International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016

EFL Learners’ Writing Progress through Collocation Awareness-raising

Approach: An analytic assessment
Elaheh Hamed Mahvelati, Assistant Professor, Department of Language and Basic Science,
Faculty of Petroleum Engineering, Petroleum University of Technology, Abadan, Iran
Mahvelati.e@put.ac.ir

Abstract
There is insufficient rigorous research examining which features of EFL/ESL learners’ writing
can be improved through their awareness/knowledge of collocation. This study, therefore,
addressed this issue and examined the effect of this awareness on Iranian EFL learners’ writing
performance with respect to the specific features of each writing sub-component (i.e., content,
organization, vocabulary and language use) in the short and long term. The study was a quasi-
experimental using a within- and between-group approach utilizing a pre-test/post-test design
with a control group supplemented with the qualitative information obtained from the interviews
with the writing raters. The results showed that developing knowledge/awareness of collocations
effectively improved the participants’ quality of writing in terms of ‘vocabulary’, ‘organization’
and ‘language use’. A deeper analysis of the results revealed that the observed improvements in
these sub-components were in terms of some particular features: the range of sophistication,
lexical choice and usage; succinctness, fluency and clarity of the expressed ideas; and the
accuracy of agreement, tense and prepositions. In addition, the sub-findings emerging from the
results accentuated the significance of instructional intervention, in general, and first language-
second language contrastive analysis, in particular, in helping L2 learners notice, note and
incorporate collocations in their output.

Keywords: Collocation, collocation awareness, writing proficiency, writing sub-components,

first language-second language (L1-L2) contrastive analysis

Introduction
Upper-intermediate L2 learners, in general, and Iranian EFL learners, in particular, have
been found to experience serious difficulties in using even the most common or already known
words in their written production (Namvar, Nor, Ibrahim, & Mustafa, 2012). Indeed, it is a
serious problem in the majority of writing classes in Iran giving rise to various research studies.
The findings of these studies (e.g. Darvishi, 2011; Dastjerdi & Samian, 2011) have indicated that
lack of sufficient knowledge of collocational fields of words is one of the main reasons for such a
deficiency. This is rooted in the fact that multi-word units including fixed expressions and
collocations are not taught sufficiently in language classrooms in Iran (Bahardoust, 2013;
Ghonsooli, Pishghadam, & Mahjoobi, 2008). Therefore, most written English of Iranian language
learners, even at upper-intermediate to advanced levels, falls short of expectations and contains
unacceptable word combinations (Zarei & Koosha, 2003). Hence, it has been strongly suggested
that a major part of writing instruction be devoted to developing learners' knowledge of
collocation (Bahardoust, 2013; Hsu, 2007).
This idea that very few utterances in a language are completely novel creations and
language mostly consists of pre-fabricated meaningful word combinations is reflected in Lewis’s
(1993) ‘Lexical Approach’. More precisely, the lexical approach advocates argue that language
learners can identify patterns in a language by the help of collocations which are the habitual co-
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 70

occurrences of words together at the syntagmatic level (Lewis, 2000; Martyńska, 2004, Nation,
2001). Hence, the lexical approach, which is at the center of the current communicative teaching
approach, has introduced new approaches to syllabus design with a focus on the importance of
learning phrases or chunking language as a unit rather than as individual words (Lewis, 1997).
Indeed, this approach has emphasized the need for making the neglect of collocations, as a subset
of multi-word units or prefabricated chunks; in English as a foreign/second language (EFL/ESL)
classrooms a big concern for language teachers (ibid).
Notwithstanding all these emphases, empirical research studies that specifically and
deeply enough investigate which features of writing, as one of the main goals of language
learning in the EFL/ESL contexts, can actually be improved through collocation instruction are
scarce. Moreover, the few available studies show little consistency in their findings. This study,
therefore, aimed to fill this gap and empirically examined the impact of improving learners'
knowledge of collocation on their writing performance in terms of each writing sub-construct
(i.e., content, organization, vocabulary and language use) in the short and long term. It is worth
noting that in comparison to the other studies in this body of research, the present study
addressed this issue more deeply by employing the qualitative method of in-depth interviews
with the writing raters.
Furthermore, the form of instruction employed in the previous studies in the related
literature was limited to the traditional teaching of the collocational fields of some words.
However, the collocation teaching method used in this study was more comprehensive. In fact, it
was presented through an awareness-raising approach, mainly based on Ying and Hendricks'
(2003) proposed model of ‘Collocation Awareness-Raising (CAR) Process’. More precisely, in
this type of instruction, students were made aware of the idea of collocation and taught how to
incorporate these word combinations in their written production through the four steps proposed
by Ying and Hendricks (2003) as well as first language-second language (L1-L2) contrastive
analysis.

Literature Review
Collocations and Writing Competence
L2 writing has always posed severe difficulties for English as Second/Foreign Language
(EFL/ESL) learners. Hence, exploring the possible ways to enhance language learners' writing
proficiency has always been a topic of concern to L2 writing specialists (Ismail, 2011). This has
inspired various researchers to conduct studies in this regard and their findings have suggested
teaching collocations as one of the most viable ways (e.g. Ashouri & Mashhadi Heidar, 2015;
Mounya, 2010; Sadoughvanini, 2012). Nevertheless, it is important to note that there are few
studies, such as Avci (2006) and Liu (2000), which have not reported the same positive results
regarding the effectiveness of collocation instruction. Avci (2006) and Liu (2000), however,
believed that their findings might be due to the interference of some other factors, such as the
subjects' level of language proficiency and the length of the study. Hence, they suggested further
research in this area.
It is also important to note that most of the conducted studies in this area merely focused
on learners' overall writing proficiency and only few of them (e.g. Eidian, Gorjian, & Aghvami,
2013; Ghonsooli et al., 2008) specifically examined what sub-components of writing could be
improved by teaching of collocations. Additionally, these few available studies show little
consistency in their findings. Ghonsooli et al. (2008), for example, in their study on a group of
Iranian upper-intermediate learners found that teaching collocation could only improve the
'fluency' and 'vocabulary' sub-components. However, Ashouri & Mashhadi Heidar (2015)
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 71

reported the positive effect of collocation instruction on the 'vocabulary' and 'mechanics' sub-
scales. Thus, further research on this issue is still needed.
In brief, as the review of this body of research reveals, knowledge of collocation can
significantly facilitate L2 writing development. As a result, L2 researchers and scholars stress the
need for integrating collocations in language syllabus. This has led into the emergence of various
pedagogical suggestions. One of these suggestions is the instructional method of awareness-
raising approach which will be dealt with in the following section.

The Awareness-raising Approach

Teaching collocations explicitly through an awareness/consciousness-raising approach
has been strongly suggested by Lewis (1993, 2000) and some other researchers, such as Hill
(2000) and Woolard (2000). This is because it is not feasible to teach all that language learners
need due to the fact that lexical elements are infinite and the time of the class is limited. More
particularly, Lewis (1993, 2000) recommends that language teachers should raise their students'
awareness of collocations and also help them to develop efficient skills for learning these chunks.
Following Lewis, some other researchers and specialists, such as Stoitchkov (2008) and
Boonyasaquan (2009), stress the significance of developing a set of skills as well as enhancing
awareness.
Researchers, therefore, have tried to propose different pedagogical methods and activities
to enhance L2 learners' knowledge of new collocations, reinforce the learnt ones and give the
learners the opportunity to practice the required skills. One of the proposed methods belongs to
Ying and Hendricks (2003). They put forward a teaching model called the 'collocation
awareness-raising (CAR) process' and identified four steps in teaching collocations within their
proposed approach. It is noteworthy that the awareness-raising approach which was applied as
the treatment of the present study was mainly based on this model.
As Ying and Hendricks (2003) put it, the first step in teaching collocations is to become
sure that students know what is meant by collocation. Using some examples from the target and
first language can be helpful. Secondly, it is suggested, in order to raise learners' collocation
awareness, to begin with those collocations which are relevant to in-class activities. In order to
raise learners' awareness of the target collocations, some reference materials for the target
collocations, such as collocation instructional books and dictionaries, articles and concordances
should be introduced and the learners should be taught how to work with them.
Thirdly, Ying and Hendricks (2003) suggest teaching learners the steps of noticing and
noting collocations and the correct ways of incorporating them in their own L2 production.
Finally, the last step is to check students' work and provide useful feedback. On this basis, it can
be concluded that, as Ying and Hendricks (2003) contend, their proposed method is a ‘threefold
process’ which ‘bridges students from noticing to noting to incorporating desired target forms’
(p. 58). Indeed, they believe that learners should be helped to notice the target features and also
the gaps in their own linguistic knowledge. This is based on what Ellis (1995) and Schmidt and
Frota (1986) suggest as the necessary conditions for acquisition to take place.
Additionally, with regard to cross-language exploration, James (1994) notes that raising
learners’ awareness of their own native language can be considered important in helping them
understand the target language by comparing and contrasting the two languages and ‘facilitating
[a] bridge between them’ (p.212). In his view, contrasting L1 with L2 can help learners to find
the differences between the two languages which can lead to recognizing new patterns in the
target language and expanding learners' knowledge of the language. The findings of a study by
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 72

Laufer and Girsai (2008) also lend empirical support to the positive outcomes of this pedagogical
suggestion for L2 learners' collocation development.

Research questions
Q1. Does collocation awareness-raising approach significantly affect the learners’ writing
performance in terms of the ‘content’ sub-component in short and long terms?
Q2. Does collocation awareness-raising approach significantly affect the learners’ writing
performance in terms of the ‘organization’ sub-component in short and long terms?
Q3. Does collocation awareness-raising approach significantly affect the learners’ writing
performance in terms of the ‘vocabulary’ sub-component in short and long terms?
Q4. Does collocation awareness-raising approach significantly affect the learners’ writing
performance in terms of the ‘language use’ sub-component in short and long terms?

Methodology
Design
In this study, a quantitative approach design (i.e. a quasi-experimental, using a within-
and between-group approach utilizing a pretest/posttest design with a control group) was
primarily employed supplemented with the qualitative information obtained from the interviews
with the writing raters in order to clarify and follow up on the quantitative findings.

Participants
Intact classes comprising sixty-three upper-intermediate EFL learners (including 37
female and 26 male) were selected to serve as the participants of this study. All were Iranian
adult learners who had enrolled in regular English courses in Safir Language Academy in
Mashhad, Iran. These classes were randomly assigned to the two groups of control and
experimental. The number of the participants in these two groups was 32 and 31 respectively.
For the qualitative part of the study, the two raters who scored the subjects' essays at both
pre and posttest stage were also interviewed. The raters were female English teachers who had a
minimum experience of ten years in teaching IELTS, TOEFL and general English courses in
various language centers and also a minimum teaching experience of three years as EFL lecturers
in some universities and higher education institutes in Iran. Both raters were native speakers of
Persian. One of them had master's degree in Teaching English as a Foreign Language (TEFL) and
the other one was a PhD candidate in Teaching English as a Second Language (TESL).

Instruments

Proficiency test
The participants had been classified as upper-intermediate learners by the placement test
of the academy. However, a TOEFL Paper-Based Test was administered in order to determine
their level of proficiency again and also ensure the homogeneity of the participants prior to the
experiment. The test was taken from Gallagher’s (2000) TOEFL test practices. The reliability of
the test was assessed using the Cronbach’s alpha as (α =.90).

Test of collocation
In order to determine the homogeneity of the learners in terms of their knowledge of
collocation, the learners took a test of collocation prior to the treatment. This researcher-made
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 73

test was composed of two parts measuring the learners’ both receptive and productive knowledge
of the same target collocations which would be taught in the experimental class during the
treatment period.. The first part of the test assessed the subjects’ productive knowledge of the
target collocations through gap-filling items, then the answers were collected. The second part of
the test, measured the participants’ receptive knowledge of the same target collocations through
multiple-choice items. Thus, the total score for the test was the average of the scores of these two
parts. The test was piloted and its reliability was assessed using the Cronbach's alpha as (α =
0.79).
In the current study, the validity of the employed instruments, as suggested by creswell
(2008), was determined by a panel of experts. The panel of experts was asked to ascertain the
content and face validity of the collocation test, interview questions and the proficiency test. In
addition, they were consulted about the selected target collocations, the reading texts, the writing
rating scale and the writing prompts. These experts were L2 specialists holding a PhD. Any
modifications to the instruments were made after consultation with these experts.

Target collocations
The decision regarding selecting the target collocations was made after consulting the
panel of experts. Based on the feedback from the panel of experts, it was decided to make a list of
the most common collocational errors made by the Iranian upper-intermediate learners in their L2
writing. Part of the data regarding these collocational errors was collected from the researcher's
own classes and the rest was gathered from her colleagues' classes in different language centers.
Due to time constraints, it was decided to teach three or four target collocations each session.
Therefore, on the whole, 50 of the most common collocational errors found in the sample essays
were selected as the target items for the present research. The panel of experts, the educational
supervisors of the language academy as well as the teacher of the classes, was all consulted in
this regard.
It is important to note that the target collocations in this study included both the lexical
and grammatical types, viz.: noun+noun, noun+verb, verb+noun, adjective+noun,
adverb+adjective, verb+adverb, noun+preposition, preposition+noun, adjective+preposition and
delexicalized verbs (become/get/turn/go).

The reading texts

The target collocations were placed into the reading texts used as one of the treatment
materials for the experimental group. Each of these reading texts, which were given to the
students every session, included three or four of the target collocations.
The students were given these texts and taught the three steps of noticing-noting and
incorporating the target collocations. More precisely, the instructor started reading the texts and
when she encountered the target collocations, she paused and drew the learners’ attention
explicitly to these word combinations and asked them to underline or highlight them. Then she
explained to them that they should not only focus on the ideas but also the linguistic features of a
text, particularly the co-occurrence of words together.

Collocation dictionaries and concordance

Oxford Dictionary of Collocation, Longman Dictionary of Contemporary English as well
as two free web-based concordancing tools with the databases selected from the Corpus of
Contemporary American English (http://corpus.byu.edu/coca/) and British National Corpus
(http://corpus.byu.edu/bnc/) were employed as the sources for teaching collocations in the
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 74

experimental group. The learners were, first, taught how to use these collocation resource
materials. Then they were given some homework assignments which required regular use of
them.
Writing tests
Each writing test, that is, pre-test, immediate- and delayed post-test, consisted of two
argumentative writing tasks. The learners were asked to write at least 250 words for each essay
within 40 minutes. For the purpose of this study, the ESL Composition Profile (Jacobs, Zingraf,
Wormuth, Hartfiel, & Hughey, 1981) was the rating scale employed to score the writing samples
of the present study. The scale comprises five ESL sub-components of writing– which help raters
evaluate learners’ writing with respect to their:
Content: development of thesis and relevance to the assigned topic,
Organization: fluent expression; clearly stated/supported, succinct, well-organized, logically
sequencing and cohesive presentation of ideas,
Vocabulary: sophisticated range, effective word/idiom choice and usage, word form mastery,
appropriate register,
Language use: effective complex constructions, few errors of agreement, tense, number, word,
order/function, articles, pronouns and prepositions, and
Mechanics: mastery of conventions, few errors of spelling, punctuation, capitalization and
paragraphing
It is important to note that the researcher would not report the findings regarding the
effect of teaching collocation on the use of mechanics in writing in the current research since
such an effect does not sound justifiable and cannot be expected.

Interview
For the purpose of this study, an interview protocol was prepared for the one-to-one semi-
structured in-depth interviews with the raters of the learners' writing tests. The protocol was
piloted to ensure the clarity of the questions and to see if these questions could serve their
purpose. This enabled the researcher to edit the interview protocol in terms of deleting the
questions that did not provide useful information and adding others. Its validity was determined
by making the necessary revisions to the questions based on the feedback from the panel of the
experts.
The protocol consisted of questions regarding the writing performance of both groups, in
general, and in terms of the sub-scales of writing (content, organization, vocabulary, language
use), in particular. The raters were asked to elaborate more on the scores they had given for the
writing tests and comment on changes, if any, in the learners' post-treatment essays in
comparison to their pre-treatment essays. They were also asked to compare the post-treatment
writing qualities of the control and experimental groups, and elaborate on their similarities and
differences.

Procedures
Data collection was carried out in the regular English courses which were held for adult
upper-intermediate learners within a period of two months and a half in Safir language academy.
The treatment period was composed of seventeen treatment sessions which were held three times
a week. The procedures for collecting the quantitative and qualitative data were carried out in
three stages, that is, the pre-treatment stage, the treatment stage and the post-treatment stage,
which will be dealt with one by one in the subsequent sections.
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 75

The pre-treatment stage

First, a test of proficiency was given to all the participants in order to ensure that they
were all truly upper-intermediate learners. In the second pre-treatment session, two writing tasks
were given to the participants to measure their writing proficiency and to ensure the homogeneity
of the participants both within and between groups. They also served as the pre-writing test for
the study. Finally, in the third pre-treatment session, a collocation test was administered in order
to check the homogeneity of the participants in terms of their knowledge of collocation within
and between the groups.

The treatment stage

The participants were randomly assigned into one experimental and one control group.
Both groups were taught by the same instructor within the same time span. Since the course
which was used for this study was a general proficiency course offering practice in the four skills,
both groups received instruction on all of these skills. The instructor exposed both groups to the
same language input (audio, video and textual) and used the same course materials except for the
materials (for example, the reading texts in which the target collocations were embedded and
concordances) which were specifically used to teach the target collocations in the experimental
groups. Therefore, the only variable that these two groups differed on was related to the teaching
of collocation.
In the case of writing, both groups received instruction on this skill in the normal run of
the class regarding paragraph development, writing topic sentences, the overall organization of
the essay, content and mechanics. Every other session, the students in both groups were assigned
a new topic and their essays were carefully corrected and were returned to them before assigning
another topic. Both groups were assigned to write exactly the same number of essays (eight
writing assignments) on exactly the same topics during this experiment. The students in the
experimental group were provided with explicit and direct feedback on their collocational errors.
However, this was not the case in the control group.
In the experimental group, the learners’ awareness of the concept of collocation, in
general, and their knowledge about the target collocations, in particular, were developed
explicitly through the four steps suggested by Ying and Hendricks’ (2003) proposed method
called ‘Collocation Awareness-Raising (CAR) Process’. In addition to these four steps, first
language-second language (L1-L2) contrastive analysis was also adopted as another step. In
particular, More precisely, the students were made aware of the idea of collocation, provided with
both positive and negative evidence of word combinations through L1-L2 contrastive analysis,
familiarized with some available resources, such as web-based concordances and self-study
materials, asked to use the target collocations in their language production, thus helping them to
notice the gap in their linguistic system, and finally given some useful feedback on their language
production, especially on their written output. As mentioned earlier, some reading texts including
the target collocations and some collocation resources (e.g. Oxford Dictionary of Collocation and
British National Corpus) were the treatment materials used for developing learners’ knowledge
of collocations.
However, the learners in the control group did not receive any treatment for collocations.
They were also given some reading texts similar to the ones given to the experimental group
which only differed in terms of linguistic features regarding the target collocations. To put it
more simply, for the purpose of the current research, the target collocations were excluded from
these texts since it was attempted to have a control class in which neither explicit nor implicit
collocation instruction happened. Therefore, even mere exposure to the collocations in the
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 76

reading texts was avoided since according to some researchers including Krashen (1989), it is an
implicit form of teaching.

The post-treatment stage

The first post-test of writing was administered in a session following the last treatment
session for both groups. Then the raters were interviewed. The interview with each of them took
one hour and a half. Two weeks after the immediate post-tests, the delayed post-test of writing
were administered. As mentioned before, the writing tests were rated by two raters. To ensure
that a high degree of agreement existed between the two raters in this regard, inter-rater reliability
was measured. The Pearson Product Moment Correlation was employed and the raters' rating
scores were found to enjoy a very high degree of agreement for all the three tests (r > 0.8). The
average values of the two raters' scores were used for data analysis.
To ensure the validity of the qualitative data obtained from the interviews, peer debriefing
technique was employed (Ary, Jacobs, Razavieh, & Sorensen, 2010), and for the purpose of
insuring the reliability issue, the inter-rater method was employed (McMillan & Schumacher,
2010). In fact, the researcher’s coding of the data was compared with that of a peer who had been
asked to code the data, and the inter-rater reliability was found to be .92.

Data analysis
The Statistical Package for Social Science (SPSS) was employed to analyze the
quantitative data in this study. To ensure that the data were normally distributed, the tests of
normality (i.e. Shapiro-Wilks, Kurtosis and Skewness) were run for all the data prior to any
statistical analysis. The results indicated that the data used in the present study had normal
distribution. Hence, repeated measures ANOVA and independent samples t-test were employed
for analyzing the quantitative results of the study.
In the case of the qualitative data, the five steps, viz. familiarization, a thematic framework
identification, indexing, charting, mapping and interpretation, suggested by Krueger (1994) in his
‘framework analysis’ were followed.

Results and discussion

Firstly, four separate independent samples t-tests were employed to compare the pre-test
scores of the control group and the experimental group in terms of each writing sub-component.
The results revealed that the differences between the groups with respect to each sub-component
[Content t(61) = .328, p >.05, Organization t(61) = 1.89, p >.05, Vocabulary t(61) = 1.90, p >.05
and Language use t(61) = .461, p >.05] were not statistically significant prior to conducting the
experiment.
Then to answer the research questions of the study regarding the within-group effects of
the collocation awareness-raising approach on the learners’ writing performance with respect to
each writing sub-component, five separate one-way repeated measures ANOVAs were
performed. The results are presented in the following tables.

Table 1. Mauchly's Test of Sphericityb for each sub-scale of writing

Approx. Epsilona
Within Subjects Effect Mauchl Chi- Greenhouse- Huynh- Lower-
y's W Square df Sig. Geisser Feldt bound
Content time 0.797 6.59 2 0.037 0.831 0.874 0.5
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 77

Organization time 0.864 4.245 2 0.120 0.880 0.931 0.5

Vocabulary time 0.742 8.648 2 0.013 0.795 0.832 0.5
Language use time 0.699 10.386 2 0.006 0.769 0.802 0.5

Table 2. Tests of within-subjects effects/Multivariate tests for each sub-scale of writing

Source d F Sig. Partial
f Eta
Square
d
Content Time Sphericity Assumed 2 164.779 0.000 0.546
Time Greenhouse-Geisser 1.663 164.779 0.000
Error(time) Sphericity Assumed 60
Error(time) Greenhouse-Geisser 49.88
8
Organizatio Time Sphericity Assumed 2 485.928 0.000 0.642
n
Error(time) Sphericity Assumed 60
Vocabulary Time Sphericity Assumed 2 444.704 0.000 0.637
Error(time) Greenhouse-Geisser 1.590 444.704
Sphericity Assumed 60
Greenhouse-Geisser 47.70
0
Language Time Sphericity Assumed 2 290.197 0.000 0.606
use Time Greenhouse-Geisser 1.537 290.197
Error(time) Sphericity Assumed 60
Error(time) Greenhouse-Geisser 46.11
7

As Table 1 demonstrates, the results of Mauchly’s test indicated that the assumption of
Sphericity was violated for all the sub-components (p <.05) except the ‘organization’ sub-
component (2= 4.24, p >.05). Therefore, the multivariate tests (i.e. sphericity assumed) were
interpreted for the ‘organization’ sub-component and the test of Greenhouse-Geisser for the other
four sub-constructs. As can be seen in Table 2, the results of the tests revealed that the writing
performance of the learners in the experimental group differed significantly over time in terms of
content (F(1.66,49.88)= 164.77, p <.05, 2= 0.54), organization (F(2,60)= 485.92, p <.05, 2= 0.64),
vocabulary (F(1.59,47.70)= 444.70, p <.05, 2= 0.63) and language use (F(1.53,46.11)= 290.19, p <.05,
2= 0.60) . Thus, the pairwise comparisons were conducted in order to see which specific means
differed significantly.

Table 3. Pairwise Comparisons for each sub-scale of writing

Mean Sig
Difference
Content Time Pre – immediate post-test -.1.09* 0.000
Pre- delayed post-test -.806* 0.000
Immediate – delayed post-test 0.290* 0.000
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 78

Organization Time Pre – immediate post-test -2.435* 0.000

Pre- delayed post-test -1.968* 0.000
Immediate – delayed post-test 0.468* 0.000
Vocabulary Time Pre – immediate post-test -2.613* 0.000
Pre- delayed post-test -2.363* 0.000
Immediate – delayed post-test 0.250* 0.006
Language Time Pre – immediate post-test -2.758* 0.000
use Pre- delayed post-test -2.500* 0.000
Immediate – delayed post-test 0.258* 0.033

According to Table 3, the immediate post-treatment mean scores of the learners were
significantly higher than their pre-treatment mean scores for each sub-component (p < .05). In
addition, it was found that although the writing scores of the learners significantly deteriorated
from the immediate post-test to the delayed post-test (p <.05), the differences between their pre-
test and delayed post-test scores were still statistically significant for all the five sub-components
(p <.05). This suggests that the treatment had durable effect on the leaners’ writing proficiency.
Furthermore, the Partial Eta Squared values show a large effect size for all the writing sub-scales.
Based on the results reported so far, the learners in the experimental group showed
significant improvement in their writing performance in terms of all the five sub-components. In
order to see if this significant progress was the result of the employed collocation teaching
method or the teacher’s writing instruction in the normal run of the class, their post-treatment
writing mean scores were compared with those of the learners in the control group for each sub-
component. For this reason, independent samples t-tests were performed.

Table 4. T-test for the independent samples of the control vs. experimental group’s immediate
and delayed writing post-tests in terms of each writing sub-scale
Groups N M SD T df P
Immediate Control 32 23.39 0.50 -1.15 61 0.254
Content post-test Experimental 31 23.52 0.41
Delayed Control 32 23.14 0.43 -.786 61 0.435
post-test Experimental 31 23.23 0.42
Immediate Control 32 15.17 0.38 -13.13 61 0.000
Organization post-test Experimental 31 16.85 0.60
Delayed Control 32 14.96 0.42 -11.93 61 0.000
post-test Experimental 31 16.38 0.51
Immediate Control 32 14.64 0.51 -17.22 61 0.000
Vocabulary post-test Experimental 31 16.87 0.51
Delayed Control 32 14.49 0.50 -16.282 61 0.000
post-test Experimental 31 16.62 0.53
Immediate Control 32 20.14 0.66 -7.947 61 0.000
post-test Experimental 31 21.81 0.98
Language use
Delayed Control 32 20.00 0.74 -7.435 61 0.000
post-test Experimental 31 21.55 0.91
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 79

Table 4 demonstrates that the immediate and delayed post-treatment mean scores of the
experimental group were markedly higher than those of the control group only in the case of
‘organization’, ‘vocabulary’ and ‘language use’ (p<.05). This means that the learners’ writing
qualities in the experimental group did not differ significantly from those of the learners in the
control group in terms of ‘content’ (p >.05). To throw more light on this issue, the writing raters
were interviewed.
The findings from the conducted interviews confirmed and elaborated the quantitative
results regarding the significant impact of the collocation awareness-raising approach on the
learners’ writing abilities in terms of these three sub-constructs which will be dealt with one by
one below. However, it is noteworthy that both raters emphasized that the observed
improvements in the writing performance of the experimental group was noticeable in
comparison to their pre-test writing quality and the writing performance of the control group.
Therefore, their positive comments regarding these learners’ writing progress did not mean that
they changed into proficient writers after the treatment or their writing quality was dramatically
enhanced. It simply meant that their writing skills were improved through the employed
treatment, but they still needed more practice to achieve the desired high standards of writing.

Vocabulary
Based on Jacob’s rating scale, the raters in the current study scored the ‘vocabulary’ sub-
component based on the following features: the range of sophistication, lexical choice, usage,
form and appropriateness of register. Both raters (Rater A and Rater B) reported that although the
essays in both groups were enhanced in terms of vocabulary, the experimental group’s
improvement in this aspect was found to be superior to the control groups. They contended that
this superiority was in terms of the range of sophistication, lexical choice and usage. More
particularly, the raters reported improvement in the lexical choice and usage since more correct
collocations and fewer unacceptable word combinations were observed in the learners' writing at
the post-test stage.
The raters, in fact, highlighted this point that they observed fewer L1-based errors (i.e.
Persian negative transfer) in the writing of the experimental group after the treatment. Such
finding suggests that contrasting Persian (L1) with English (L2) gave these learners the required
opportunity to find out the differences between the two languages and this led into recognizing
new collocational patterns in English and expanding their knowledge about both acceptable and
unacceptable English word combinations. These findings, therefore, empirically espoused James’
(1994) and Stoitchkov’s (2008) claims regarding the significant role of L1-L2 contrastive
analysis in helping language learners understand the target language better. Moreover, these
findings match those of Laufer and Girsai (2008) who found out that the contrastive analysis of
L1 and L2 had a significant effect on L2 collocation knowledge enhancement.
Furthermore, the raters observed an increase in the learners’ lexical sophistication level as
a result of using more collocations, such as granted the custody to, a child from a deprived home,
which consequently made their writing sound more sophisticated.

Organization
Following Jacob‘s rating scale, the raters scored the ‘organization’ sub-component based
on the following features: fluency, the clarity of the expressed ideas, supporting the expressed
ideas, succinctness, cohesion, arrangement and sequencing of the ideas. Both raters believed that
although the organizations of the essays in both groups were enhanced, the improvement in the
experimental group was found to be superior in terms of fluency, the clarity of the expressed
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 80

ideas and succinctness. That is, the raters found that the ideas in the post-treatment essays of the
experimental group were expressed and also supported more clearly and in fewer words. They
also emphasized that this group produced more fluent and natural-sounding texts after the
treatment. Indeed, the enhancement of their knowledge about some appropriate collocations
helped these learners to produce fewer unnecessary long sentences to express their ideas. As a
result, the reader could gain a better and clearer understanding of the expressed ideas. One of the
raters, for example, stated that
In line with such finding is Sadoughvanini’s (2012) study that indicated making the
learners aware of collocations and teaching them how to use them effectively can help them “to
convey what they have in mind as well as possible” in their L2 written utterances (p. 36).

Language use
The raters scored the ‘language use’ sub-scale based on the following features: the degree
of sentence construction complexity, the accuracy of agreement, tense, number, word order,
articles, pronouns and prepositions. The raters believed that the experimental group could get
higher scores for this sub-component because they made fewer errors of agreement, tense and
prepositions. In particular, the raters stated that, in comparison to the control group, the learners
in the experimental group used more collocations. This, in turn, helped them produce shorter or
fewer unnecessary sentences to express their ideas. As a result, they made fewer errors in terms
of agreement and tense. One of the raters’ remarks is as follows:
These findings empirically support Lewis’ (2000) claim that without sufficient knowledge
of collocations as short cuts to clearly convey what they mean, L2 learners often produce longer
sentences in order to express their views precisely. This, therefore, increases the chance of
committing more errors and deviating from native speaker norms.
Furthermore, the raters referred to the learners’ improvement in using more correct
prepositions and less L1 negative transfer which could be due to the fact that one of the concerns
of teaching collocation in the present study was developing the learners’ knowledge of
grammatical collocations, including ‘noun + preposition’, ‘adjective + preposition’, ‘preposition
+ noun’ and ‘verb + preposition’. In Jacobs’ rating scale, the use of correct prepositions is
considered to be part of grammatical competence. Hence, in the current study, fewer
prepositional errors in the learners’ written works were regarded as improvement in their
grammatical aspect of writing.
In all, in addition to Ying and Hendricks’ (2003) own study, the findings of the present
research can be regarded as further empirical proof of the effectiveness of their proposed
collocation teaching model. Moreover, consistent with the findings of the present study, Ying
and Hendricks’ study proved the significant role of collocation instruction in helping learners
make fewer grammatical and usage errors and reach a higher level of language sophistication.
Moreover, these findings are, to a certain extent, in agreement with the results of the
research carried out by Ghonsooli et al. (2008) on a group of Iranian EFL writers. The students in
Ghonsooli et al.’s (2008) study showed progress only on the ‘vocabulary’ and ‘organization’
(which was called fluency in their study) sub-components. They, indeed, found no significant
improvement in the grammar of the learners. The researchers attributed the learners’ lack of
progress in the grammatical aspect of their writing to the short period of the treatment.
Notwithstanding the similarity between the finding of Ashouri & Mashhadi Heidar’s
(2015) and Eidian et al.’s (2013) conducted studies on Iranian EFL learners and that of the
present research regarding the positive effect of collocation instruction on the 'vocabulary' sub-
scale, these findings differ with respect to the other sub-constructs of writing. In fact, contrary to
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 81

the findings of the present research, the findings of these two studies showed that while the effect
on the learners' ‘grammar’ and ‘fluency’ failed to be of great significance, it was significant for
the ‘mechanics’ sub-component. Their justifications for their findings regarding the sub-scales of
‘grammar’ and ‘fluency’ were the short period of the treatment and lack of emphasis on
grammatical collocations during the experiment. However, in the case of ‘mechanics’, they did
not elaborate on the reasons for the reported improvement. In fact, such a relationship between
knowledge of collocation and ‘mechanics’ sub-scale (i.e. spelling, punctuation and capitalization)
does not sound justifiable.
Moreover, the findings of this study regarding the significant role of learners’ knowledge
of collocations in their writing development accord with those of Bahardoust (2013) and Tang
(2012). Similarly, Mounya (2010) stressed the significance of this role and suggested “a
Communicative-Collocational Approach to teaching writing (or speaking) in which the aim is to
teach writing following a communicative approach by concentrating on attracting students’
attention mainly to collocations” (p. 134). The findings of Namvar et al.’s (2012) study also
accentuated the importance of teaching collocations to L2 writers by noting that the sufficient
knowledge of these pre-fabricated chunks could save learners’ time and energy. In addition, their
stress about their limited knowledge of grammar and lexis, as an affective filter, which led into
poor writing performance, could be significantly reduced.
Inconsistent with the findings of this study in terms of the effectiveness of collocation
instruction on enhancing L2 writing quality are the results of the research studies conducted by
Avci (2006) and Liu (2000). The results of these studies indicated that although the employed
treatments for teaching collocation could help the learners to produce more correct collocations,
they did not effect any significant changes in the learners’ writing performance at the post-test
stage. However, these researchers attributed lack of improvement in their subjects’ post-
treatment writing to the interference of other factors such as the subjects’ level of language
proficiency, the likelihood of their inability to organize ideas in compositions as well as the
length of the study. The researchers, therefore, admitted that their findings should be generalized
with caution. Moreover, it can be also claimed that part of their failure was due to their employed
collocation teaching method which was limited to mere awareness of collocation. Such treatment
cannot bring about the desired learning results in the learners’ L2 output since the learners need
enough practice and corrective feedback in order to not only notice and note the collocations but
also learn how to incorporate them efficiently in their L2 production.

Conclusion
Both quantitative and qualitative findings indicated that the learners’ quality of writing
was significantly improved through collocation awareness-raising approach. The long-term
writing performance of the learners indicated that the employed treatment had durable effect. A
deeper analysis on separate aspects of the learners’ written production revealed that this
improvement was with respect to ‘vocabulary’ (in terms of the range of sophistication, lexical
choice and usage), ‘organization’ (in terms of fluency, the clarity of the expressed ideas and
succinctness) and ‘language use’ (in terms of the accuracy of agreement, tense and prepositions).
That is, the employed collocation teaching method did not affect the ‘content’ sub-component. It
sounds quite justifiable that developing knowledge of collocation cannot bring any significant
changes in this writing sub-component.
In all, these findings accentuate the crucial role of collocation awareness and use in
enhancing L2 writing competence and therefore strongly suggest teaching collocations through
Ying and Hendricks' (2003) proposed CAR model in writing classes. To put it more simply,
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 82

comparing the post-treatment essays of the experimental group with those of the control group
revealed the necessity of instructional intervention to help L2 learners notice, note and
incorporate collocations in their written output. In fact, the findings of this research show that
heightening the learners’ awareness of collocations and teaching them how to incorporate them in
their writing can lead into producing written works which include fewer unacceptable word
combinations and grammatical errors and enjoy a higher level of sophistication. Furthermore, the
clarity, succinctness and fluency of the expressed ideas increase.
The qualitative findings of this study also revealed that bringing some examples from the
learners’ L1 in the form of L1-L2 contrastive analysis helped the learners produce more correct
and natural-sounding word combinations. This, in turn, improved the fluency and accuracy of
their written production appreciably. Such findings suggest that using L1-L2 contrastive analysis
in writing classes can be one of the effective techniques in enhancing L2 learners’ knowledge of
collocations which consequently affects the quality of their writing positively.
On the whole, this study made a contribution to the series of empirical research studies
investigating the relationship between collocational competence and writing proficiency. The
present study, in fact, proved that such relationship exists. The findings of the current research,
therefore, can be of great benefit to language educators, including teachers, instructional
designers, L2 specialists and material developers, as well as language learners. The main
pedagogical implication of this study is that teachers should make learners aware of these word
combinations and teach them how to use them in their language production. Indeed, if learners
know that their writing competence is closely related to their collocational knowledge, they will
make more effort to increase their bank of collocations and they will become more willing to use
them.
Collocations do not receive the attention they deserve in the current textbooks and
instructional methodologies. The findings of this study show that they deserve more attention and
emphases. In fact, these findings suggest material developers produce writing textbooks which
mainly focus on the significance of collocations. They also suggest instructional designers devise
teaching methods in which collocations play the central role.
This study was limited to only one level of proficiency, that is, upper-intermediate. Future
research, with larger groups, can include L2 learners of higher or lower levels of language
proficiency. Furthermore, future experiments can be conducted in a longer period of time and
examine the long-term effect of the treatment within a longer interval between the immediate and
delayed post-tests or through more than one delayed post-test to determine its efficacy over time.
Moreover, this study focused on writing as one of the L2 productive skills. Future
researchers can examine the efficacy of the employed treatment on speaking. In addition, the
qualitative data collection technique employed in this study was semi-structured interviews with
the writing raters. Future researchers can collect the required qualitative data through some other
techniques, such as classroom observation, journal writing, etc. to gain more insight into the
issue.

References
Ary, D., Jacobs, L., Razavieh, A., & Sorensen, C. (2010). Introduction to research in
education (7th ed.). Canada: Tho mson Wadswort.
Ashouri, S., & Mashhadi Heidar, D. (2015). The Impact of Teaching Corpus-based
Collocation on EFL Learners' Writing Ability. International Journal of Foreign Language
Teaching and Research, 3(10), 53-62.
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 83

Avci, G. B. (2006). The effects of collocation instruction on writing (Unpublished

dissertation). Istanbul Technical University, English Preparatory Programm. Retrieved March 16,
2010, from www.bergeler.com
Bahardoust, M. (2013). Grammatical Collocation in Writing Production of EFL
Learners. The Iranian EFL Journal, 266.
Boonyasaquan, S. (2009). The lexical approach : An emphasis on collocations. Retrieved
June 1, 2010, from http://ejournals.swu.ac.th/index.php/hm/article/viewFile/626/627Creswell, J.
W. (2008).Educational research: planning, conducting, and evaluating quantitative and
qualitative research. Upper Saddle River, New Jersey.
Darvishi, S. (2011). The Investigation of Collocational Errors in University Students’
Writing Majoring in English. In International Conference on Education, Research and
Innovation 18 (12), 52-57.
Dastjerdi, H. V., & Samian, S. H. (2011). Quality of Iranian EFL Learners’ Argumentative
Essays: Cohesive Devices in Focus. Mediterranean Journal of Social Sciences, 2.
Eidian, F., Gorjian, B., & Aghvami, F. (2013). The impact of lexical collocation instruction
on developing writing skill among Iranian EFL learners. International Journal of Language
Learning and Applied Linguistics World, 4(3), 273-283.
Ellis, R. (1995). Interpretation tasks for grammar teaching. Tesol Quarterly, 29(1), 87-105.
Ghonsooli, B., Pishghadam, R., & Mahjoobi, A. M. (2008). The Impact of collocational
Instruction on the Writing Skill ofIranianEFL Learners: A Case of Product and Process Study.
Iranian EFL Journal, 2, 36-59.
Hill, J. (2000). Revising priorities: From grammatical failure to collocational success. In:
Lewis M (Ed), Teaching Collocation. Hove, England: Language teaching Publications, pp. 47–
70.
Hsu, J. (2007). Lexical collocations and their relation to the online writing of Taiwanese
college English majors. Electronic Journal of Foreign Language Teaching, 4(2), 192-209.
Ismail, S. A. A. (2011). Exploring students’ perceptions of ESL writing. English Language
Teaching, 4(2), 73.
Jacobs, H., Zingraf, S., Wormuth, D., Hartfiel, V. F., & Hughey, J. (1981). Testing ESL
composition: A practical approach. MA: Newbury House Publishers.
James, C. (1994). Explaining grammar to its learners. In: Bygate, M., Tonkyn, A. &
Williams E. (Eds.), Grammar and the Language Teacher. Hertforshire: Prentice Hall
International Ltd, pp. 203-214.
Krashen, S. D. (1989). We acquire vocabulary and spelling by reading: Additional
evidence for the Input Hypothesis. The Modern Language Journal, 73, 440-464.
http://dx.doi.org/10.1111/j.1540-4781.1989.tb05325.x
Krueger, R. A. (1994). Focus Groups: A Practical Guide for Applied Research. Thousand
Oaks, CA: Sage Publications.
Laufer, B. & Girsai, N. (2008). Form-focused instruction in second language vocabulary
learning: A case for contrastive analysis and translation. Applied Linguistics, 29(4), 694-716.
Lewis, M. (1993). The lexical approach. Hove: Language Teaching Publications.Lewis, M.
(1997). Implementing the lexical approach: Putting theory into practice. Hove, England:
Language Teaching Publications.
Lewis, M. (2000). Language in the lexical approach. In: Lewis, M. (Ed.), Teaching
collocation: Further developments in the lexical approach. London: Language Teaching
Publications, pp.155-184.
 International Journal of Foreign Language Teaching & Research – Volume 4, Issue 16, Winter 2016 84

Lewis, M. (2001). There is nothing as practical as a good theory. In: Lewis M (Ed.),
Teaching collocation: Further developments in the lexical approach. London: Language
Teaching Publications, pp. 10-27.
Liu, C. P. (2000). An empirical study of collocation teaching. Proceedings of the
seventeenth conference on English teaching and learning in the Republic of China. Taipei:
Crane, pp. 165-178.
Martyńska, M. (2004). Do English language learners know collocations? Investigationes
Linguisticae 6, 1-12.
McMillan, J.H., & Schumacher, S. (2010). Research in education: Evidence-based inquiry
(7th ed.). US: Pearson Education.
Mounya, A. (2010). Teaching lexical collocations to raise proficiency in foreign language
writing (Unpublished master’s thesis). Guelma University, Algeria.
Nation, I.S.P . (2001). Learning vocabulary in another language. Cambridge: Cambridge
University Press.
Namvar, F., Nor, N. F. M., Ibrahim, N., & Mustafa, J. (2012). Analysis of collocations in
the Iranian postgraduate students’ writings. 3L; Language, Linguistics and Literature, The
Southeast Asian Journal of English Language Studies, 18(1), 11-22.
Sadoughvanini, H. (2012). The Relationship between Use of Collocations and
Expressiveness: Fulfilling the Referential Function of Language. International Journal of English
Language and Literature Studies, 1(2), 28-37.
Schmidt, R., & Frota, S. (1986). Developing basic conversational ability in a second
language: A case study of an adult learner of Portuguese. Talking to learn: Conversation in
second language acquisition, 237-326.
Stoitchkov, R. (2008). Building collocation competence. Retrieved April 15, 2010, from
http://www.beta-iatefl.org/1749/blog-publications/building-collocation-competence/
Tang, J. (2012). An Empirical Study on the Effectiveness of the Lexical Approach to
Improving Writing in SLA. Journal of Language Teaching and Research, 3(3), 578-583.
Woolard, G. (2000). Collocation – encouraging learner independence. In: Lewis, M. (Ed.),
Teaching Collocation. Further developments in the lexical approach. Hove: Language Teaching
Publications, pp. 28-46.
Ying, Y., & Hendricks, A. (2003). Collocation awareness in the writing process.
In international conference, Hong Kong University.