TELL, Vol. 5, No. 1, 2011, pp.

45-71

Consciousness-raising versus explicit grammar

teaching and their impact on Iranian EFL learners’
lexical inferencing
Mojgan Rashtchi1
Associate Professor, Islamic Azad University, North Branch
Saiideh Jalili
M.A. in TEFL, Islamic Azad University, North Branch

Received July 28, 2011

Accepted April 25, 2012

Abstract
The present study attempted to investigate the impact of
sentence-level grammar instruction, explicitly and through
consciousness-raising, on Iranian EFL learners’ lexical
inferencing behavior and subsequent learning of inferred
vocabulary. It also aimed at examining the relationship
between the number of used knowledge sources and success
in lexical inferencing. To address these issues, a quasi-
experimental study with a mixed design involving two
treatment groups and one control group was conducted. The
participants were 47 adult learners at intermediate
proficiency level. A counterbalanced design was utilized to
eliminate the effect of the topic familiarity on the learners’
performance. Retrospective data were collected from eight
participants, as well. Further, the eight participants were
divided into two groups (successful and less successful) in

1
Corresponding author at: Islamic Azad University, North Branch
Email address: [email protected]
46 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

order to identify the differences between the two groups’

lexical inferencing behavior. The findings indicated that
sentence-level grammar instruction could place learners in a
better position to infer the meaning of unfamiliar words
more successfully and also could help them recall the
inferred vocabulary better. The results are significant due to
the controversy surrounding teaching grammar; moreover,
they confirm the positive role of the number of knowledge
sources in the lexical inferencing success.

Keywords: consciousness-raising, explicit teaching, lexical

inferencing, lexical inferencing success, retention,
retrospective data, sentence-level grammar

1. Introduction
The most preferable strategy among EFL learners is lexical
inferencing when exposed to incomprehensible input (Dycus, 1997;
Fraser, 1999; Paribakht, 2004; Paribakht & Wesche, 1999).
Nevertheless, learners’ lexical inferencing is not always successful
and sometimes goes astray (Bensoussan & Laufer, 1984; Frantzen,
2003; Huckin & Bloch, 1993; Hulstijn, 1992; Kaivanpanah & Alavi,
2008a; Nassaji, 2003). Determining a wrong meaning for an
unknown word and consequently misunderstanding the whole text is
not the only inherent drawback of lexical inferencing. The worse is
that learners “may learn the wrong meaning which they have to
unlearn subsequently” (Hulstijn, 1992, p.114). Another problem
touched upon by Hulstijn (1992) is that although the possibility of
learning the meanings of inferred vocabularies is higher than
learning those words which their meanings have been presented to
learners, they still have a slight chance of recalling the meanings of
words upon a single exposure.
With regard to these studies, the present study is an attempt to
make an empirical contribution to the assumption proposed by
Bengeleil and Paribakht (2004) that teaching learners the knowledge
 TELL, Vol. 5, No. 1, 2011 47
Rashtchi and Jalili

sources they use while struggling to infer the meaning of unknown

words may increase the chance of more accurate guesses.
Specifically, the study focuses on teaching sentence-level
grammatical knowledge which is, according to Paribakht (2004), the
major knowledge source used by learners. The study, also, tries to
manifest whether teaching sentence-level grammar explicitly or
through consciousness-raising is more beneficial in improving
learners’ grammatical knowledge in dealing with unfamiliar words
while reading. Finally, this study aims at finding the relationship, if
any, between the number of knowledge sources which learners use
and their lexical inferencing success.

2. Background
EFL learners have different aims for learning English such as
performing better on academic tasks, learning more about subject
matters, or improving language skills. Whatever their goal might be,
if they want to be autonomous language learners, reading is the
main means (Grabe & Stoller, 2001) and is the most important way
for transmitting academic knowledge (Huckin & Bloch, 1993).
Reading as a language skill used for obtaining knowledge is of high
importance for students. As Richards and Renandya (2002) state,
reading comprehension is the only skill that most EFL students ever
need during their lives. Furthermore, most foreign students rely on it
to compensate for the weaknesses in their speaking ability during a
course of study (Huckin & Bloch, 1993). Lee and Krashen (as cited
in Krashen, 2003) contend that the more you read, the less you have
writing apprehension. One reason for such significance can be that
it provides learners with various models of language and introduces
new cultural issues to them. Further, through reading, learners are
exposed to vocabulary, grammar, idioms, proverbs, and culture that
are the essential components of language. Therefore, it is an
advantageous way for bringing EFL learners into contact with
language and consequently increasing the chance of learning new
materials. Moreover, learners can do reading activities on their own
at any time and wherever they are. However, when learners are
48 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

exposed to authentic language, incomprehensible input is inevitable

(Kaivanpanah & Alavi, 2008b).
Nation (1996) and Hedge (2000) confirm that one of the major
obstacles to comprehending new materials is unknown vocabularies.
In other words, how well learners are able to read in a foreign
language depends heavily on their vocabulary repertoire. In the
same line of argument, Schmitt (2008), reviewing current research
on second language vocabulary learning (Bonk, 2000; Schmitt &
Zimmerman, 2002), concludes that “learners must learn a very large
number of lexical items to be able to operate in English” (p. 332).
To make it more challenging, most learners do not have such a good
command of vocabulary knowledge (Schmitt, 2008).
Hence, to compensate for such lack of knowledge exposure is
not sufficient; learners should be equipped with strategies to make
use of learning opportunities. Thus, teachers should not only
maximize the amount of exposure but also pave the way for better
understanding by teaching effective strategies to their learners.
Research on lexical processing strategies indicates that learners use
different strategies when they face new words including ignoring
unknown words, consulting a dictionary or another person, and
inferring meaning from context (Fraser, 1999). Fraser (1999) and
Paribakht (2004) have found that the participants in their studies use
lexical inferencing in 58% and 80% of the cases respectively when
they encounter unknown words while reading.
Hasstrup (1991) explains lexical inferencing as “making
informed guesses as to the meaning of a word in the light of all
available linguistic cues in combination with the learner’s general
knowledge of the world, her awareness of the co-text and her
relevant linguistic knowledge” (p. 40). She considers “inferencing
as a central procedure in receptive language use as well as in
language learning” (p. 16). Dunmore (1989) states that lexical
inferencing is a catalyst in the dynamic and reversible relationship
of vocabulary and reading; that is, reading broadens vocabulary
repertoire and vocabulary, in turn, facilitates reading. Decarrico
(2001) calls guessing meaning from context a “self-study strategy”
for learning vocabulary and adds that lexical inferencing is
necessary “as students are encouraged to make the transition to
 TELL, Vol. 5, No. 1, 2011 49
Rashtchi and Jalili

independent learning by determining meanings of the less frequent

words they read or hear” (p. 290). Similarly, Liu and Nation (1985)
are in favor of lexical inferencing especially for determining the
meaning of low frequency words in written texts. They state that
such words “because of their narrow range and the low probability
of meeting them again soon” are not worth effort and time (p. 33).
Hasstrup (1991) contends that lexical inferencing in reception is
comparable to strategic competence in production in the sense that
as strategic competence helps learners fill gaps in their productive
vocabulary, lexical inferencing, likewise, helps them fill gaps in
receptive vocabulary.
Drawing on the work in cognitive psychology (Craik & Lockart,
1972; Craik & Tulvig, 1975), Hedge (2000) suggests that the more
actively learners work on words, the more likely they retain them.
In other words, retentiom of an inferred word meaning is superior to
retention of a given word meaning. In Hulstijn and Laufer’s (2001)
terminology “words that are processed with higher involvement
load will be retained better than words that are processed with lower
involvement load” (p. 552). Their motivational-cognitive construct
of involvement comprises three components: need, search, and
evaluation. The higher learning stands to the reasons that depth of
processing enhances the probability of transferring a word from
short-term to long-term memory (Decarrico, 2001) and also
increases the amount of input which becomes intake. Besides, Hu
and Nassaji (in press) have discovered that there is an inverse
relationship between ease of inferencing and retention.
In spite of its dominance and popularity, unfortunately, lexical
inferencing is not always an effective strategy. Learners may make
wrong guesses that may lead to not only failure in the
comprehension of the whole text but also to learning wrong
meanings. Liu and Nation (1985) believe that for inferencing
strategy to apply successfully, a large proportion of words in a text
(at least 95%) should be known to learners. However, Nassaji
(2003) believes that despite knowing 19 out of 20 words, his
participants’ rate of inferential success has still been low.
Fraser (1999) maintains that incidental vocabulary learning
happens through lexical inferencing in the course of reading for
50 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

comprehension. Vocabulary expansion through incidental

vocabulary learning has received widespread support in L2
literature (Hulstijn, 1992; Hunt & Beglar, 1998; Paribakht &
Wesche, 1997; Schmitt, 2008). Schmitt (2000) supports incidental
vocabulary learning to such an extent that he asserts “a viably-sized
vocabulary cannot realistically be taught exclusively through
explicit study” (p. 143). Paribakht (2005) takes a step further and
states that lexical inferencing is “the first stage in learning a new
word” (p. 702).
Furthermore, as Bengeleil and Paribakht (2004) argue, making
L2 learners aware of different types of knowledge sources and
contextual cues that can be used in inferring the meaning of
unknown words may help them have more accurate guesses. In fact,
having a good command of metalanguage can facilitate making
successful inferences. That’s to say, if learners are equipped with
explicit knowledge of the nonlinguistic and linguistic features used
in lexical inferencing, the chance of inferring true meanings may be
increased. In addition, Alavi and Kaivanpanah (2009) propose
consciousness-raising as a way to increase the likelihood of
learners’ lexical inferencing success, particularly that of field-
dependent ones. They state that, “contextual clues can be made
more salient through underlining or different type faces” (p. 4). In
this way learners pay more attention to details in context which is a
prerequisite for more effective guesses. In fact, according to
Schmidt’s (1990) Noticing Hypothesis, implicit and explicit
knowledge connect through noticing or awareness. However, as
Walters (2004) points out, empirical studies in this area are sparse.
Kern (1989) has investigated the effectiveness of explicit teaching
of second language reading comprehension strategies on improving
L2 learners’ abilities in inferring the meaning of unknown words.
The results of the study have revealed that students especially less
proficient ones could benefit from instruction on inferencing
strategy. Additionally, Huckin and Jin (1987) have attempted to
investigate the efficacy of training in using context. In their study,
the experimental group receives a 15-minute training in guessing
from context consisting of reading the pretest passage and
explaining the context clues. The results of the posttest show that
 TELL, Vol. 5, No. 1, 2011 51
Rashtchi and Jalili

although both experimental and control groups have improved, the

experimental group outperforms the control group. Also, Fraser
(1999) has made an attempt to empirically test the effects of
instruction on lexical processing strategies (i.e. ignoring, consulting,
and inferring) and their subsequent impact on vocabulary learning.
The results of the study signify that instruction is effective in
improving students’ guesses and can decrease the number of times
the participants ignore unknown words; instead, inferencing is used
more frequently and thus vocabulary learning is enhanced. Yet, as
Walters (2004) points out, since students have had other options at
their disposal (i.e. ignoring or consulting); it is difficult to isolate
the effects of instruction on inferencing.
Contrary to the results of the above-mentioned studies, strategy
training, as Nassaji (2004) states, is not always successful (e.g.,
Barnett, 1988). He claims that although strategy training is
necessary, it is not enough and maintains that lexical inferencing
success depends on many other variables such as proper use of
strategies and knowledge sources. Learners’ language and
comprehension abilities, types of tasks, texts, and words are also
involved. Following this line of argument, the present study focused
on teaching sentence-level grammatical knowledge.
Sentence-level grammatical knowledge, according to Paribakht
(2004), is the major knowledge source which learners use while
inferring the meaning of unfamiliar words. It seems that by teaching
explicitly or through consciousness-raising sentence-level
grammatical knowledge, the chance of successful meaning
inference and learning vocabularies may increase, as well. Teaching
grammar appears to help learners internalize grammatical
knowledge easier. Paribakht (2004) using concurrent think-aloud
and immediate retrospective data collection procedures
demonstrates that EFL readers benefit from two sources of
knowledge in their attempts to infer the meaning of unknown
words, that is, linguistic and extralinguistic knowledge sources.
Linguistic knowledge can be further divided into L2-based
linguistic knowledge (intralingual sources) and L1-based linguistic
knowledge (interlingual sources). Intralingual sources include
sentence-level grammatical knowledge (SG), word morphology
52 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

(WM), punctuation (P), discourse/text (D), homonymy (H), word

association (WA). It is noteworthy that this taxonomy was used as a
framework in the present study. Nassaji (2003), analyzing
introspective and retrospective think-aloud protocols of 21
intermediate ESL learners, distinguishes between the use of
knowledge sources and strategies. He defines strategies as,
“conscious cognitive or metacognitive activities that the learner
use[s] to gain control over or understand the problem without any
explicit appeal to any knowledge source as assistance” (p. 655).
Thus, having considered the long-standing controversy over
inferencing strategy, along the aforementioned line of inquiry, the
present study tried to throw light on how to possibly reduce its
potential disadvantage, that is, wrong guesses and consequently
learning wrong meanings of words. Examining the relationship
between the number of used knowledge sources and lexical
inferencing success and finding the knowledge sources used more
frequently by successful learners, the study also attempted to help
less successful learners. To investigate the issues at hand the
following research questions were proposed:

1. Does sentence-level grammar instruction explicitly and through

consciousness-raising affect Iranian EFL learners’ lexical
inferencing success in seen and unseen texts?
2. Does sentence-level grammar instruction explicitly and through
consciousness-raising affect the retention of inferred
vocabularies by Iranian EFL learners?
3. Is there any relationship between the number of knowledge
sources that Iranian EFL learners use and their success in lexical
inferencing?

3. Method
3.1 Participants
A total number of 47 university undergraduates (13 males and 34
females) between the ages of 19 and 24 in three intact classes were
selected based on convenience sampling and randomly assigned to a
 TELL, Vol. 5, No. 1, 2011 53
Rashtchi and Jalili

control group (n = 14) and two experimental groups (n = 17 & n =

16). They were all Persian native speakers and their English
proficiency was at the level of Interchange, book 2. The participants
were selected from one single language school to control the
intervening effect of contextual factors such as school atmosphere
and principal leadership on students’ achievement. Classes met
twice a week for 11 weeks and each session lasted for 105 minutes.

3.2 Instrumentation
To check the homogeneity of the participants, Interchange/Passages
Placement Test was used. Since the focus of this study was on
reading, the reading subsection of the ESOL Preliminary English
Test (PET) was also run. Two texts namely ‘Big City Dreams’ and
‘Genetic Engineering’ developed at the same difficulty level by
Paribakht (2005) were used as the pre and posttests to measure
lexical inferencing success. Similar to Nassaji’s (2004) study, in
each text 10 unknown words were used to focus on EFL learners’
inferencing behavior. The target words were all content words
consisting of nouns, verbs, adjectives, and adverbs. To make sure
that the words were unfamiliar to the participants, a list of the words
was given to the participants before the pre and posttests and they
were asked to write down their meanings (either in Persian or
English). Subsequently, the known words were omitted for those
participants.
Moreover, Vocabulary Knowledge Scale (VKS) developed by
Paribakht and Wesche (1996, 1997) was used to monitor the effect
of grammar instruction, both explicitly and through consciousness-
raising, on the retention of inferred vocabularies by EFL learners
and were given ample time to complete the task. This instrument,
utilized by Wesche and Paribakht (1996), elicits self-perceived and
demonstrated knowledge of the target words in written form from
learners with five self-report and performance items. The scale
ratings range from complete unfamiliarity, through recognition of
the word and recall of its meaning, to the ability to use the word
productively in a sentence with grammatical accuracy and semantic
appropriateness. According to Paribakht and Wesche (1996), the
strength of VKS lies in the fact that, unlike most tests of vocabulary
54 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

size, particularly YES/NO type, VKS allows verification by

eliciting actual knowledge of learners’ vocabulary with performance
items. Elicitation of this actual knowledge makes VKS more
trustworthy than other vocabulary size tests which are only based on
“learners’ dubious knowledge” (p. 34). Furthermore, the fact that
vocabulary learning is an incremental process (De Bot, Paribakht, &
Wesche, 1997) makes the VKS even a better choice for measuring
incidental word learning. The reason behind this is that the VKS is
sensitive enough to capture the initial development of basic
knowledge of specific words (Wesche & Paribakht, 1996). It is
noteworthy that the lists of target words given to the participants
prior to the main passages in the pre and posttests were initially in
the format of VKS. Nonetheless, since the researchers’ intention
was to have only those words that were completely unfamiliar to the
participants, following Fraser’s procedure, the five-point VKS score
was collapsed to two-point score (familiar vs. unfamiliar). That is
why the word ‘list’ was used for those tests that were given to
participants before the main passages.
Another instrument utilized in this study was immediate
retrospective think-aloud protocols. The data about the use of
knowledge sources were derived from immediate retrospective
protocols. The passage used to elicit a retrospective think-aloud
protocol of used knowledge sources was the one developed by
Haastrup (1991) named ‘Health in the Rich and Poor’. It had 25
content words underlined as target words. First, the participants
were given a list of the target words and were asked to write down
the meaning of the familiar words. Second, they were given the
main passage wherein they were to write down the meaning of the
underlined words. They had as much time as they needed to
complete the task. Finally, the participants were interviewed
individually and were asked to verbalize and report on their
thoughts as they were solving the lexical inferencing task.

3.3 Procedure
As mentioned earlier, Interchange/Passages Placement Test was
administered to three intact classes to examine whether they
belonged to the same population. Since the construct of interest in
 TELL, Vol. 5, No. 1, 2011 55
Rashtchi and Jalili

this study was reading, the reading subsection of the PET was also
administered to ensure that the participants were homogeneous in
their reading ability. Furthermore, to eliminate the effect of
teachers’ experience, teachers were chosen from novice English
teachers. To this end, Tschannen-Moran and Hoy’s (2007) cut-off
point for distinguishing between novice and experienced teachers
(i.e. three years of teaching experience) was followed. Thus, the
teachers’ experience ranged from nine months to two years.
Besides, all the three teachers held M.A. in English-related majors.
The control group was taught by one of the researchers.
The three intact classes were randomly assigned to a control
group and two experimental groups. In sessions one to three all of
the participants in the three groups took the placement test, were
trained on lexical inferencing, and took the pretest respectively. The
instruction on lexical inferencing strategy focused on developing
students’ awareness of the use and viability of lexical inferencing.
They were taught how to use inferencing as a strategy in their
reading when encountering unfamiliar words using Clarke and
Nation’s (1980) model:

1. identifying the part of speech of the unknown word

2. looking at the immediate context
3. looking at the wider context
4. I) inferring the meaning of unknown word
II) checking the guess
(a) checking that the guess is the same part of speech as
the unknown word
(b) checking if the word can be analyzed into prefix, root,
or suffix
(c) substituting the guess for the unknown word in the
context and then looking it up in a dictionary

As the pretest, the participants were given two passages to read

and infer the meaning of the underlined words while they were free
to ask the meaning of other unfamiliar words in the passage (i.e.
those not underlined). One half of each class took the passage
named ‘Big City Dreams’ while the other half took ‘Genetic
Engineering’. The rationale was to eliminate the effect of the topic
56 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

familiarity on the students’ performance which has been proven to

affect lexical inferencing and retention of inferred vocabulary
(Pulido, 2007). The reason for the inclusion of an unseen text (i.e.
the counterpart of the pretest) in the posttest was to counteract any
possible practice effect of having the pretest before the treatment.
Sentence-level grammar instruction for the experimental groups
was integrated in the regular content of their course. The instruction
started from the fourth session and took about 30 minutes each
session. It focused on building up students’ sentence-level
grammatical knowledge which is one of the major knowledge
sources used by learners when they try to construct the meaning of
unknown words (Paribakht, 2004). The learners received instruction
on the speech parts (i.e. verb, noun, adjective, adverb, preposition,
determiner, pronoun, and linking word) and the syntactic
relationship among the words within a sentence like word order and
word class. The only difference was that the first experimental
group (ETG) was taught sentence-level grammar explicitly and the
second experimental group (CRG) was instructed through
consciousness-raising tasks. The control group (CG), however,
practiced lexical inferencing through its common method; that is,
the participants were encouraged to guess the meanings of unknown
words from context rather than use their dictionaries.
ETG was instructed based on Oxford Practice Grammar by
Eastwood (2006). In this group, the grammatical points were taught
explicitly; that is, the teacher provided the learners with
metalinguistic explanations of the target grammatical points and
then the learners were engaged in doing some exercises. The CRG,
on the other hand, received a reading passage taken from Select
Readings (Lee & Gundersen, 2002) in which a related grammatical
point was underlined. Also, some other grammatical points were
taught to the group through some inductive consciousness-raising
tasks which were adapted from Ellis (1993, 2002). To do these
exercises, the class either acted as a group or was divided into small
groups. Alternatively, the CG was given a passage each session and
the learners were encouraged to guess the meaning of unfamiliar
words from the context not using their dictionaries.
 TELL, Vol. 5, No. 1, 2011 57
Rashtchi and Jalili

The posttest was administered in session 18 to the groups.

Followed by the list of the target words the main passages, which
were the counterpart of what the participants had received before
the instruction, were given to them. Afterwards, the participants
were given the same passage they had as the pretest and were asked
in their mother tongue to infer the meaning of unknown words
while there was no time limit for task completion. Moreover, they
were free to ask the meaning of any unknown words except the
target ones.
Two weeks later, the VKS comprising 20 words (i.e. the target
words in the posttests) was administered to the three groups to
assess the effect of sentence-level grammar instruction, either
explicitly or through consciousness-raising, on the retention of
inferred vocabulary. To reduce the contextual clues, the listed words
were presented in a scrambled order. It is worthwhile to mention
that the chance of meeting the target words outside testing sessions,
as Hulstijn and Laufer (2001) suggest, was low on the grounds that
the target words were low frequent and English is a foreign
language in Iran.
Finally, at the end of the term, immediate retrospective think
aloud data collection procedure was carried out in the ETG in
individual sessions. Nine students were asked to read the passage
‘Health in the Rich World and in the Poor’ developed by Haastrup
(1991) and tried to infer the meaning of the underlined words with
no time limit for completing the task. Subsequently, they were
asked in individual sessions to verbalize every knowledge source
they had used to guess the meaning of an unknown word. At the
beginning of each session, the participants were trained in think-
aloud procedures. Each individual session conducted in the
participants’ first language (Persian) took about an hour. All
auditory evidence was recorded by the participants’ permission and
transcribed for further analysis. Paribakht’s (2004) taxonomy of
knowledge sources was used to examine think-aloud protocols for
different kinds of knowledge sources and the frequency of use. It
should be noted that six participants knew the meanings of nine
target words altogether and three of these words were familiar for
only one participant. To save more words, instead of removing the
58 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

nine words, six words and the participant with the knowledge of the
three words were excluded from the study. Eventually, this stage
was carried out with total number of eight participants and 19
words.
Accordingly, all the answer-sheets gathered from 47 participants
were scored by one of the researchers. To measure the degree of
success in lexical inferencing in the three passages (i.e. the pretest,
posttest, and think-aloud task), a three-point scale (0 to 2)
representing unsuccessful, partially successful, and successful
inferences respectively was used. Successful inferences were
defined as those that were semantically, syntactically, and
contextually appropriate. They could be synonyms or definitions.
Answers that were semantically appropriate but syntactically
deviant, or vice versa, or were approximate were classified as
partially successful. Responses were evaluated as unsuccessful
when they did not meet any of these conditions or no meaning was
offered at all. In this way, the maximum score possible for each
participant was 20, the results of which were compared after adding
them up.
To score VKS items, Paribakht and Wesche’s (1997) guidelines
for scoring VKS were followed. The minimum and maximum
scores possible for any participant were 20 and 100. It is noteworthy
that more or less all students inferred the meaning of some target
words approximately or even wrongly but they remembered exactly
those inferred meanings in the vocabulary test. In scoring VKS, all
inferred meanings which students could recall successfully were
considered correct even if they were approximate or incorrect
meanings since what counted was the retention of inferred meanings
no matter whether they were right or wrong. The logic behind is that
of Fraser’s (1999) framework, who asserted that vocabulary
learning from inferred vocabulary happened to the extent that
learners could recall those meanings which they assigned to
unknown words in reading before. However, in scoring students’
answer-sheets in the pretest, posttests, and VKS those target words
which participants had indicated as known were deleted for those
participants.
 TELL, Vol. 5, No. 1, 2011 59
Rashtchi and Jalili

4. Results
The data obtained from Interchange/Passages placement test,
reading component of the PET, and the pretest to three separate one-
way ANOVAs confirmed the homogeneity of the three groups with
respect to their general proficiency, reading ability, and lexical
inferencing prior to the instruction. Table 1 displays the mean
scores and standard deviations for the three groups’ performance on
the pretest, posttests, and the VKS.

Table 1: Descriptive statistics of the groups’ performance on the

pretest, posttests, and VKS

Groups Test Text M SD N

Pretest 6.74 1.71 17
Seen 12.33 2.76 17
ETG Posttest Unseen 9.04 3 17
Total 10.68 3.30 34
VKS 39.06 6.22 17
Pretest 6.78 1.81 16
Seen 10.35 3.17 16
CRG Posttest Unseen 8.60 2.07 16
Total 9.47 2.78 32
VKS 37.94 4.69 16
Pretest 5.50 1.79
Seen 5.97 1.95 14
CG Posttest Unseen 4.98 1.99 14
Total 5.47 2 28
VKS 32.14 7.09 14

Furthermore, to address the first research question a two-way

between-groups analysis of variance was carried out. As it is
apparent from Table 3 below, the interaction effect between
sentence-level grammar instruction and text type was not
statistically significant, F (2, 88) = 1.62, p = 0.204. There was a
statistically significant main effect for sentence-level grammar
instruction, F (2, 88) = 333.53, p = 0.001. The effect size, using eta
squared was 0.432, indicating a relatively large effect size, which
means that teaching grammar by itself accounted for 43.2% of the
overall variance. In order to see which mean differences were
60 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

significant the Scheffe test was performed. The results revealed that
the mean scores for the ETG and CRG were significantly different
from the CG. The ETG and CRG did not differ significantly from
each other. The main effect for text type [F (2, 88) = 14.22, p =
0.0005] was also significant. The obtained η2 value was 0.139 which
can be said to be a large value. According to post-hoc comparisons
of the three groups the CG was significantly different from both
ETG and CRG although ETG and CRG did not differ significantly
from each other.

Table 2: One-way analysis of variance of the groups’

performance on the retest

SS df MS F Sig.
Between
15.61 2 7.80 2.5 .09
Groups
Pretest
Within Groups 137.48 44 3.12
Total 153.1 46

Table 3:Tests of between-subjects effects

Dependent variable: Lexical inferencing success
Source SS df MS F Sig. Partial Eta
Squared
Corrected 567.44a 5 113.49 17.13 .000 .493
Model
Intercept 6814.90 1 6814.90 1028.415 .000 .921
Group 444.37 2 222.18 33.529 .000 .432
Text 94.23 1 94.27 14.219 .000 .139
Group*Text 21.468 2 10.73 1.620 .204 .036
Error 583.140 88 6.63
Total 8296.27 94
Corrected 11.50.58 93
Total
a. R Squared = .493 (Adjusted R Squared = .464)

To address the second research question, that is, whether teaching

sentence-level grammar explicitly and through consciousness-
raising affect the retention of inferred vocabularies, the data from
the students’ performance on the VKS were submitted to a one-way
ANOVA. The results presented in Table 4 show that the mean
performance of the three groups was different, [F (2, 44) = 5.63, p =
 TELL, Vol. 5, No. 1, 2011 61
Rashtchi and Jalili

0.007, η2 = 0.203]. As a result, it may be concluded that grammar

instruction promoted the retention of inferred vocabulary. The
Scheffe test, again, was run to trace where the significant
differences laid. The results revealed that there was a significant
difference, on the one hand, between the ETG and the CG, and on
the other hand, between the CRG and the CG; though, there was no
significant difference between ETG and CRG.

Table 4: One-way analysis of variance of the groups’

performance on the VKS

SS df MS F Sig.
Between
409.63 2 204.814 5.63 .007
Groups
VKS
Within Groups 1601.95 44 36.408
Total 2011.58 46

Finally, to address the third research question, all the immediate

retrospective think-aloud protocols were transcribed and examined
qualitatively for any knowledge source. To code knowledge
sources, Paribakht’s (2004) taxonomy was used as a framework.
Successively, the number of knowledge sources which learners
drew upon and their scores on the third passage were calculated.
Then, the scores on lexical inferencing and the numbers of used
knowledge sources were correlated using Spearman correlation. The
results revealed that there was a strong, positive correlation between
the two variables r = 0.93, p = 0.001 with greater numbers of used
knowledge sources associated with greater success of lexical
inferencing.
In order to get additional information, the eight participants
were divided into two successful and less successful groups
according to their percentile rank on lexical inferencing scores
obtained from the think-aloud task. There were four participants in
each group. Therefore with 19 target words and four participants in
each group there were 76 cases to guess in each group. Table 5
displays the type, frequency, and percentage of knowledge sources
which the participants in each group called upon when inferred the
unfamiliar words. Counting the number of the used knowledge
sources by each group, it is clearly evident that the successful group
62 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

used more knowledge sources than the less successful group (139
vs. 93).

Table 5: Relative frequencies of knowledge sources used by

successful and less successful learners
Type of knowledge Successful vs. less Frequency of % of knowledge
sources successful knowledge sources sources
Sentence-level Successful 71 50.08
grammatical knowledge Less successful 56 60.22
Successful 21 15.11
Word morphology
Less successful 10 10.75
Successful 0 0
Punctuation
Less successful 0 0
Successful 26 18.71
Discourse
Less successful 17 18.28
Successful 10 7.20
Homonymy
Less successful 3 3.22
Successful 2 1.44
Word association
Less successful 0 0
Successful 0 0
Cognate
Less successful 0 0
Successful 9 6.47
World knowledge
Less successful 7 7.53
Successful 139 100
Total
Less successful 93 100

As it is apparent from Table 6 both groups used an almost similar

sequence of relative knowledge source frequency while deriving the
meaning of unknown words (with the exceptions of WK and P).

Table 6: The order of relative frequency of knowledge sources use

Groups Knowledge sources

SG > D > WM > H > WK > WA >
Successful
P=C
SG > D > WM > WK > H > WA =
Less successful
P=C
SG > D > WM > WK > H > WA >
Total
P=C

As stated before, a list of target words was given to the participants

to write down the meanings of known words (if any) to ensure that
 TELL, Vol. 5, No. 1, 2011 63
Rashtchi and Jalili

the target words used in the think-aloud were unfamiliar to them. In

data analysis, the known words were excluded for all of the
participants. However, some learners provided wrong meanings for
some words in the decontextualized list. A more detailed analysis of
these wrong responses to the target words in the decontextualized
list and the main passage revealed two patterns in their responses.
One pattern was that they mistook the words for other similar-
looking but semantically irrelevant words for example squalor for
solar and scholar, precipitating for perception, and permeate for
permit which is called synophone/synograph by Bensoussan and
Laufer (1984) and mistaken ID by Huckin and Bloch (1993).
Another pattern was that sometimes learners misanalyzed a word to
its morphological components for instance translating waver as the
sum of wave and er called morphological troublemaker by
Bensoussan and Laufer (1984).

Table 7: The comparison between responses to the

decontextualized list and the think-aloud task

Restructuring lexical knowledge

Partially
Successful Unsuccessful Total
successful
Freque Freque Freque Freque
% % % %
ncy ncy ncy ncy
Gro
Succes 9. 6. 1. 17.
ups 7 5 1 13
sful 21 58 32 11
Less
1. 2. 3.9
succes 1 2 0 0 3
32 63 5
sful

5. Discussion
The findings of the present study revealed that teaching sentence-
level grammar both explicitly and through consciousness-raising
result in learners’ significant success in applying lexical
inferencing. These findings lend support to Paribakht (2004) and
Kaivanpanah and Alavi’s (2008b) findings that grammatical
knowledge contributes to lexical inferencing success. Paribakht
(2004) in her study discussed the possible effect that grammatical
64 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

knowledge might have on vocabulary learning; however, she did not

directly address it. Therefore, this study provides valid grounds that
sentence-level grammatical knowledge may influence the retention
of inferred vocabulary. All in all, these findings add support to
Schmitt’s (2000) suggestion that “lexical knowledge and
grammatical knowledge are inextricably interrelated in a kind of
lexicogrammar” (p. 58). The findings also provide evidence for
Bengeleil and Paribakht’s (2004) assumption that teaching learners
the knowledge sources which are called upon in lexical inferencing
while reading increase the likelihood of an efficient and accurate
guess. The result that CRG outperformed the CG expands upon
Alavi and Kaivanpanah’s (2009) speculation that through
consciousness-raising activities it is likely that learners pay more
attention to details that consequently lead to more successful lexical
inferences. The finding that ETG and CRG performed equally is in
line with Fotos and Ellis’s (1991) and Fotos’ (1994) studies that
both metalinguistic descriptions of the target features and
consciousness-raising tasks are effective methods to enhance
learners’ grammatical knowledge. Although the effect sizes of 0.432
and 0.139 for the posttests attested that teaching sentence-level
grammar is effective for increasing the likelihood of a correct guess,
a glance at the mean scores in the posttests of both ETG (seen and
unseen texts 12.32 and 9.04 respectively) and CRG (seen and
unseen texts 10.35 and 8.60 respectively) reveals that the mean
scores are not high. It can be concluded that desirable results are
achieved when teaching sentence-level grammatical knowledge is
coupled with instruction on other knowledge sources. Likewise,
although, the effect size of 0.203 for VKS testified the significant
effect that sentence-level grammar instruction can have on
promoting the retention of inferred vocabulary, the mean scores of
the ETG and CRG are not high. Bearing in mind that learners in this
study were exposed to the target vocabulary only two times, the
medium mean scores (39.06 and 37.94 for ETG and CRG
respectively) seem to reflect Schmitt’s (2008) conclusion drawn
from reviewing the related literature (e.g., de Groot, 2006; Nation,
2001) about the importance of recycling and repeated exposure in
vocabulary learning. Therefore, teachers should take repeated
 TELL, Vol. 5, No. 1, 2011 65
Rashtchi and Jalili

exposure into consideration while practicing incidental vocabulary

learning in their classrooms. This implies that better results are
achieved if instruction on knowledge sources is combined with
recycling previously learned words.
Using more knowledge sources by the successful group can be
attributed to two reasons; first, the successful group attempted to
infer 72 vocabularies while the less successful group managed to
infer 64 words. Second, the successful group in their attempts to
infer unknown vocabulary used multiple (more than one)
knowledge sources rather than a single one for a word more
frequently while the less successful group used single knowledge
sources more frequently. To be more precise, the successful group
used single knowledge sources and multiple knowledge sources 24
and 48 times respectively while the less successful group used
multiple and single knowledge sources 28 and 30 times
respectively. Attempting to infer more words and using more
multiple knowledge sources by the successful group are consistent
with Fraser’s (1999) and Kaivanpanah and Alavi’s (2008b) findings
respectively in which more proficient learners tried to infer more
words and used more multiple knowledge sources than limited
proficient learners respectively.
Table 6 echoes Paribakht’s (1985) finding that proficient and
less proficient learners used largely similar sequences of knowledge
sources in oral communication. A quick look at Table 5 shows that
the less successful group used sentence-level grammatical
knowledge and world knowledge more than the successful group. A
justification can be that because the less successful group had less
knowledge sources at their disposal (as is apparent from not using
word association and using less proportion of other knowledge
sources) they resorted more frequently to their background
knowledge as well as sentence-level grammatical knowledge for
which they received instruction. Considering that both groups
received sentence-level grammar instruction explicitly, it seems that
the less successful learners benefited more from the instruction;
thus, they seemed to compensate for the unavailability of other
knowledge sources by the knowledge source they received
instruction on. The pattern of used knowledge sources for all
66 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

participants as a group concurs with that of Paribakht (2004) except

for discourse, cognate, and punctuation. That is, whereas discourse
was a minor knowledge source in Paribakht’s (2004) study, it was a
major one in this study. The other differences were that participants
in this study did not use cognate and punctuation knowledge sources
at all. Since using cognate knowledge source was also missing in
Kaivanpanah and Alavi’s (2008b) study wherein their participants
were Iranian, it seems reasonable to conclude that the first language
of the participants in both studies (i.e. Persian) could be the source
of difference. The finding that participants in this study did not use
punctuation knowledge source may reflect the fact that Iranian
teachers are seriously negligent in providing EFL learners with
punctuation instruction; unquestionably, an exception could be
those who are majoring in English, with formal instruction on
punctuation. It should be recalled that participants in this study were
undergraduate students from various majors other than English. For
pedagogical purposes, teachers should focus on those knowledge
sources which learners did not use very much as a total group and
also those knowledge sources which learners in successful group
used more frequently than those in less successful one. Confusing a
word with an unrelated but similar-looking word (Table 7) provides
support for the importance that Schmitt (2008) attaches to the form
of words in vocabulary learning. However, the promising point was
that when learners saw these words in context, contextual cues in
some cases helped them to restructure their lexical knowledge
which happened more frequently in the successful group than the
less successful one. This confirms the role of context in assessing
inferences posited by Huckin and Bloch (1993) and Bensoussan and
Laufer (1984). More restructuring in successful group reflects
McLaughlin’s (1990) claim that more successful learners are more
flexible in restructuring than less successful ones.

6. Conclusion

The present study supports the findings of the previous studies that
sentence-level grammatical knowledge contributes to lexical
 TELL, Vol. 5, No. 1, 2011 67
Rashtchi and Jalili

inferencing success and retention of inferred vocabulary. It also

verifies that both explicit teaching and consciousness-raising are
effective. Additionally, the findings indicate that in teaching lexical
inferencing emphasis should be placed on those knowledge sources
which less successful learners fail to use whatsoever or use less than
successful learners. Further study can be carried out to discern the
results of teaching other knowledge sources, which learners use to
infer the meaning of unknown words, on lexical inferencing
behavior of learners.

References
Alavi, S. M., & Kaivanpanah, S. (2009). Examining the role of
individual differences in lexical inferencing. Journal of Applied
Sciences, 9(15), 1-6.
Bengeleil, N. F., & Paribakht, T.S. (2004). L2 reading proficiency
and lexical inferencing by university EFL learners. The
Canadian Modern Language Review, 61(2), 225-249.
Bensoussan, M., & Laufer, B. (1984). Lexical guessing in context
in EFL reading comprehension. Journal of Research in
Reading, 7(1), 15-32.
Bonk, W. (2000). Second language lexical knowledge and listening
comprehension. International Journal of Listening, 14, 14-31.
Clarke, D. F., & Nation, I. S. P. (1980). Guessing the meaning of
unknown words from context: Strategy and techniques. System,
8(3), 211-220.
Craik, F. I., & Lockhart, R. S. (1972). Levels of processing: A
framework for memory research. Journal of Verbal Learning &
Verbal Behavior, 11, 671-684.
Craik, F. & Tulving, E. (1975). Depth of processing and the
retention of words in episodic memory [Electronic version].
Journal of Experimental Psychology: General, 104, 268-294.
De Bot, K., Paribbakht, T. S., & Wesche, M. B. (1997). Toward a
lexical processing model for the study of second language
vocabulary acquisition: Evidence from ESL reading. Studies in
Second Language Acquisition, 19, 309-329.
68 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

Decarrico, J. S. (2001). Vocabulary learning and teaching. In M.

Celce-Murcia (Ed.), Teaching English as a second or foreign
language (pp. 285-299). Boston, MA: Heinle & Heinle.
De Groot, A. M. B. (2006). Effects of stimulus characteristics and
background music on foreign language vocabulary learning and
forgetting. Language Learning, 56(3), 463-506.
Dunmore, D. (1989). Using contextual clues to infer word meaning:
An evaluation of current exercise types. Reading in a Foreign
Language, 6(1), 337-347.
Dycus, D. (1997). Guessing word meaning from context: Should we
encourage it? Retrieved October 25, 2009, from
http://www2.aasa.ac.jp/dcdycus/ LAC97/ guessing.htm
Eastwood, J. (2006). Oxford practice grammar – intermediate.
Oxford: Oxford University Press.
Ellis, R. (1993). Second language acquisition research: How does it
help teachers? An interview with Rod Ellis. ELT Journal, 47(1),
3-11.
Ellis, R. (2002). Grammar teaching – practice or consciousness-
raising? In J. C. Richards & W. A. Renandya (Eds.),
Methodology in language teaching: An anthology of current
practice (pp. 167–174). Cambridge: Cambridge University
Press.
Fotos, S. (1994). Integrating grammar instruction and
communicative language use through grammar consciousness-
raising tasks. TESOL Quarterly, 28(2), 323-351.
Fotos, S., & Ellis, R. (1991). Communicating about grammar: A
task-based approach. TESOL Quarterly, 25(4), 605–628.
Frantzen, D. (2003). Factors affecting how second language
Spanish students derive meaning from context. Modern
Language Journal, 87(2), 168-199.
Fraser, C. (1999). Lexical processing strategy use and vocabulary
learning through reading. Studies in Second Language
Acquisition, 21(2), 225-241.
Grabe, W., & Stoller, F. L. (2001). Reading for academic purposes:
Guidelines for the ESL/EFL teacher. In M. Celce-Murcia (Ed.),
Teaching English as a second or foreign language (pp. 187-
203). Boston, MA: Heinle & Heinle.
 TELL, Vol. 5, No. 1, 2011 69
Rashtchi and Jalili

Haastrup, K. (1991). Lexical inferencing procedures or talking

about words. Tübingen: Gunter Narr.
Hedge, T. (2000). Teaching and learning in language classroom.
Oxford: Oxford University Press.
Hu, M., & Nassaji, H. (in press). Ease of inferencing, learner
inferential strategies, and their relationship with retention of
word meanings inferred from context. Canadian Modern
Language Review.
Huckin, T., & Bloch, J. (1993). Strategies for inferring word
meaning in context: A cognitive model. In T. Huckin, M.
Haynes & J. Coady (Eds.), Second language reading and
vocabulary learning (pp. 153-176). Norwood, NJ: Ablex.
Huckin, T., & Jin, Z. (1987). Inferring word meanings from
context: A study in second language acquisition. In F. Marshall
(Ed.), ESCOL ’86. Columbus, OH: Ohio State University
Department of Linguistics.
Hulstijn, J. H. (1992). Retention of inferred and given word
meanings: Experiments in incidental vocabulary learning. In P.
J. L. Arnaud & H. Be’joint (Eds.), Vocabulary and applied
linguistics (pp. 113–125). London: Macmillan.
Hulstijn, J. H., & Laufer, B. (2001). Some empirical evidence for
the involvement load hypothesis in vocabulary acquisition.
Language Learning, 51(3), 539-558.
Hunt, A., & Beglar, D. (1998). Current research and practice in
teaching vocabulary. In J. C. Richards & W. A. Renandya
(Eds.), Methodology in language teaching: An anthology of
current practice (pp. 258-266). Cambridge: Cambridge
University Press.
Kaivanpanah, S., & Alavi, M. (2008a). Deriving unknown word
meaning from context: Is it reliable? RELC Journal, 39(1), 77-
95.
Kaivanpanah, S., & Alavi, M. (2008b). The role of linguistic
knowledge in word-meaning inferencing. System, 36(2), 172-
195.
Kern, R. G. (1989). Second language reading strategy instruction:
Its effects on comprehension and word inference ability. The
Modern Language Journal, 73(2), 135-149.
70 TELL, Vol. 5, No. 1, 2011
Consciousness-raising versus explicit grammar teaching

Krashen, S, D. (2003). Explorations in language acquisition and

use. Portsmouth, NH: Heinemann.
Lee, L., & Gundersen, E. (2002). Select readings – pre-
intermediate: Student book. New York, NY: Oxford University
Press.
Lesly, T., Hansen, C., & Zukowsky/Faust, J. (2005). Interchange
(3rd ed.)/placement and evaluation package. Cambridge:
Cambridge University Press.
Liu, N., & Nation, I. S. P. (1985). Factors affecting guessing
vocabulary in context. RELC Journal, 16(1), 33-42.
McLaughlin, B. (1990). Restructuring. Applied Linguistics, 11(2),
113-128.
Nassaji, H. (2003). L2 vocabulary learning from context: Strategies,
knowledge sources, and their relationship with success in L2
lexical inferencing. TESOL Quarterly, 37(4), 645-670.
Nassaji, H. (2004). The relationship between depth of vocabulary
knowledge and L2 learners' lexical inferencing strategy use and
success. The Canadian Modern Language Review, 61(1), 107-
134.
Nation, P. (1996). Best practice in vocabulary teaching and
learning. In J. C. Richards & W. A. Renandya (Eds.),
Methodology in language teaching: An anthology of current
practice (pp. 267-272). Cambridge: Cambridge University
Press.
Nation, P. (2001). Learning vocabulary in another language.
Cambridge: Cambridge University Press.
Paribakht, T. S. (1985). Strategic competence and language
proficiency. Applied linguistics, 6(2), 132-146.
Paribakht, T. S. (2004). The role of grammar in second language
lexical processing. RELC Journal, 35(2), 149-160.
Paribakht, T. S. (2005). The Influence of L1 Lexicalization on L2
Lexical Inferencing: A Study of Farsi-Speaking EFL Learners.
Language Learning, 55(4), 701-748.
Paribakht, T. S., & Wesche, M. (1996). Enhancing vocabulary
acquisition through reading: a hierarchy of text-related exercise
types. Canadian Modern Language Review, 52, 250-273.
Paribakht, T. S., & Wesche, M. (1997). Vocabulary enhancement
 TELL, Vol. 5, No. 1, 2011 71
Rashtchi and Jalili

activities and reading for meaning in second language

vocabulary acquisition. In J. Coady & T. Huckin (Eds.), Second
language vocabulary acquisition: A rationale for pedagogy
(pp.174-200). Cambridge: Cambridge University Press.
Paribakht, T. S., & Wesche, M. (1999). Reading and incidental L2
vocabulary acquisition: An introspective study of lexical
inferencing. Studies in Second Language Acquisition, 21(2),
195-224.
Preliminary English Test, Handbook for Teachers (2009)
[Electronic version]. Cambridge: University of Cambridge,
ESOL examinations.
Pulido, D. (2007). The effects of topic familiarity and passage sight
vocabulary on L2 lexical inferencing and retention through
reading. Applied Linguistics, 28(1), 66-86.
Richards, J. C., & Renandya, W. A. (2002). Methodology in
language teaching: An anthology of current Practice.
Cambridge: Cambridge University Press.
Schmidt, R.W. (1990). The role of consciousness in second
language learning. Applied Linguistics, 11(2), 129-158.
Schmitt, N. (2000). Vocabulary in language teaching. Cambridge:
Cambridge University Press.
Schmitt, N. (2008). Review article: Instructed second language
vocabulary learning. Language Teaching Research, 12(3), 329-
363.
Schmitt, N., & Zimmerman, C. B. (2002). Derivative word forms:
What do learners know? TESOL Quarterly, 36(2), 145-171.
Tschannen-Moran, M., & Hoy, A. W. (2007). The differential
antecedents of self-efficacy beliefs of novice and experienced
teachers. Teaching and Teacher Education, 23(6), 944-956.
Walters, J. M. (2004). Teaching the use of context to infer meaning:
A longitudinal survey of L1 and L2 vocabulary research.
Language Teaching, 37(4), 243-252.
Wesche, M., & Paribakht, T.S. (1996). Assessing second language
vocabulary knowledge: Depth versus breadth. The Canadian
Modern Language Review, 53(1), 13-40.