Cross-linguistic Influence: Its Impact on L2 English Collocation Production

This research study investigated the influence of learners’ mother tongue on their acquisition of English collocations. Having drawn the linguistic data from two groups of Thai EFL learners differing in English proficiency level, the researcher found that the native language (L1) plays a significant role in the participants’ collocation learning as it is regarded as a primary learning strategy on which they depend. Such L1 transfer seems to lead the learners to a number of collocational problems in the target language (TL), e.g. preposition omission, preposition insertion, non-targetlike word choice, and collocate redundancy. It was discovered that high-proficiency learners relied heavily upon their L1, a bahavior claimed to be characteristic of those with limited TL knowledge (Ellis, 2008; Odlin, 1989). That the L1 evidently has a great impact on high-proficiency learners’ use of English collocations found support for some past studies (e.g. Boonyasaquan, 2006; Koya, 2003).


Introduction
Collocation has become a challenging issue not only for L2 learners but also for language researchers (Schmitt, 2010).To successfully acquire vocabulary in a target language (TL), In addition to mastering a good number of useful words and their principal meanings, learners also need to be capable of possible and likely word combinations, referred to as collocation (Firth, 1957).
Collocation can be defined as "items that occur physically together or have stronger chances of being mentioned together" (Sinclair, 1991, p. 170).In a similar vein, Lewis (1997) gives a definition of collocation as "the readily observable phenomenon whereby certain words co-occur in natural text with greater than random frequency" (p.8).Collocation, in other words, exists in statistically significant manners (Lewis, 2000).Furthermore, Hill (2000) observes that collocation deals with word combinations that are predictable.For instance, when foot is used as a verb, it is highly likely that a following collocate, i.e. a word that has a strong tendency to co-occur with another, is the noun bill, as in foot the bill.Hill (2000), moreover, maintained that some collocations, known as strong collocations, are fixed or not much generative in the sense that they allow a very limited number of collocates.A clear example of strong collocations is the adjective rancid, which exclusively co-selects a very small group of collocates, e.g.butter or oil.Conversely, there appear certain word combinations that are considered weak collocations because a great number of words can serve as their word partners (Hill, 2000).For example, the adjective nice can obviously co-occur with any number of nouns, e.g. a nice man, a nice condominium, a nice dog, a nice hairstyle, a nice car, etc.
The present study was aimed at an examination of how learners' native language can influence their English collocation learning, with an emphasis on the actual problems caused by an application of L1 transfer as a major learning strategy (Selinker, 1992).

Significance of Collocations for EFL Learners
It is undeniable that learners of English as a foreign language (EFL), for example Thais, require solid collocational knowledge of English in order to effectively utilize vocabulary in the TL (Phoocharoensil, 2010;Schmitt, 2000).Hill (2000) suggested some reasons why collocations should be emphasized in an EFL classroom.
Firstly, vocabulary choice is often predictable in such a way that a word uttered by a speaker could make the listener expect a range of possible word partners.For instance, the verb drink is supposed to be followed by a drinkable kind of liquid, e.g.water, milk, lemonade, etc. rather than inedible liquid, e.g.engine oil, or even solid food, e.g.bread, rice, meat, etc.For this reason, an awareness of collocate predictability should be raised for EFL students.
Secondly, collocations in English are of enormous size, taking up almost 70% of what is really spoken and written in authentic English.Consequently, collocation teaching seems to benefit EFL learners to a great extent.
Another reason why collocations are viewed as important concerns its role as a facilitator of the human thinking process, which allows more successful, fluent communication.According to Hill (2000) and Wray (2000), whereas native English speakers recognize multi-word units with ease, EFL learners with limited exposure to L2 English often process L2 word-by-word owing to a lack of a vast repertoire of ready-made or formulaic language like collocations.Thus, adequate collocational knowledge seems to substantially expedite learners' ability to process L2 (Moon, 1997).This idea gains support from Kuiper, Columbus & Schmitt (2009), who remarked that formulaic sequences offer processing efficiency since single memorized units composed of a string of words are processed more quickly and easily than word sequences generated creatively.In a similar way, Conklin and Schmitt (2008) indicated an advantage in reading by non-native speakers who read formulaic sequences faster than non-formulaic ones.
In addition to the above reasons, collocations are very useful to advanced EFL students since knowledge of chunks or prefabricated units (Wray, 2002) simply enables these learners to naturally and accurately express complex ideas, as stated by Hill (2000), "The more exposure students have to good quality input and the more awareness they develop of the lexical nature of language, the more they will recognize and eventually produce chunks themselves."(p.55).
The subsequent subsection pertains to the impact of learners' native tongue on their acquisition of English collocations.
Evidence of negative transfer from L1 is available in several past studies.Bisk-up (1992) examined Polish and German EFL learners' performance in English collocations.The learners in this study apparently produced L2 collocations based on risk taking.That is, they seemed to try using L2 combinations, not being afraid of whether their own creations constituted well-formed, acceptable collocations in the TL.They often transferred from L1 collocations, which often resulted in errors in the TL.An example of Polish transfer errors is *to state a record instead of the targetlike one to set a record.As for the German speakers, they seemed to use erroneous L2 collocations reflecting L1 counterparts, e.g.*to lend a bookshop, as opposed to the standard version to run a bookshop.
Like Bisk-up (1992), Bahns (1993) and Bahns & Eldaw (1993) revealed native language interference in German speakers' English collocation production through translation.Those whose L2 collocational competence was limited evidently resorted to German, which was a major source of the errors found.Similarly, according to Huang (2001)'s study of collocational competence of Taiwanese EFL university students, the participants, asked to do a sentence completion test, frequently produced collocational deviations related to L1.For example, they used *a black horse when meaning a dark horse or *to eat a bite in lieu of to take a bite.Nevertheless, Huang also discovered positive transfer when an English collocation has an exact Chinese equivalent, e.g. a paper tiger.Nesselhauf (2003), consistent with all the aforementioned studies, has shown that L1 influence on the production of English collocations by German speakers is remarkably high.Overall, the collocational problems seemed to emanate from L1 transfer.Nesselhauf maintains that the degree of difficulty of collocational learning is associated with the degree of congruence.More precisely, congruent collocations, i.e. combinations which sound natural in both L1 and L2 if rendered word for word, apparently cause less difficulty than non-congruent counterparts, which could lead to L1-based errors.According to Nesselhauf (2003Nesselhauf ( , 2005)), the role of L1 as to collocation should not be downplayed in foreign language teaching and learning as it is L1 that seems responsible for collocational problems in TL.
One of the very interesting works on cross-linguistic influence and collocation acquisition is Koya (2003)' study on Japanese learners' receptive and productive knowledge of English collocations, the main focus of which concerns noun-verb combinations.Despite the fact that low-proficiency learners are generally found to rely heavily on L1 knowledge in the process of L2 acquisition, Koya surprisingly found that it was high-proficiency learners who relied more on L1 transfer, while those with lower proficiency appeared to adopt an avoidance strategy, refraining from providing an answer to almost all the questions having no direct translation from L1 Japanese.
Transfer from native language was also discovered in Thai EFL learners' use of collocations.Having analyzed English majors' collocational violations in their translation of business articles from Thai to English, Boonyasaquan (2006) concluded that the learners' L1 was one of the most dominant contributing factors in the collocational error production.A clear example of interference errors is *expensive price, which is presumably derived from direct translation from Thai.
In support of Boonyasaquan (2006), Mongkolchai (2008) also showed that Thai EFL university students majoring in English seemed to use lexical and grammatical collocations based on L1 Thai, thus ending up producing erroneous combinations in TL English.For instance, with respect to grammatical collocations, as in the sentence *I became skillful in (rather than at) drawing, the use of the preposition in instead of at is considered ungrammatical, reflecting the prepositional pattern in Thai (Mongkolchai, 2008, p. 32).Moreover, the learners also had problems with lexical collocations, as can be seen in *She has built a well-known reputation as an entertaining speaker.(Mongkolchai, 2008, p. 32).The co-occurrence of the adjective well-known and the modified noun reputation is redundant and unacceptable in English.The combination as such, however, is allowed in Thai.Hence, the learners might transfer such a collocation from Thai into English.
L1 transfer was also very noticeable in Fan's study (2009), which examined the use of collocations in writing by Hong Kong ESL learners.It demonstrated that the production of collocations was adversely affected by L1 Chinese.There occurred some combinations that seemed to stem from word-for-word translation of Chinese collocations which are unacceptable or non-existent in the native corpus reference.These deviant collocations are claimed to result from L1 influence.For example, the non-standard collocation like *circle-eye, as opposed to round-eye, could be attributed to the fact that the concepts of round and circle are expressed by the same lexical item in Chinese.
Juan and Ge-ling ( 2009) lent support to Fan (2009) in that their research revealed interlingual errors on collocations, i.e. errors caused by L1-L2 differences.Like Boonyasaquan (2006), Juan and Ge-ling reported similar transfer errors, e.g. the incorrect use of the adjective expensive to modify the noun price, as in *I am afraid that the price of the food is a little expensive (Juan & Ge-ling, 2009, p. 14).An additional example of interference errors is the verb + noun collocation happen + traffic accident, as in During the trip, you may happen traffic accident (Juan and Ge-ling, 2009, p. 15).As claimed by Juan and Ge-ling, the use of expensive to modify price and the use of happen with traffic accident reflect transfer from Chinese in that the students probably equated these collocations with Chinese ones, committing collocational errors in TL.Ying (2009), furthermore, studying collocations in English writing by Chinese learners who are English majors and non-English majors, has borne out Nesselhauf's (2003) finding that collocations which have no translation equivalents in L1 are apparently more difficult for learners to master, compared to those which are congruent or translatable from L1 to L2.As predicted, the total number of errors produced by non-English majors was far greater than that committed by English majors.More detailed analysis revealed that, for both groups of learners, errors in lexical collocations obviously outnumbered those in grammatical collocations.The most common type of collocational errors for non-English majors was a grammatical one: preposition + noun, whereas English majors seemed to produce the highest number of errors on the verb + noun combination, which is a type of lexical collocation.According to Ying, one of the important reasons for collocational errors in this study is an overgeneralization or lexical transfer from L1.In using English collocations, the learners probably searched for L1 equivalents without any awareness of L1-L2 incongruity which may result in deviations from the target language.
In addition, Laufer and Waldman (2011) examined the use of English verb-noun collocations in the writing of Hebrew learners of English at three proficiency levels.They interestingly discovered strong evidence of L1 influence on their collocation use in L2.In particular, around half of the erroneous collocations produced by the learners at all proficiency levels seem to originate from native language transfer and are clearly recurrent.These interlingual errors, in a nutshell, do not tend to decrease over time.They proposed that, unlike native English speakers, learners construct messages from individual words rather than from prefabricated patterns, with dependence on L1 transfer and tendency to disregard restrictions on word combinations (Laufer & Walderman, 2011, p. 665).
A recent study by Yumanee (2012), who investigated Thai secondary-school students' use of collocations, indicates cross-linguistic influence as well.Through a multiple-choice collocation test and a translation test, the results revealed that the performance of Thai EFL students in both the receptive test and the productive test appeared to be influenced by mother-tongue transfer when confronted with difficulties in dealing with English collocations.Moreover, it was discovered that there existed other potential factors that contributed to the high degree of collocational errors, such as the synonymy strategy, the learners' creative invention and the strategy of analogy, the paraphrasing strategy, and low knowledge of collocational skills.

Research Question and Hypothesis
The present research was aimed at answering the question below: Research question: What are the deviant collocations influenced by Thai EFL learners' native language?
In response to the above research question, the hypotheses below were therefore formulated: Hypothesis: Thai EFL learners' mother tongue has a negative impact upon their English collocation acquisition.

Participants
The subjects of the study were 90 freshman Thai EFL students at Thammasat University, Thailand.They were of two proficiency groups of equal numbers.The students from a variety of faculties were assigned to different classes based on their University Entrance Exam English scores.With regard to the cut-off scores, the participants whose O-NET scores were between 69-80 studied Foundation English Course 3, i.e. the high group, whereas the students whose scores were between 32 and 56 were assigned to English Foundation Course 2, i.e. the low group.The score ranges above were determined by the Registration Office of the university for the purpose of classroom arrangement according to English proficiency level.The participants' first language is Thai, and they had learned English as a foreign language (EFL) for at least 12 years.

Research Instrument
In order to draw real linguistic data from the participants, a timed (60 minutes) in-class written essay was selected as the elicitation method.The two groups of subjects were asked to write a descriptive essay, which should effectively elicit a number of different types of collocation as this written task contains a wealth of modifications usually formed by word combinations (Langan, 2007).The researcher also served as the teacher to facilitate the data collection procedure in the hope that the participants would cooperate well to complete the task because they should be more acquainted with their teacher than a researcher in general.The subjects were also well aware that this written task would not be scored according to grammatical correctness.Thus, they should feel comfortable and relaxed enough to produce L2 English combinations which best reflect their actual ability to use collocations in English.

Data Analysis
The data were then analyzed in quantitative and qualitative ways.That is, the researcher analyzed the data using the collocational framework of Benson et al. (1987), who have proposed two major types of collocations, i.e. lexical and grammatical collocations.To be more precise, the present study concentrated on three types of lexical collocations, i.e. noun + verb, verb + noun, adjective + noun.With respect to grammatical collocations, four types were investigated: noun + preposition, verb + preposition, adjective + preposition, and preposition + noun.
Only the deviant combinations were examined.In addition, a combination would be labeled as incorrect or erroneous on the condition that it does not follow at least one of the following: 1.) four collocation dictionaries, i.e.The BBI Dictionary of English Word Combinations (1987), The LTP Dictionary of Selected Collocations (1997), Oxford Collocations Dictionary (2009), and Macmillan Collocations Dictionary (2010); 2.) the British National Corpus (BNC), which is a 100 million word collection of samples of written and spoken English from a broad range of sources, created to represent a wide cross-section of British English from the later part of the 20 th century (http://www.natcorp.ox.ac.uk/); and 3.) two native-speaker EFL teachers, one of whom speaks British English and the other represents American English.
The entire number of deviant collocations based on L1 transfer was counted and then compared across both proficiency groups.Next, the errors are explained in relation to the equivalents in Thai.

Cross-linguistic Influence in Collocation Use
Table 1 reveals that not only did low-proficiency students depend on their L1 collocational knowledge ending up using L2 deviant combinations (46.55%), but surprisingly those with high proficiency level also exhibited their heavy reliance on L1 collocations, despite their production of fewer errors (46.55%).It seemed that both groups produced erroneous combinations negatively impacted by L1 interference, which finds support for Koya (2003).
In the present study, the collocational problems apparently concern both lexical (High: 44; Low: 53).and grammatical collocations (High: 37; Low: 40), with higher token numbers of the former type in both proficiency groups.

Collocational Errors Involving Prepositions
This type of collocational errors is associated with preposition omission, preposition addition, and incorrect use of preposition.The first type of problem probably occurs as a result of the differences in transitivity between English and Thai.That is, while some verbs in Thai are transitive, i.e. requiring a direct object, some of their English counterparts are intransitive verbs, which do not need any object but a prepositional phrase (Panthumetha, 1982).An error can be committed once an obligatory preposition in English is omitted.The following instances illustrate how an omission of preposition causes a deviation in L2 English.
In (1), the verb  'listen' in Thai subcategorizes for a direct object, whereas the English counterpart listen does not.Instead, listen requires that the following construction be a prepositional phrase beginning with to.Thai students may use listen, based on their L1 subcategorization knowledge, resulting in an ungrammatical structure, like (1), in the target language.
(2) *He always takes care me when I need help.
The deviant collocation in (2) perhaps stems from a lack of the preposition of as part of the verb phrase take care of me.In Thai, duul 'take care of' is a transitive verb.It is probable that Thai students transfer the transitive feature of duul from Thai to English and then omit the preposition of in L2 English.
The problem in ( 3) is caused by an absence of the preposition at or in right after the verb stay, which is an intransitive verb requiring a locative prepositional phrase introduced by at or in (Longman dictionary of contemporary English, 2009English, , p. 1723)).In contrast, it is common in Thai to use the verb phrase yu u h  'stay at a dormitory' (yu u 'stay' and h  'dormitory').Put simply, the verb yu u in Thai is transitive, so Thai learners probably transfer the transitive pattern of this verb to English, leading to the miscollocation in (3).
In addition to the preposition deletion, Thai EFL students also appear to employ L2 English collocations on the basis of those in L1, inserting an extra preposition to a verb in L2.More precisely, where a verb in Thai subcategorizes for a prepositional phrase, its English counterpart sometimes does not need any preposition to follow.This contrast may explain why some Thai learners of English produce a collocational error by adding an unnecessary preposition right after a verb.
(4) *I never left from my home.
The errors in (4) emanate from an addition of the preposition from.In particular, the transitive verb leave in English usually precedes an object; the existence of the preposition from between left and the object my home in (4) constitutes a deviant verb-phrase construction in English.
(7) *When I arrived at there, I saw my old friend who standing alone.
It is worth noticing that the collocational deviations in ( 5)-( 7) result from an occurrence of the unnecessary preposition at.As a matter of fact, the word there in this context functions as an adverb modifying the preceding verb.There is actually no need for any preposition to be positioned between the verb and there.This problem is probably ascribed to the students' native language influence, i.e.Thai.In Thai, the preposition thi  i 'at' in the prepositional phrase thi  i na n 'at that place' or 'there' is generally used to refer to a place, as in thi  i ba an 'at home' and thi  i rorian 'at school', etc.In short, Thai speakers usually use thi  i to refer to a place and this may account for why they are likely to transfer thi  i to its equivalent at in English.Unfortunately, this strategy does not work with the adverb there since the co-existence of at and there is unacceptable in standard English.
The other type of collocational problem regarding prepositions connected with learners' L1 concerns a wrong choice of preposition.First of all, an incorrect preposition is found with a group of adjectives, as shown in ( 8)-( 12).
(8) *When you're tired from working and studying, you may need to take a rest.
(9) *When I'm tired with life and all the problems, I normally consult my close friend.
(10) *My brother can easily get bored from reading.
(11) *I am really impressed in that special time.
(12) *I am close with paddy field and farmers' lives.
In (8), the adjective tired is an equivalent to nya y 'tired' in Thai, which is usually followed by the preposition ca ag 'from' or ka b 'with', as in ( 13) and ( 14) respectively.
( Possibly Thai learners dealing with the adjective tired could transfer the use of ca ag 'from' and ka b 'with' to their English collocation production, causing an ungrammatical construction in L2 English.Likewise, the students may depend on Thai when using *bored from in (10), as opposed to the correct combination bored with, perhaps because the word bya (na y) 'bored' in Thai can co-occur with ca ag 'from'.The incorrect preposition from in (10) may therefore originate from a direct translation, whereas standard English allows only with.
According to (11), the error arises because the preposition in is employed instead of by or with.Such an erroneous use of preposition may also be attributed to L1 transfer.In Thai, the word pra tha bcaj 'impressed', apart from being followed by a noun phrase, can subcategorize for the preposition naj 'in'.
As regards ( 12), the preposition with should be changed to to in order for the sentence to become grammatical.The use of with is perhaps associated with an interference from Thai.In Thai, the word kla jchi t 'close' almost always collocates with the preposition ka b 'with', and this may be the reason why the use of with is present here.

Collocational errors involving nouns and verbs
The data have also demonstrated deviations in relation to the use of a noun and a verb that do not have collocability in the target language.It seems that the erroneous combination production is a reflection of the students' mother tongue.
(16) *Her house stayed at Roi-Et same with me.
In ( 15) and ( 16), the problem apparently arises from the miscollocations of the noun home/house and the verb stay.Such non-standard combinations are presumably influenced by L1 Thai, where the collocation as such is very common.
It is discovered that the learners whose English proficiency as well as exposure is relatively low may have a tendency to use stay with house/home as they heavily rely on L1 collocational knowledge, not knowing a proper passive construction like be located.This is not unusual since passive voice occurs in a limited fashion in Thai, i.e. in a negative sense (Panthumetha, 1982).Furthermore, passive voice is also considered a difficult grammar point to be acquired by EFL learners in general (Celce-Murcia & Larsen-Freeman, 1999;Cowan, 2008).It appears very likely that Thai learners with low L2 collocational competence may resort to the subject-verb-object word order in L1, thereby ending up with an erroneous collocation like *house/home stays.
In addition, Thai EFL learners are also found to create verb + noun combinations that probably reflect L1 collocations, as in ( 17).
The collocation move many places in ( 17) is non-standard.The correct one should be the fixed phrase move house or move home, generally prevalent in British English (Longman dictionary of contemporary English, 2009, p. 1141), as in (18).Another possibility is to use move as an intransitive verb, as in ( 19).
(18) My parents kept moving house because of my dad's job.
(Longman dictionary of contemporary English, 2009, p. 1141) (19) We've moved seven or eight times in the last five years.
(Longman dictionary of contemporary English, 2009, p. 1141) Nonetheless, Thai speakers permit the verb phrase ya j thi i 'move', as illustrated in (20), and this may be a reason why Thai students depend on L1 collocational knowledge in the production of the incorrect combination move places, as in ( 17).
(20) khr bkhrua cha n ya j la aj thi i family my move many place 'My family moved for several times.'

Collocational Errors Involving Adjectives
Although the errors on collocations in which an adjective is a component do not seem to trouble Thai learners as seriously as the others previously discussed, an analysis of these combinations has revealed some interesting co-occurrences of incongruent word partners.
(22) * The dwelling place of my uncle is in Korat.
The collocations in ( 21) and ( 22) are understood in English; however, they are different from what is really used in native speaker corpora, e.g.BNC, thereby being considered deviant.To be more precise, the collocation *working room in (21) should be replaced by office.The former appears to be caused by cross-linguistic influence from Thai.In other words, the sequence h  thamaan (h  'room' and thamaan 'work') in Thai may be literally translated into *working room.
In a similar way, dwelling place in ( 22) is considered unnatural in standard English as the noun dwelling alone refers to a place where people live (Longman dictionary of contemporary English, 2009, p. 529).The production of the word place together with dwelling is apparently viewed as redundant.Again, the miscollocation dwelling place should originate from L1 transfer, being translated from the combination thi  i yu u (thi  i 'place' and yu u 'dwell') in Thai.

Pedagogical Implications
The main findings of this research study demonstrate the authentic collocational errors produced as a consequence of Thai EFL learners' application of native language transfer.Not merely does such a learner strategy result in Thai students' problems, but also other EFL learners speaking other native languages frequently seem to suffer from negative transfer in L2 English collocation acquisition, as indicated in several past studies (e.g.Bahns & Eldaw, 1993;Fan, 2009;Huang, 2001;Koya, 2003;Laufer and Waldman, 2011;Mongkolchai, 2008;Nesselhauf, 2003Nesselhauf, , 2005)).Accordingly, teachers of English for speakers of other languages are highly likely to benefit from the present-study findings in such a way that they should be able to prepare vocabulary lessons and materials with more emphasis on the concept of collocation, as this is globally accepted as an essential element in EFL vocabulary curricula (Nation, 2001(Nation, , 2008;;Schmitt, 2010).
Teachers, moreover, who speak the same L1 as the students, e.g.Thai, are advised to supply students with a list of common, frequent English collocations that are incongruent with L1 counterparts, which can hopefully raise their awareness of a mismatch between L1-L2 collocations (Nesselhauf, 2003).This means students should be informed of the fact that transfer from L1 for several times leads them to deviant combinations in the target language.As suggested by Nation (2008, p. 121), a L2 multi-word unit which has no L1 equivalent needs "to be met and remembered", while deliberate learning is unnecessary where there appears a perfect match between L1 and L2 formulaic sequences.In a similar vein, it has been revealed that a brief explanation of L1-L2 differences in specific collocations and suitable translation practice of such collocations proves far more effective than other instructional methods that ignore the crosslinguistic differences (Laufer & Girsaw, 2008).

Conclusion
To sum up, it has emerged that native language transfer is evidently a very common strategy Thai students adopt when they produce collocations in L2 English.The learners, on the whole, apparently depend on L1 equivalents when they are unable to find the appropriate lexical items, i.e. nouns, verbs, adjectives, or the grammatical items, e.g.preposition, in the target language.Their great reliance on L1 often causes collocational errors in L2.It is also worth noting that not only low-proficiency learners but also those with high proficiency levels rely on transfer from L1 collocational knowledge.This finds support for Koya (2003), who also discovered that even high-proficiency Japanese EFL learners resorted to L1 in selecting possible collocations for L2.Such a finding also bears out Laufer & Waldman (2011), who indicated intermediate and even advanced English learners' dependence on L1 transfer, accounting for approximately half of the collocational errors.
As clearly shown in this article, the learners' native language plays a key role in their L2 collocational acquisition.In other words, most of the students' collocational errors are attributed to L1 transfer.The findings of the present study are in consistent with those of a large number of previous studies (e.g.Bahns, 1993;Bahns & Eldaw, 1993;Fan, 2009;Huang, 2001;Koya, 2003;Nesselhauf, 2003Nesselhauf, , 2005;;Ying, 2009).Moreover, the research findings of this project also gives support to studies on Thai EFL learners which indicated strong evidence of L1 interference (e.g.Boonyasaquan, 2006;Mongkolchai, 2008;Yumanee, 2012).

Table 1 .
The types and number of L1-based (transfer) collocational errors