Cambridge: Cambridge University Press. Mahmoud, A. (2005). Collocation errors made by Arab learners of English. Asian EFL Journal Press. The Professional Teacher Articles, 5. 117-126. Retrieved January 18, 2021 from https://www.asian-efl-journal.com/monthly-editionsnew/collocation-errors-made-by-arab-learners-of-english/. Men, H. (2018). Vocabulary increase and collocation learning: A corpus-based crosssectional study of Chinese learners of English. Singapore: Springer. Nattinger, F. (1988). Some current trends in vocabulary teaching. In R. Carterm & M. McCarthy (Eds.), Vocabulary and language teaching. London (pp.62-82). Longman. Nesselhauf, N. (2003). The use of collocations by advanced learners of English and some implications fro teaching. Applied Linguistics, 24 (2), 223-242. Rahimi, M. & Momeni, G. (2012). The effect of teaching collocations on English language proficiency. Procedia - Social and Behavioral Sciences 31, 37-42. Richards, J., Schmidt, R., Platt, H. & Schmidt, M. (2002). Dictionary of language teaching and applied Linguistics. Essex: Longman. 25
Shin, D. & Nation, P. (2007). Beyond single words: the most frequent collocations in spoken English. ELT Journal, 62(4), 339-348. Taiwo, R. (2004). Helping ESL learners to minimize collocational errors. The Internet TESL
http://iteslj.org/Techniques/Taiwo-Collocation.html. Vasiljevic, Z. (2008). Developing collocational competence of second language learners. The EastAsian Learner, 4 (1), 46-50. Vasiljevic, Z. (2014). Teaching collocations in a second language: Why, what and how?. ELTA Journal, 2(2). 48-73. Youmei, Y. & Yun. Z. (2005). A tentative corpus-based study of collocations acquisition by Chinese English language learners. Canadian Social Science, 2(2), 105-113.
Genişletilmiş Özet Amaç İkinci dil öğrenenlerin eşdizim yeterliliği, ikinci/yabancı dil edinimi alanında son yıllarda oldukça dikkat çeken bir konudur. En temel tanımıyla, sözcük bileşenleri olarak nitelendirilen eşdizimler, ikinci/yabancı dil öğretiminde önemli bir rol oynamaktadır (Alexander, 1984; Men, 2018; Nattinger, 1988; Nesselhauf, 2003; Shin & Nation, 2007). Yapılan çalışmalar çoklu sözcük öbeklerinin ikinci dil öğrenen bireylere bir şeyler söylemenin veya yazmanın alternatif yollarını sunduğunu ve okuma, yazma, dinleme ve konuşma becerilerinin
öğrenimindeki önemine rağmen, alan yazında yalnız bir kaç çalışma eşdizim öğretimine ilişkin önemli ipuçları vermektedir (Örn. Taiwo, 2004; Vasiljevic, 2014). Yabancı dil öğretmenleri dil sınıflarında hangi tip eşdizimleri nasıl öğretecekleri konusunda kararsızlık yaşamaktadır. Bu çalışmanın amacı İngilizceyi yabancı dil olarak öğrenen ve farklı İngilizce düzeylerine sahip üniversite öğrencilerinin sınıf içi kompozisyonlarında kullandıkları eşdizim miktarını ve türlerini belirlemektir. Bazı çalışmalar, öğrencilerin yabancı dil genel yeterlilik seviyesi ile eşdizim performansı arasında pozitif bir korelasyon olduğunu ortaya koymuştur (Örn. Al-zahrani; 1998; Bahn, 1993; Chen, 2008). Bunun yanı sıra, bir kısım araştırma, anadil transferinin yabancı dil eşdizim performansında önemli bir rol oynadığına ilişkin kanıtlar sunmuştur (Chen, 2008; Fan, 2008; Keshavarz & Salimi, 2007; Mahmoud, 2005; Nesselhauf, 2003). Buna paralel olarak, bu çalışmanın bir diğer amacı katılımcıların anadilinin ikinci dil eşdizim kullanımı üzerindeki etkisini araştırmaktadır. Bu çalışmada aşağıdaki araştırma sorularına cevap aranmıştır:
Yöntem
Küçük ölçekli, kesitsel araştırma deseninin benimsendiği bu çalışmada, alt orta ve ileri düzey dil öğrencilerinin sınıf içi kompozisyonlarda kullandıkları İngilizce eş dizimler ve anadilden transferin eşdizim performansındaki etkisini ölçen çoktan seçmeli test sonuçlarını incelenmiştir. Çalışmada üç alt grup bulunmaktadır: Anadili Türkçe olan ve İngilizce düzeyi alt orta olan katılımcılar (Katılımcı sayısı=12), Anadili Türkçe olan ve İngilizce düzeyi ileri olan katılımcılar (Katılımcı sayısı=12) ve anadili İngilizce olan katılımcılar (Katılımcı sayısı=6). Toplam katılımcı sayısı 30’dur. Çalışma, İstanbul’da bulunan prestijli iki devlet üniversitesinde gerçekleştirilmiştir. Bu araştırmada, iki farklı veri toplama aracı kullanılmıştır. Sınıf içi kompozisyon yazımında katılımcılardan verilen süre içerisinde iki şehri karşılaştırdıkları
kompozisyon yazmaları istenmiştir. Çoktan seçmeli testte ise katılımcılardan verilen cümledeki boşluğu en iyi tamamlayan sözcüğü seçmeleri istenmiştir. Bu testte katılımcıların farklı tip eşdizim performansını ölçen test maddeleri yer almaktadır (Örn. Fiil+İsim, Sıfat+İsim, Sıfat+Edat ve İsim+Edat). Çalışmada katılımcılar tarafından kullanılan sözcük gruplarının eşdizim olup olmadığına karar vermede Oxford Advance Learner’s Dictionary (OALD), Oxford Collocation Dictionary, ve Longman Active Study Dictionary (LACD) kullanılmıştır. Ayrıca, anadili İngilizce olan bir uzmandan görüş alınmıştır. Katılımcıların kullandığı sözcük gruplarının sınıflandırılmasında Nesselhauf (2003)
sınıflandırmalarından yararlanılmıştır. Katılımcıların İngilizce seviyelerinin kompozisyonlarında kullandıkları eşdizim miktarında istatistiksel olarak önemli bir farka neden olup olmadığını analiz etmek için Kruskal Wallis Testi kullanılmıştır. Analizlerde tamamlayıcı 28
test olarak kullanılan Mann-Whitney U testleri ise katılımcıların oluşturduğu üç gruptan hangi ikisi arasında istatistiksel olarak anlamlı bir fark olduğunu göstermiştir. Ayrıca, katılımcıların çoktan seçmeli testte verdikleri doğru cevaplar yüzdelik olarak hesaplanmıştır. Katılımcıların performansında anadil transferinin etkili olduğu yanıtlar da yine yüzdelik olarak hesaplanmış ve sonuçlar şekillerle sunulmuştur.
Bulgular
eşdizim performansında istatistiksel olarak anlamlı farklılıklar tespit edilmiştir ( H(2)= 24.108, p= .000). Ayrıca, Mann-Whitney U sonuçlarına göre, anadili İngilizce olan katılımcılar (Mdn=14), ileri düzey İnglizceye sahip olan katılımcılardan (Mdn=11.5, U= 7.500, p<.01) ve alt orta düzey İnglizceye sahip olan katılımcılardan (Mdn= 2, U= 0.000, p<.001) istatistiksel olarak daha iyi eşdizim performansı sergilemiştir. Benzer bir şekilde, ileri düzey İngilizceye sahip olan katılımcıların eşdizim perfromansının(Mdn= 11.5) alt orta düzey İngilizceye sahip olan katılımcılardan (Mdn= 2), U= 0.00, p<.001) daha iyi olduğu tespit edilmiştir. Katılımcıların yazdıkları kompozisyonlar incelendiğinde, ileri düzey İngilizceye sahip katılımcıların alt orta düzey İngilizceye sahip olan katılımcılara oranla daha fazla eşdizim kullandığı görülmüştür. Tüm grupların kompozisyonlarında kullandıkları eşdizimler incelendiğinde, anadili İngilizce olan katılımcılar ve İngilizceyi yabancı dil olarak öğrenen katılımcılar arasında hem benzerlik hem de farklılıklar gözlemlenmiştir. Tüm gruplarda katılımcılar Fiil+İsim, Fiil+Edat, Sıfat+Edat, Sıfat+İsim ve Edat+İsim tipinde eşdizim kullanmıştır. Bununla birlikte, tüm gruplarda katılımcılar en çok Fiil+Edat, Sıfat+İsim, Sıfat+Edat ve Edat+İsim türünde eşdizimleri tercih etmişlerdir. Öte yandan, anadili İngilizce olan katılımcıların İngilizceyi yabancı dil olarak öğrenen gruplardan farklı olarak Zarf+Sıfat ve Fiil+Sıfat türünde eşdizimler de kullandıkları görülmüştür. Ayrıca, ileri düzey İngilizceye sahip katılımcıların alt
orta düzey İngilizceye sahip katılımcılardan farklı olarak Fiil+Zarf türünde eşdizim kullandıkları tespit edilmiştir. Yapılan analizler, hem alt orta hem ileri düzey İngilizceye sahip katılımcıların ikinci dilde eşdizim kullanımında anadil Türkçeden aktarım yaptıklarını göstermiştir. Her iki grupta, anadil Türkçenin etkisi İngilizce eşdizim kullanımında en fazla Sıfat+İsim kategorisinde görülmüştür. Fiil+İsim ve edatların oluşturduğu eşdizim kategorilerinde yine alt orta düzey İngilizceye sahip katılımcıların Türkçeden İngilizceye daha fazla aktarım yaptığı görülmüştür. Sonuç Bu çalışma, alanyazında yapılan çalışmalara paralel olarak, öğrencilerin sahip oldukları İngilizce düzeyinin kullandıkları eşdizim miktarını ve çeşidini etkileyen faktörler arasında yer aldığını göstermiştir. Çalışmada ayrıca katılımcıların anadilinin ikinci dilde eşdizim kullanımını etkilediği tespit edilmiştir. Tüm bu sonuçlara göre, yabancı dil öğretiminde oluşturulan kapsamlı programlar, öğrencilerin eşdizim fakındalığını sistematik olarak destekleyen ve onların sözcük dağarcığının vazgeçilmez unsurlarından biri olan eşdizim konusunda pratik yapmalarını sağlayan etkinlik ve materyaller içermelidir. Ayrıca, yabancı dil öğretmenleri anadil ve yabancı dil eşdizim farklılıklarına vurgu yapmalı ve bu konuda öğrencilerde çeşitli materyal ve etkinlikler yoluyla farkındalık oluşturmalıdır. Yabancı dil öğrencilerinin öğrendikleri dilde eşdizim düzeylerini arttırmaları onların ikinci dilde daha akıcı ve doğru cümleler kurmalarını sağlayacaktır.
References
- İngilizce seviyesi alt orta ve ileri olan katılımcılar ne çeşit eşdizimler üretmektedir? 27 Yıldız Journal of Educational Research, 2020, 5(2), 1-30.
Share and Cite
ÖZATA, H. Turkish EFL Learners Use of English Collocations. Yildiz Journal of Educational Research 2020, Vol. 5, pp. 1-30. https://doi.org/10.51280/yjer.2020.006