COLLOCATION REVIEW

REVIEW ARTICLE ABOUT COLLOCATION

Post by        :GÜLAY KOÇ, Department of Teaching English as a foreign language, Bilke University, Ankara

A review by : Agnesia Elvi Wisnita S

Pages           : 127

Title             : Developing Collocational Awareness

The article give us the explanation about Collocation, such as:

This study aimed to investigate to what extent explicit instruction of vocabulary

in collocations, using different techniques, develops collocational awareness in students,

and whether such instruction has any enhancing effect on the retention of vocabulary.

Eight intact groups of 160 EFL students of upper-intermediate proficiency level

under the supervision of their regular course teachers participated in this study. Four of

the groups were assigned as the experimental group and received vocabulary instruction

focusing on collocations, while the remaining four were assigned as the control group

and received instruction concentrating on single words. For this investigation, a

vocabulary retention test , which was administered as the pre-and post-test, three tasks

for the three treatment sessions, transcriptions of verbal processes of one of the

experimental groups, and retrospective interviews with the participant instructors were

used as data collection devices.

The analyses of the qualitative data showed that the participants developed

awareness to the extent that they could identify collocations in any text and categorize

lexical collocations. The analyses of the quantitative data revealed that vocabulary

instruction in collocations yielded far better results in terms of vocabulary retention.

In the light of the findings of this study, explicit instruction of vocabulary in

collocations, using different techniques, is highly recommended for developing

collocational competence and better retention of vocabulary.

Background of Collocation

            The term “collocation”, which means word combinations, such as catch a cold, commit suicide, bitter disappointment, safety belt, was originally introduced by Firth (1951cited in Cowie and Howarth, 1996), directing the attention of  ELT practitioners, theorists, linguists and researchers to the highly significant phenomenon of lexicon. However, the importance of it was realized far later.

           As Zimmerman (1997) states, especially with the introduction of work in the area of corpus analysis, computational linguistics and lexical approaches, a growing number of scholars (e.g., Sinclair, 1991; Nattinger and Decarrico, 1992; Lewis, 1993), representing a significant theoretical and pedagogical  shift from the past with their work, pushed collocations to the center of language acquisition. Today, it has widely been acknowledged that collocations constitute an important part of native speaker competence, and therefore should be integrated into second and foreign language

teaching (e.g., Cowie, 1992; Bahns, 1993; Wei, 1999; Lewis; 2000).

Research on collocations falls into a broad spectrum. Nattinger (1988), Sinclair (1991), Willis (1990) and Lewis (2000), among the pioneers of research on collocation, have described and categorized collocations and produced seminal studies which have contributed considerably to our understanding of lexis.

There have been many published studies evaluating the collocational proficiency of EFL learners from various levels, in order to investigate the correlation between English proficiency and knowledge of collocations. Huang (2001), Bonk (2000), Biskup (1992) and Zughoul and Abdul-Fattah (2003) are among those who developed and administered small-scale elicitation tests and attempted to reflect on the learners’ actual production problems of English collocations. Moreover, some of these studies have contributed to literature revealing what strategies are being used by non-native learners of English when they cannot find the appropiate collocates of words.  

Furthermore, some scholars have offered a distinction between receptive and productive vocabulary skills of EFL/ ESL learners. They have emphasized that receptive knowledge enables students to comprehend word meanings appropriately; however, the productive knowledge entails using a wide variety of ways that words collocate with each other. They have focused on learner errors in production, analysed them and made suggestions towards solutions for minimizing these types of collocational errors (e.g., Pawley & Syder, 1983; Meara,1984; Carter, 1987; Nation, 1990; Wei,1999; Lewis 2000; Nesselhauf, 2003).

Another field, which has been fruitful for collocation studies is closely related to computer technology. As Kita and Ogata (1997) state, rapid advances in computer technology have caused a shift in natural language applications from a knowledge based to a corpus based or data intensive approach. This new trend has highly affected the field of computer assisted language learning/teaching (CALL/CALT). A growing number of researchers have dealt with the problem of collocation from various aspects yielding aid for practitioners and learners (e.g., Kita & Ogata, 1997; Shei & Pain, 2000; Nesselhauf & Tschichold, 2002; Sun & Wang, 2003, among others).

Additionally, lexicographers and linguists have also expanded the spectrum of studies on collocations. Dictionaries are the most important sources of lexical information for learners and instructors. Carter (1987: 157) says, “Dictionaries have a good image” and he notes that almost every learner of a language as a second or foreign language owns one and it is one of the few books retained after following a language course. However, conventional dictionaries are used for decoding- finding the meaning of unknown words- rather than encoding. Since collocations were recognized by many scholars as one of the most significant aspects of lexicon, some researchers have diverted attention to the need of developing more sophisticated  phraseological dictionaries. The BBI Dictionary of English Word Combinations, The LTP Dictionary of Selected Collocations and Longman Essential Activator are some of the products of this period.

In summary, collocations have been researched from various aspects. However, only a few researchers have attempted to develop insights towards the needs of learners and practitioners in EFL classroom settings. The focus of this study is on explicitly raising students’consciousness towards collocations, and providing learners with different techniques for the retention of these word combinations.

Methodology of Collocation

The instruments used in the data collection process included the Vocabulary. Retention Test, its subsections, its piloted version, the tasks and techniques used in the treatments and the materials delivered in the treatments. In addition, classroom sessions were audiotaped and transcribed, and interviews were held with the participant teachers.

1.      The Vocabulary Retention Test as the Pre- and Post-test

As this was an experimental research study with two groups (experimental and control groups) and with three treatment or instruction types for each group, a vocabulary test with three subsections was designed by the researcher and administered before and after the treatments in order to see the preliminary collocational knowledge of the participants and to assess the difference, if any, stemming from the effect of treatment types focusing on collocations in the experimental group. Another reason for the vocabulary retention test to serve as the pre- and post-test was that similar testing procedures have been widely used in experimental studies with the same design as the present one (e.g., Tokaç, 2005).

The test was comprised of 75 collocations, both grammatical and lexical. In the literature, mostly lexical collocations have been emphasized as the source of problems for learners of EFL. However, although grammatical collocations have received more attention than lexical collocations and taught explicitly in most EFL classrooms, students still fail to produce adjectives, verbs and nouns when they form combinations with prepositions because they are not transparent; therefore, they require special attention as much as lexical collocations. For example, Bonk (2000) posits that there is a combination of semantic, syntactic and lexical knowledge in these expressions that make them hard to acquire. Considering these reasons and to prevent the negligence towards this type of collocations, one of the sections in the test was alloted to grammatical collocations and the other two to lexical collocations. In choosing this type of collocations for inclusion, the suggestion made by Woolard (2000) was also taken into consideration. What he suggests is that teachers should keep a record of mis-collocations students make in their production of language and provide a platform for raising awareness of collocations focusing on the selection of their mis-collocations. Therefore, in this section, the researcher used some of the mis-collocations students generally produce, drawing on her experience of teaching students of the same proficiency level.

The first subtest was a simple recognition test in multiple choice format with 25 verb + preposition, noun + preposition, and adjective + preposition type collocations and the participants were asked to choose the best alternative among the choices of prepositions. As the focus of this section was on grammatical collocations and there were only a few alternative prepositions as collocates of the headwords, multiple choice type of testing was found more appropriate due to the fact that it would be more reliable and provide more practicality for later evaluation.

The second subtest was a fill-in-the-blank test adapted by the researcher, leaving the collocates of 25 words blank in two selected passages of an article from Newsweek Magazine, which was also used in the corresponding treatment session. The collocations targeted here were mainly of the lexical type, and to provide more contextual clues for the participants to make closer guesses and to enhance reliability in evaluation they were presented in context. Additionally, this section aimed at more free production of the participants when compared with the first and third subtests because choices were not provided.

The third subtest was a cued fill-in-the-blank test. In this section, a passage on pollution with sentences containing blanks for the collocates of 25 words and the synonyms of the blanked collocates in parantheses at the end of each sentence was used, and the participants were required to provide the missing part of the collocations. The answers of the participants were kept under control by providing as many synonyms of the targeted collocates as possible in parantheses to raise the reliability. This section also mostly concentrated on lexical collocations and except for the last two sentences all of them were adapted from a reading text which was designed to teach collocations and retrieved from a website. (http://esl.about.com/od/vocabularylessonplans/a/chunk.htm).

With regard to validity of the test, the participants were all at the same proficiency level (as revealed on table 2 in chapter 4) and the test was piloted with two classrooms who were at the same level as the participants. Reliability was assured by having two native English teachers check the items in all sections. Additionally, the researcher evaluated the results at different times.

2.      Piloting the Vocabulary Retention Test

The test was piloted two weeks prior to the first treatment session with two classes who were at the same proficiency level as the control and experimental groups but did not take part in the experimental study. The test was reduced to 45 items excluding ten items from each section, which were answered by the majority of the pilot group. Thus the items to be used in the pre-test, three treatment sessions and post-test were determined. (See Appendix A for the final version of the pre- and post-test).

3.      Materials Used in the Treatment Sessions

Three tasks for three treatment sessions were designed, and to ensure unity in classrooms and to inform the participant teachers about how to administer the lessons, a training session was held one day prior to each treatment session. The instructors were also provided with lesson plans, which showed all the steps to be followed in detail. (See Appendix B for the lesson plans.)

The first treatment session, participants of both control and experimental groups were exposed to the same reading text, the article on anorexia nervosa. The participants in the control group read the article twice, worked on the vocabulary unfamiliar to them, answered comprehension questions and summarized the article in their own words. On the other hand, in the experimental classrooms, the participants were first asked to work in groups of four to translate some Turkish sentences, all of which contained targeted collocations, into English in groups of four. The teachers then introduced some information about collocations and asked them to read the text to reshape their actual versions. (See appendix C for the information given about collocations.) The groups worked on their versions after reading the text on anorexia nervosa and corrected their mistakes on collocations. Then, they completed a table drawn by their teachers with types of lexical collocations they encountered in the text. They also summarized the text using the word combinations they worked on.

The second treatment session, a dictaglossing technique was used with a short text on pollution, which was retrieved from a web site. The text was read aloud by the instructors twice or three times and the participants in the control group were asked to collect words while it was being read and work in groups of four to reconstruct the passage using the words they compiled. On the other hand, the participants in the experimental group were asked to collect words during the first round of the reading to reconstruct the passage and their collocates in the second round to reshape their first versions. Their attention to collocations was also ensured by three questions eliciting some of the targeted phrases after the first round of the reading.

The last treatment session, grammatical collocations were focused on and participants in both groups were delivered a handout consisting of fifteen sets of sentences. In each set, there were sentences containing one or two collocational mistakes or none. The participants in the control group were asked to work in pairs to mark and correct the mistakes looking up the headwords in their dictionaries, while those in the experimental group were provided with concordancing extracts for the same procedure. The experimental group participants were also asked to find the most frequent collocates of the headwords looking at the extracts as well as making funny sentences with the collocations they corrected. Thus they had the chance to recycle the target collocations and to solidify their learning.

The treatment sessions were conducted at the same class hour in all classrooms under the supervision of their regular course teachers on subsequent days. As for the time allotted to the treatments, in both groups, it was 100 minutes to the first treatment, and 50 minutes each for the second and third treatment sessions.

Conclussion of Collocation

The findings of this study indicate that students have difficulty in dealing with collocations, which shows consistency with a great body of research conducted previously. As revealed by other studies  (e.g., Martin, 1984; Farghal and Obiedat, 1995; Gitsaki, 1996; Bahns and Eldaw, 1993; Bonk, 2000; Conzett, 2000; Huang, 2001; Nesselhauf, 2003; Zughoul and Abdul-Fattah, 2003), although learners develop knowledge of collocations to some extent, and it increases steadily as the level of proficiency increases, it lags far behind their knowledge of single vocabulary items. In

this study, the analysis of some segments from the verbal processes of the participants has shown that although the students have a good stock of words, since their attention has not been directed to collocations, they have mostly failed to re-combine elements to produce the targeted collocations, and they produced combinations such as, “eating irregularities” instead of  “eating disorders”. Therefore, for teachers of EFL the highlighting of collocations should be as important as teaching vocabulary individually. It seems to be the only way to enhance the active use of language, and to help the learners construct lexically acceptable sentences rather than making sentences full of unnatural-sounding elements or grammatical mistakes with simple vocabulary.

            In a number of previous studies, researchers have investigated and identified the strategies used by the learners of EFL. The analyses of the responses of the participants on the pre-test and the transcriptions of the verbal processes have revealed that similar strategies were employed by the participants of this study. The students either on the pre-test or during the treatments resorted to overgeneralization and analogy, expanding a feature or form to a different contextual use in the target language with phrases such as, “put off weight” “obsessed to” rather than “lose weight”, and “obsessed with”. Literal translation was another mostly relied on strategy, which led to errors of L1 interference. Along this line, most participants used phrases such as “destroy/ damage her body”, “heart crisis”, “drink a drink”, “give weight” instead of “abuse her body” “heart failure”, “take a drink”, and “lose weight”. There were traces of derivativeness as well. The use of “destroyable eating habits” and “seeming bones” instead of  “disastrous eating habits” and “protruding bones” may serve as examples of this strategy. Assumed synonymity and paraphrase and circumlocution were the other two most commonly used strategies. Although the participants knew many synonymous words, they looked for a synonym or a near-synonym, which resulted in production of odd phrases, such as  “voice pollution”, “battle problems”,  “suit the profile”, “earn respectability” instead of “noise pollution “, “combat problems”, “fit the profile “, and “gain respectability”. This can be attributed to the fact that they were not fully aware of the selectional restrictions of these words because of the type of instructional input they had and bilingual or less sophisticated monolingual dictionaries with limited contextual distinctions they relied on. Similarly, most participants in this study widely employed paraphrasing while performing the translation task to convey the intended message with phrases such as “treat her body badly”, “ with a bigger and bigger ambition“, “ to take control of her appetite ” rather than “ abuse her body“, “with growing enthusiasm” and “to keep her appetite in check”, which resulted in grammatical mistakes. Taking into consideration these strategies employed by learners would be beneficial for instructors, textbook developers, and course designers in providing solutions for learners’ problems and facilitating teaching collocations in EFL settings.

            Some researchers who hold the belief that collocations are of great significance for both learners and instructors of EFL find instructors helpless in developing materials and techniques to teach collocations and encouraging student autonomy, and as a result, some have focused on development of dictionaries (e.g., Benson et al., 1997) while others have designed concordancer programs (e.g., Sinclair, 1991). One of the findings related to the use of concordancing extracts and dictionaries in teaching collocations generated consistent results with their beliefs. In the last treatment session of this study, the instructors confronted the experimental group participants with more naturally occuring contexts of the targeted collocations in concordancing extracts. As a result, these students could easily find the frequent collocates of those words and practically

corrected the mis-collocations. However, the participants in the control group, having less sophisticated monolingual dictionaries, could find only a few collocates of the targeted words, had to rely solely on their teachers explanations, and could not achieve so well as the experimental group on the post-test. Therefore, teachers of EFL can guide their students to own dictionaries which present more naturally occurring data or make use of available concordancing extracts.

            Most researchers who have conducted studies on collocations have labeled grammatical collocations less problematic than lexical collocations, since they have been mostly focused on by instructors. However, the findings of this study have revealed that there is no superiority between these two types in terms of difficulty. On the contrary, when explicitly taught, although grammatical collocations are easily learned, they are not as retainable as lexical collocations, presumably because grammatical collocations require more recycling for long-term retention. Therefore, both types should be treated equally in EFL settings but providing students with more recycling exercises or tasks.

            This study has revealed that explicit instruction of vocabulary in collocations with different techniques develops awareness in students towards collocations to the extent that they can recognize collocations and categorize them, ask questions related to collocations, find collocates for the headwords in the texts they encounter in their actual courses, as well as finding collocates for the headwords in these texts to expand their vocabulary. It has also showed that students, when taught vocabulary in collocations, retain more words than they do when they are taught vocabulary as isolated items. Additionally, in this study, the three techniques used in the treatment session were compared and the translation technique was found more effective that is, the words taught with translation technique were retained better than those taught with dictaglossing and error correction with concordancing ex