Productive and Receptive Collocational Knowledge of Iranian EFL Learners at Different Proficiency Levels

In the present study, an attempt was made to probe into the probable difference between Iranian intermediate and advanced EFL learners’ receptive and productive collocational knowledge. To this end, 60 EFL learners studying at Islamic Azad University, Isfahan Branch, including 30 advanced and 30 intermediate learners, were chosen through the Oxford Placement Test (OPT). The participants at each level of proficiency received two tests of collocations, namely receptive collocation test and productive test of collocations. Paired-samples t test showed no statistically significant difference between productive and receptive knowledge of collocations of the advanced EFL learners. However, the mean comparison between the receptive and productive collocation test scores of intermediate EFL learners revealed a significant difference. Pedagogical implications emanating from the obtained results are elaborated in the study.


INTRODUCTION
The origin of the term collocation was the Latin verb collocare, meaning to arrange.However, Firth (1957) who is known as the father of collocations first introduced this term to refer to "the company that words keep" (p.183).Firth believed that it is essential to know what words come with words.Benson (1986) states that collocations are a subbranch of formulaic language, and many language researchers have paid attention to the acquisition of collocations.
According to Sadat Kiaee, Heravi Moghaddam, and Moheb Hosseini (2013), collocations are "words that 'fit together' intuitively with great expectation in the syntagmatic and paradigmatic levels.The syntagmatic relation of lexical words, which is horizontal, refers to the collocability of words" (p.2).In terms of paradigmatic, connections refers to sets of words in the same class.For instance, the word 'dog' is in syntagmatic relation with 'hairy' and in paradigmatic relation with 'cat'.Collocations are predictable patterns and phrases or groups of words that typically co-occur.They include what have traditionally been considered lexical items, as well as structural patterns which may seem closer to grammar and combinations of words that simply 'go together.' In addition, McCarthy (1990) believes that "in vocabulary teaching there is a high importance of collocation."He also suggests that "the relationship of collocation is fundamental in the study of vocabulary, and collocation is an important organizing principle in the vocabulary of any language" (p.12).Furthermore, learning collocations is essential for language learners because collocations can be used both in oral and written language (Lennon, 1996).Considering the importance of collocations, lexicographers take into consideration that collocations should be completely explained to L2 learners because a little knowledge of these vocabulary items can be dangerous to speech and writing production.
There are three approaches about the phenomenon of collocation.The first approach, which is the lexical approach, believes that lexis is a separate issue, which is distinct from grammar.The second approach is the semantic approach, which, similar to the lexical view, overlooks grammar, but emphasizes the semantic aspects of the words that control their meaning.The last approach is the structural approach, which emphasizes the importance of both lexis and grammar in the study of collocations.In addition, so far just a few number of collocations have been studied by the researchers from different perspectives.Therefore, just some limited results have been gained.Contrary to this, more patterns of collocations have been studied by the structural approach of collocation.The importance of collocations has been the focus of a number of studies in the field of language learning.According to Brown (1974), learning collocations improves the learners' language skills and sub skills.
Accordingly, the present paper aimed to compare Iranian EFL learners' receptive collocational knowledge and productive collocational knowledge at advanced and intermediate levels.In fact, the research questions of the present study were: (1) Is there a significant difference between the 12 IJALEL 6(7): [11][12][13][14][15][16] Persian-speaking advanced EFL learners' productive and receptive knowledge of collocations?and (2) Is there a significant difference between the Persian-speaking intermediate EFL learners' productive and receptive knowledge of collocations?

LITERATURE REVIEW
As mentioned, three major approaches have studied collocations.The lexical approach, which is the oldest one and was developed by Firth (1951), believes that collocation is an "abstraction at the syntagmatic level" and is not directly linked to the "conceptual or idea approach to the meaning of words " (p. 196).This framework was adjusted by Halliday (1966) and Sinclair (1991).The second view toward collocation was the semantic approach, which focuses the form of collocations.Other issues were also studied under this approach like the issue of "why words collocate with certain other words, and how the meaning of a word is reduced to its ultimate contrastive elements resulting in the atomization of meaning" (Bahns, 1993, p. 175).
The third approach of collocations is the structural approach.According to this approach, the structural patterns govern collocations.There are some contradictions between the grammatical outlook and the two aforementioned approaches.The difference is that this approach mainly focuses on grammatical and lexical structure (Gitsaki, 1999).Lexis cannot be separated from grammar, because the two are distinctive but related aspects of one phenomenon (Bahns, 1993).Kjellmer (1990) stated that articles, prepositions, and the base form of verbs are collocational.In contradiction, adjectives, singular proper nouns, and adverbs are not collocational in nature.Gitsaki (1996) through the analysis of collocations identified 37 categories of collocation overall: 8 lexical and 29 grammatical.
Some empirical studies have so far been conducted in the field of collocations.Dechert and Lennon (1989) came to the conclusion that advanced English learners who had an experience of at least ten years living with native speakers could not speak and write like native speakers.Furthermore, their production caused misunderstanding and interrupted comprehension.Dechert and Lennon maintain that the errors made by the subjects are not mainly grammatical, but lexical ones.
In another study, Bahns and Eldaw (1993) studied advanced EFL learners' productive knowledge of English ver-b+noun collocations.The participants were classified into two groups.One group took a cloze test containing 10 sentences, each of which had a verb+noun collocation in which the verb was missed.The other group received a translation test in which they were supposed to translate 15 sentences, each made up by a verb+noun collocation in a sentence.The results showed that around 50% of learners' responses were acceptable English collocations.Finally, Bahns and Eldaw concluded that "collocation is problematic, even for advanced students" (1993, p. 102).
Similarly, Gitsaki (1999) intended to measure post-beginner, intermediate, and post-intermediate ESL learners' knowledge of collocations.Three tasks were employed including essay writing, translation, and fill-in-the-blank.The results showed a positive correlation between proficiency and the knowledge of collocation.It was found that frequency of collocations lead to better learning of collocations.
In another study, Nesselhauf (2005) investigated the use of verb/noun collocations among advanced German learners of English in free writing.It was found that production of collocations is affected by their L1.It was also shown that the most frequent error was the wrong choice of the verb.
In the Iranian EFL context, as far as learners' general proficiency is concerned, Koosha and Jafarpour (2006) studied the collocational proficiency of prepositions across various levels of EFL proficiency.In addition, they studied the influence of EFL learners' L1 on their collocational proficiency of prepositions.Two hundred EFL learners were chosen through an English language proficiency test.Two completion tests of collocations were utilized.The results showed that EFL learners' performance in the test of collocation had a positive correlation with their general language proficiency.Finally, it was shown that Iranian EFL learners transferred their L1 collocational patterns to their L2 production.
In the same vein as the above studies, Bagherzadeh, Hosseini, and Akbarian ( 2007) studied the relationship between collocational competence and general language proficiency among thirty Iranian EFL learners.The results showed that there was a relationship between the collocation test and TOEFL and between the vocabulary section of TOEFL and the collocation test.
In another study, Keshavarz and Salimi (2007) employed open-ended, multiple choice cloze tests, and TOEFL to measure collocational competence and language proficiency of one hundred Iranian students.A test of collocation was used.The results showed a significant relationship between the results of the cloze tests and collocational competence.
Similarly, Shokouhi and Mirsalari (2010) studied the relationship between collocational proficiency and general linguistic proficiency of EFL learners.A 90-item multiple-choice test was run among thirty-five subjects.The results revealed no significant correlation between the general linguistic proficiency and collocational proficiency of EFL learners.Lexical collocations were found to be easier than grammatical collocations.
Along with studies on the impact of language proficiency on collocation knowledge, the colocations have been studied in other fields, too.For instance, contrary to the above researches, Bazzaz and Samad (2011) investigated the effects of collocational proficiency on the use of verb-noun collocations in writing.The participants were twenty-seven Iranian PhD students in a Malaysian university.The number of collocations that the students used in their essays were calculated.The results showed a positive relationship between proficiency of collocations and the use of verb-noun collocations in the stories.
In addition, Bahardoust (2012) studied lexical collocations in L1 and L2.Midterm and final tests were used as sources of data.The results showed that the rate of verbnoun and adjective-noun were higher than other collocation 13 Productive and Receptive Collocational Knowledge of Iranian EFL Learners at Different Proficiency Levels types, while the rate of noun-verb was lower.It was shown that L1 collocations had a higher rate and frequency, and L1 had both positive and negative effects on collocations.
The above-mentioned studies in Iran investigated collocational proficiency of EFL learners from different points of view; however, some contradictions are observed because some studies indicated that collocational proficiency increases along with improvement in language proficiency.Other studies showed that language proficiency has no effect on collocational proficiency like the study by Shokouhi and Mirsalari (2010).Some other studies worked on language transfer from L1 and came to the conclusion that negative transfer can cause a problem (e.g.Koosha & Jafarpour, 2006).
In addition to the aforesaid studies, Nesselhauf (2005) investigated the relationship between language proficiency and collocational proficiency.Nesselhauf (2005) studied the use of verb-noun collocations by advanced German learners of English in free writing.He showed there was a correlation between language proficiency and collocational proficiency.
Similarly, Shehata (2008) studied the use of collocations by advanced Arabic-speaking learners.Two production tests and one reception test were used.The results proved the strong influence of learners L1 and the language learning environment on learning collocations.In addition, the results revealed that students' productive proficiency of collocations was less than their receptive proficiency of collocations.
Thus, the review of literature shows that not much study has so far been conducted on the relationship of language proficiency and receptive versus productive knowledge of collocations across different levels of proficiency; therefore, to fill the existing gap, the present research was set out to investigate the relationship between language proficiency and receptive versus productive knowledge of collocations among Iranian intermediate and advanced EFL learners.

Design of the Study
The present study had an ex post facto design (alternatively called causal-comparative design) since quantitative data were collected and analyzed from two groups of learners while no treatment or intervention whatsoever was carried out on them.In fact, the learners at both levels of proficiency took a receptive test of collocation as well as a productive collocation test and the difference between the two sets of test scores was investigated for each proficiency level.

Participants
The participants in this study were 60 learners majoring in English at the English Department at Islamic Azad University, Isfahan Branch.They were 30 available MA students who were considered to be advanced EFL learners and 30 BA students who were coded as intermediate EFL learners.The participants' mean age was calculated and it was 24.56 years.These participants were both male and female students.Their L1 background was Persian (Farsi).In or-der to assure the homogeneity of the participants in terms of their general proficiency at each level of proficiency, an Oxford Placement Test (OPT) was employed.

Instruments
In order to establish the general language proficiency of the participants, the first instrument, i.e., Oxford Placement Test (2004), was used.In order to measure the participants' receptive knowledge of collocation, a receptive test of English lexical collocations was employed.The receptive test was adapted from (Haqiqi, 2007), and it was comprised of 50 items.The items in this test included different types of lexical collocations like noun+ noun, verb+ noun.The reliability index of the test was calculated using Cronbach alpha (r =.92).The last instrument used was a productive English collocation test (Haqiqi, 2007).The productive collocation test consisted of fill-in-the-blank items with the initial letters of collocations as clues to the right answer.The test was highly reliable and produced a reliability estimate of.89through Cronbach alpha.

Procedures
The study intended to assess intermediate and advanced Iranian EFL learners' productive and receptive knowledge of English collocations.At the outset of the study, an OPT was run among 100 male and female EFL undergraduate and postgraduate students.Then, from among them, 30 advanced and 30 intermediate EFL learners were selected according to the scoring rubrics of the test.The participants at each level (i.e.intermediate and advanced) received two tests of collocations, namely receptive and productive tests of collocations.Then, the gained scores were analyzed through SPSS.

RESULTS
The results obtained for each of the research questions are presented in what follows.

Research Question One
The first research question of the study was: Is there a significant difference between the Persian-speaking advanced EFL learners' productive and receptive knowledge of collocations?In order to answer this research question, two tests of collocations, namely productive and receptive tests were given to the advanced learners and the results were compared.Table 1 presents the descriptive statistics.
As shown in Table 1, the mean difference of the receptive and productive collocation tests is not very much (the mean scores of the receptive test being 36.03 and that of productive test being 34.21. Figure 1 compares the mean scores of the two tests.
The mean difference between the receptive and productive tests does not seem to be significant; however, in order to be more objective, a paired-samples t test was run.Table 2 presents the results.
The results of the paired samples t test indicated that there was no significant difference between the advanced IJALEL 6( 7):11-16 EFL learners' productive and receptive knowledge on collocations, t(29) =.46, p >.05.Thus, the first null-hypothesis could not be rejected.

Research Question Two
The second research question of the study asked: Is there a significant difference between the Persian-speaking intermediate EFL learners' productive and receptive knowledge of collocations?In order to compare the intermediate EFL learners' mean scores on productive and receptive lexical English collocations, the mean scores were compared.The results are presented in Tables 3 and 4.
As shown in Table 3 and Figure 2, the mean score of the receptive knowledge (M = 23.95) of intermediate learners was higher than their productive knowledge (M = 21.23).Figure 1 shows the results in pictorial form.
The mean difference between the two groups is obvious; however, in order to be more objective, a paired-samples t test was run, the results of which are presented in Table 4.
According to the results presented in Table 4, the mean difference between the collocational receptive and productive tests by intermediate EFL learners was significant, t(29) = 2.92, p <. 05; this led to the rejection of the second null hypothesis of the study.

DISCUSSION
In this section, it is attempted to present some reasons behind the findings, and to compare the findings with other studies in the field.

Research Question One
One of the objectives of the present research was an attempt to find out whether there was a significant difference between productive and receptive collocational knowledge of Iranian intermediate EFL learners.The results of the study revealed that the intermediate learners' performance on the receptive collocation test scores were significantly higher than their productive collocation test scores.
The higher scores of the intermediate EFL learners in the receptive test can be attributed to the fact that intermediate EFL learners in receptive tests could take advantage of their passive knowledge which is easier to access compared with their active knowledge.However, for advanced EFL learners access to passive knowledge seems not to be out of reach, and the passive knowledge is as great as active knowledge.Furthermore, in receptive tests, the test takers have the opportunity to guess the meaning of the collocations form context. Various contextual clues may be at work when test takers deal with receptive tests of collocations.However, when production of collocations is observed in productive tests, no contextual clues is at work, and the test takers have to start from their own knowledge.
In the Iranian EFL context, the findings of the present research are in line with the study by Koosha and Jafarpour (2006), who intended to discover whether collocational proficiency of prepositions could be examined at the differ-ent levels of EFL learners' proficiency.It was revealed that learners' performance in the test of collocation preposition was positively related to their level of language proficiency.
On the contrary, the findings of the present research are in contrast with the study by Bazzaz and Samad (2011), who indicated that there was a large positive relationship between general language proficiency of intermediate EFL learners' and their productive collocation knowledge in writing tasks.In the present study, the receptive and productive collocation knowledge of intermediate EFL learners was investigated, while in Bazzaz and Samad's (2011) study the general proficiency level of EFL learners and no their productive knowledge of collocation was surveyed.

Research Question Two
The second research question intended to compare the productive and receptive collocational knowledge of Iranian advanced EFL learners.The results revealed no significant  The results for such finding may be attributed to the fact that advanced EFL learners can easily make a link between their productive and receptive knowledge of vocabulary.This improves their scores in productive tests of collocations.
The results of the present research also are in contradiction to the study by Al-Amor (2006), who evaluated the productive and receptive collocational knowledge of Saudi EFL learners.In his research, it was found that there was a significant relationship between the EFL learners' receptive and productive knowledge of collocations.In addition, the participants in his research gained better results on the productive test compared with the receptive test.The reason for such findings according to Al-Amor was the fact that the target collocations in his receptive test were of lower frequency than those in the productive test.As it was mentioned earlier, in the present research, it was revealed that the receptive knowledge of collocations is stronger than productive knowledge of collocations, and the frequency of the collocations in the tests employed in the present research was similar.The findings of the present study also are in the same vein with the study by Shehata (2008) who found that advanced EFL learners performed better on the receptive test.

CONCLUSION
The contradictory results obtained in previous studies on collocations provided the motive to conduct the present study.The results of the study indicated a slight difference between the receptive and productive collocational knowledge of advanced EFL learners, while intermediate learners' receptive collocations test scores were significantly higher than their productive collocations test scores.According to the findings of the present study, some implications for teachers and L2 learners can be assumed.The results can help language teachers to attribute the problems which learners have in the development of their language proficiency partly to the lack of collocational knowledge.In fact, teaching collocations to EFL learners should be granted more attention.Inspired by the findings of the present research, language teachers and learners should take into account that knowing a word or a collocational combination, is not just to identify the meaning of the word or collocation in tests, rather to be able to use the collocation in language production.In addition, the people who are in charge of language teaching in EFL contexts, are suggested to make their best to bridge the existing gap between EFL learners' receptive and productive knowledge.In order to achieve this goal, some exercises like developing paragraphs can be suggested.Teaching collocations in language classes seems not to be sufficient; therefore, language teachers should ask language learners to use the learned collocations productively.

Figure 1 .Figure 2 .
Figure 1.Receptive and Productive Tests' mean comparisons of Advanced Learners

Table 1 .
Descriptive statistics of receptive and productive collocational knowledge of advanced learners

Table 2 .
Results of paired-samples t test between advanced learners' productive and receptive tests

Table 3 .
Descriptive statistics of receptive and productive collocational knowledge of intermediate learners

Table 4 .
Results of paired-samples test between intermediate learners' productive and receptive tests