International Journal of Lexicography, Vol. 25 No. 2, pp.

216–247
doi:10.1093/ijl/ecr031 Advance access publication 2 December 2011 216

DICTIONARY USE AND

VOCABULARY LEARNING IN THE
CONTEXT OF READING

Yuzhen Chen: College of Foreign Languages and Cultures of Xiamen University, Fujian,
China and Department of Languages of Putian University, Fujian, China
([email protected])

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Abstract

This empirical study attempts to explore the role of dictionary use in L2 vocabulary
learning in reading context. It involved the use of English-Chinese bilingualized diction-
aries (BLDs) for EFL vocabulary task completion and incidental vocabulary acquisition
by undergraduate English majors in Chinese universities. The subjects were asked to
read an English passage and perform a reading task under one of three conditions: with
the aid of a paper BLD (PBLD) or an electronic BLD (EBLD), or without access to any
dictionary. After task completion, they were given an unexpected retention test on the
target lexical items included in the reading passage. The same retention test was re-
peated one week later. The study found that BLD use can effectively facilitate vocabu-
lary comprehension and enhance incidental vocabulary acquisition, suggesting that
dictionary use is a more effective strategy of vocabulary learning than contextual gues-
sing. There was no significant difference in dictionary effectiveness between the PBLD
and the EBLD, yet the latter showed some advantage over the former for vocabulary
retention. Students varying on vocabulary proficiency levels and reading conditions
fared differently on incidental vocabulary acquisition.

1. Introduction

Dictionary use has long been recognized as one of vocabulary learning strate-
gies (Gu and Johnson 1996, Scholfield 1997, Nation 1990, 2001, Gu 2003,
Nation and Meara 2010). Yet despite the important role of the dictionary
for L2 learning and the relatively long history of the research on vocabulary
learning through dictionary use, in the domain of L2 vocabulary acquisition,
‘interest from a research perspective has been limited and sporadic over the
years’ (Ronald 2003: 285).1 Fortunately, recent years have witnessed steady
development of dictionary use research which includes investigations of the

# 2011 Oxford University Press. All rights reserved. For permissions,

please email: [email protected]
 Dictionary Use and Vocabulary Learning in the Context of Reading 217

use and usefulness of dictionaries for various language activities. This study
attempts to evaluate the effectiveness of dictionary use for L2 vocabulary
learning in reading context. It examines the use of English-Chinese bilingual-
ized dictionaries (henceforth BLDs) for EFL vocabulary task completion and
incidental vocabulary acquisition during reading. This type of dictionary is
hugely popular with Chinese EFL learners, yet has received little attention
from researchers of dictionary use studies. By evaluating the effectiveness of
such dictionaries for EFL vocabulary learning and identifying the problems
with dictionary use, this research attempts to shed some light on vocabulary
pedagogy and dictionary use instruction in a Chinese EFL environment.

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

2. Literature review

2.1 Dictionary use and vocabulary comprehension

Dictionaries are supposed to be useful aids to reading comprehension.

However, studies comparing dictionary use and non-dictionary use during
reading comprehension have yielded different or even contradictory findings.
Some researchers (e.g. Bensoussan, Sim and Weiss 1984, Nesi and Meara 1991,
Neubach and Cohen 1988 cited in Nesi 2000: 37–39, Heijnen 2000 cited in
Welker 2010: 177–178) identified a non-significant relation between dictionary
accommodation and test scores of reading comprehension. Such results were
ascribed to the following reasons:

(a) the test itself was made up of items which were not likely to be affected
by the availability of a dictionary,
(b) the dictionary did not include information needed to answer the compre-
hension questions, and
(c) the user failed to identify the words in the text which were most
crucial for correct answering of the test questions (Nesi and Meara
1991: 643).

In addition, dictionary users’ limited lexical knowledge, their lack of ability to

infer and their lack of experience in dictionary use also accounted for the
failure of dictionary access to have positive effect on test scores (Heijnen
2000, cited in Welker 2010: 177–178). Albus et al (2001) reported that a sig-
nificant benefit of dictionary use was found only for the intermediate-level
students, but not for lower or higher level proficiency students. In Chang’s
study (2002), the use of glosses or dictionaries did not bring about substantially
better results of reading comprehension than non-dictionary use, yet the case
with vocabulary retention was different. Padron and Waxman (1988, cited in
Szczepaniak 2006: 6) indicated that among fourteen strategies related to read-
ing achievement, looking up words in the dictionary turned out to be a negative
218 Yuzhen Chen

one. Aizawa (1999) even found that subjects in the non-dictionary condition
achieved significantly better results of reading comprehension than those using
dictionaries.
In contrast, some studies demonstrated a positive correlation between
dictionary use and vocabulary comprehension. Summers (1988) revealed that
compared with no-entry use, the use of dictionary entries yielded substantially
better results of comprehension as well as production. Tono (1989), later
republished as Tono (2001: 75–83), showed that a significant difference in
performance existed between reading comprehension with dictionaries and
that without dictionaries. Similar findings were obtained by Bogaards (2002,
cited in Welker 2010: 178–179) and Hayati and Pour-Mohammadi (2005). In
Szczepaniak (2006), the monolingual dictionary was found to be effective

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

for the successful completion of a paraphrase task as a source of the hardly
inferable canonical meaning of idioms. Nevertheless, despite a 50% rise in
scores after dictionary consultation, in only two cases (of the four idioms)
was the difference between the dictionary and non-dictionary condition
statistically significant (Szczepaniak 2006: 84). In other words, the study
confirmed a mildly positive influence of the dictionary on the comprehension
of contextually modified idioms.
Electronic dictionaries have proved to be very useful tools for researchers to
gain more accurate and non-intrusive information about dictionary use. One of
the most frequently cited study is Knight (1994) which will also be mentioned
in the next section due to its major focus on incidental vocabulary acquisition.
The results of Knight’s study concerning reading comprehension indicated that
the dictionary group obtained significantly higher scores than the
non-dictionary group. However, this overall result was due to the low verbal
ability group only (Knight 1994: 293). In contrast, Koga (1995, cited in
Kobayashi 2006: 59) showed that, for the low-reading-ability group, there
was no significant difference in reading comprehension scores across three
dictionary groups while the case for the higher-reading-ability group was dif-
ferent. Koga attributed the advantage of the on-line dictionary to the fact that
it had less interference in the reading process and thus facilitated students’
reading comprehension. The benefit of on-line dictionaries was substantiated
by Zucchi (2010, cited in Welker 2010: 310–311) who found that the scores of
the dictionary groups were significantly higher than those of the non-dictionary
group.
The findings of dictionary-supported comprehension studies reviewed above
seem to point to one fact: there is no simple yes-or-no answer to the question
whether the dictionary is useful for vocabulary comprehension or not. A var-
iety of factors should be taken into consideration during dictionary assessment,
one of which is the proficiency level of learners. In addition to the mere pres-
ence of a dictionary, the accessibility of information and the presentation style
of dictionary entries may also play a role. Furthermore, even if similar
 Dictionary Use and Vocabulary Learning in the Context of Reading 219

evaluation instruments such as multiple-choice comprehension questions are

used, it would be inappropriate to assume equivalence between different tests,
as items can depend to a greater or lesser degree on overall text comprehension
or the comprehension of individual vocabulary items, and such balance is dif-
ficult to control (Lew 2004: 28).

2.2 Dictionary use and incidental vocabulary acquisition

From its origin in stimulus-response psychological studies to the popular

strand of research on L2 vocabulary acquisition, incidental learning has been
given various interpretations and investigated under different research para-
digms, usually as a construct opposed to intentional learning (Hulstijn 2001,

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

2003, 2005, in press, Laufer and Hulstijn 2001, Laufer and Hill 2000, Schmidt
1994, 2001). In this study, incidental vocabulary learning refers to the learning
mode in which participants are not informed, prior to their engagement in a
learning task, that they will be tested afterward on their retention of lexical
information (Hulstijn 2005). In other words, the author chooses to use inciden-
tal learning as a technical term instead of an academic construct, for, theoret-
ically, the distinction between intentional and incidental learning has become
difficult to maintain (Hulstijn 2001, 2003).
Compared with the studies on the role of dictionary use in reading compre-
hension, researchers seem to show more interest in the association between
dictionary use and incidental vocabulary learning. Most studies have identified
a positive impact of dictionary use upon vocabulary retention. Krantz (1991,
cited in Welker 2010) showed that students retained 15.5% of the target words
after reading an English book and that the more proficient readers tended to
learn more words than the less proficient ones. As reported by Luppescu and
Day (1993), students using dictionaries fared significantly better on the vocabu-
lary retention test than those who did not. The finding was corroborated by a
more carefully designed study by Knight (1994) who also showed that high
verbal ability students learned more words than those with low verbal ability.
Cho and Krashen (1994, cited in Laufer 2003) suggested that reading plus
dictionary use resulted in better vocabulary retention than reading only. In
his assessment of the impact of lexical process strategies on vocabulary learn-
ing, Fraser (1999) found that the strategy of dictionary consultation led to
better comprehension than inferring and the combined use of dictionary
consultation and inferring could significantly improve vocabulary retention.
Some studies presented a more complicated picture of the effect of dictionary
use on vocabulary learning. In a well-conceived study by Hulstijn et al. (1996),
the group reading with marginal glosses had the highest scores of vocabulary
tests while the bilingual dictionary group, due to a scarce use of the dictionary,
did not perform significantly better than the control group. However, in the
few cases in which the students did use the dictionary, their retention scores
220 Yuzhen Chen

were even higher than those of the students with marginal glosses. Aizawa
(1999) showed a different role of dictionary use for vocabulary comprehension
and vocabulary retention: for the former, the non-dictionary group scored
significantly higher than the dictionary group while there was a reverse result
for the latter. Nevertheless, for those more proficient learners, there was almost
no difference in vocabulary retention. In the same vein, Chang (2002) found
that reading with different conditions did not produce significant effects on
reading comprehension but for vocabulary retention, the use of marginal
glosses and electronic dictionaries yielded different results.
Different from the above-mentioned studies, Conceição (2004, cited in
Welker 2010) concluded that dictionary use does not contribute significantly
to vocabulary retention as there was no significant difference in retention

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

scores between subjects who did consult the dictionary during reading and
those who did not. It should, however, be pointed out that the study suffers
from a methodological flaw. The author used the VKS (Vocabulary
Knowledge Scale) developed by Paribakht and Wesche (1993) as a measure
of vocabulary retention, but such a test form is appropriate for tracking the
states of knowledge for particular words, not for summing up different scores
of different words. Furthermore, the small sample size (35 vs. 16 subjects) also
makes it premature to generalize the claim. In Laufer (2011), the retention
scores of the looked up collocations measured one week later turned out to
be very low, which was believed to result from learners’ failure to locate the
target collocations in the dictionary entries or their overconfidence which pre-
vented them from seeking dictionary help.
In Chinese EFL learning context, only a few studies have probed into the
relationship between dictionary use and incidental vocabulary acquisition. By
considering factors such as students’ learning strategies, learning motivation,
vocabulary proficiency, gender and family background, Wu et al (2007) un-
veiled a rather complicated picture about the effects between the use of a
monolingual dictionary and the use of a word list on vocabulary retention.
Zhang (2007) showed that in terms of vocabulary gains, the provision of mar-
ginal glosses was the more beneficial than the availability of dictionary and
non-dictionary use. However, when readers did consult the dictionary, the
result was even better than the use of marginal glosses. In terms of vocabulary
retention, these three groups showed no significant difference, yet words
inferred or looked up can be remembered slightly better than those provided
directly with meanings. Shi (2008) revealed that consulting lexical information
in the dictionary tended to be associated with better retention than contextual
guessing, yet a combined use of learning methods was more conducive to in-
cidental vocabulary learning than separate use of them. This finding was sup-
ported by Ji (2009).
Despite the differences in research designs, the above-reviewed studies prove
that incidental vocabulary acquisition does occur and most of them point to
 Dictionary Use and Vocabulary Learning in the Context of Reading 221

the advantage of dictionary use for vocabulary retention. Yet with the only
exception of Laufer (2011), all studies mentioned above did not involve any
BLD use. This is one of the reasons that initiated the present study.

2.3 Paper dictionaries vs. electronic dictionaries

There is now a large body of literature on the use of various kinds of electronic
dictionaries covering such topics as the usefulness of dictionaries for learning
tasks, the comparison of dictionary effectiveness between different types, and
lookup preferences and behavior of dictionary users in CALL context etc. Yet
on the whole, the contrastive studies between electronic dictionaries and paper

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

dictionaries with a focus on vocabulary learning are relatively few.
Some researchers (e.g. Koyama and Takeuchi 2003, Iso and Osaki 2004,
cited in Kobayashi 2006: 63, Kobayashi 2007, Chen 2010) found that there
were no significantly different effects on vocabulary learning between pocket
electronic dictionaries and paper dictionaries. Laufer (2000) compared the ef-
fects of paper glosses and electronic glosses and found that on both retention
tests, the electronic glosses yielded significantly higher scores than the paper
ones, which may be ascribed to the visual impact produced by a word
embedded in a pop-up window and appearing in a prominent position on
the computer screen, or the more involvement load induced by the consultation
of electronic glosses which led to better retention. In Koga (1995), the on-line
dictionary was found to be more effective than the paper dictionary for the
higher-reading-ability group, yet for the low-reading-ability group, there was
no significant difference.
Dziemianko (2010, 2011, in press) has presented some interesting findings
about the comparison between on-line e-dictionaries and their book versions.
In Dziemianko (2010), e-COBUILD6 was found to be much more effective for
the receptive task, the productive task and the retention of both meaning and
collocations than COBUILD6 in book form. The author (Dziemianko 2010:
265) assumed that the form of presentation on the computer screen, possibly
more captivating and less distracting than the view of headwords on a page in a
paper dictionary, may account for the superiority of electronic dictionaries
over paper ones. However, such findings were not confirmed by the author’s
own replication studies (Dziemianko 2011, in press) which adopted exactly the
same experimental conditions except for the dictionaries used, i.e. the on-line
and paper versions of LDOCE5 and OALDCE7. It was found that, the
e-versions of LDOCE5 and OALDCE7 were no better for language reception,
production and learning than the dictionaries in book form (Dziemianko in
press). In other words, the medium had no statistically significant bearing on
reception or production and it did not affect the retention of meaning and
222 Yuzhen Chen

collocation either (Dziemianko 2011). As explained by the author, compared

with the clearer and more neatly organized website of e-COBUILD6,
e-LDOCE5 is characterized by an excess of unsolicited information such as
colorful widgets, banners, animation, and noise etc. which distract users’
attention and make dictionary information much less salient on the glutted
website. The studies by Dziemianko seem to suggest that dictionary form
may not be the decisive factor accounting for dictionary effectiveness; instead,
form-independent factors such as layout, font and line spacing may play an
even more important role.
As far as the present author knows, there is scarcely any research comparing
the use of dictionaries on computer desktop and paper form, neither is there
any study involving BLDs of different forms except for the one the author did,

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

i.e. Chen (2010) in which BLDs on pocket electronic dictionaries and paper
form were compared in the context of sentence use. Therefore, in the present
study, the author chose a desktop BLD and its print version and compared
their effects on vocabulary learning in the context of reading.

3. The study

3.1 Research hypotheses

Based on the literature review given above, the author proposed the following
research hypotheses concerning the use of a paper BLD (henceforth PBLD)
and an electronic BLD (henceforth EBLD) for vocabulary learning (including
vocabulary comprehension and incidental vocabulary acquisition).

Hypothesis 1: Students using the PBLD or the EBLD achieve significantly

better results of vocabulary learning than those without access to the
dictionary.

Hypothesis 2: There is no significant difference between the PBLD and the

EBLD in terms of dictionary effectiveness for vocabulary learning.

Hypothesis 3: Students varying on vocabulary proficiency levels and

reading conditions fare differently on incidental vocabulary acquisition.
Here reading condition refers to reading with the assistance of the PBLD
or the EBLD, or with no access to any dictionary.

3.2 Design issues

Subjects of the study were asked to finish a reading task, which was followed by
an unexpected vocabulary retention test that was repeated one week later. The
following is a detailed introduction to design issues.
 Dictionary Use and Vocabulary Learning in the Context of Reading 223

3.2.1 Selection of the reading material and target lexical items. Several factors were
taken into consideration for the selection of reading material. The priority is
to ensure that the reading text is of an appropriate level of difficulty with a lexical
density that would allow general comprehension through contextual guessing.
Therefore, the author adopted a density of 98% known words as advocated by
Hu and Nation (2000) and Nation (2001). To arouse students’ interest in reading
under experimental conditions, narratives were preferred over other types of
writings such as argumentation, exposition or description. Furthermore, texts
which are too long or too short would not be considered as appropriate for
practical reasons of test administration. In view of these considerations, two
texts of similar levels of difficulty, length and type were chosen from a corpus
that is not accessible to the subjects. Three teachers of extensive English reading

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

programs were consulted about the suitability of the reading material and finally
one of the two texts was agreed upon as the instrument of the study.
The target lexical items were chosen according to four criteria. First, they are
agreed upon by the above-mentioned three teachers as unfamiliar to most of
the participants based on the teachers’ teaching experience and their estimate of
the general vocabulary proficiency levels of the participants. Secondly, there is
more or less context which provides some chance to guess the meaning of the
lexical items so that the non-dictionary group would not be placed at a disad-
vantage. Thirdly, they are not proper nouns or technical terms which are of
relatively less experimental value. And fourthly, they are not too long or too
difficult in spelling which may otherwise induce a heavier memory burden on
the participants. It needs to be mentioned that the lexical items include single
words and a chain of words such as phrasal verbs, collocation and idioms
which are characterized by co-occurrence. Altogether, there are six single
words and four multi-word units, namely conjure, wand, rope in, wind up,
wobble, a dab hand, rapt, cabaret, augment, and play up to.

3.2.2 Design ofthe reading task. The reading text is accompanied by 13 compre-
hension questions. Different from the regular type of reading comprehension
questions which mostly involve global text comprehension, this study adopts a
word-focused approach: most of the questions are related to the comprehen-
sion of the target items. This approach can be justified by two considerations.
One is to avoid the pitfall of test design of some previous studies such as
Bensoussan et al. (1984) and Nesi and Meara (1991): the test may be made
up of items which are not likely to be affected by dictionary use. By incorpor-
ating target items into comprehension questions, a closer correlation between
word comprehension and dictionary use will be established. Meanwhile, some
of the comprehension questions are also connected with the overall text com-
prehension so as to prevent participants from doing skipping reading only.
To be specific, the questions concerning the comprehension of target lexical
224 Yuzhen Chen

items account for nearly 70% of the total, i.e. among 13 questions, nine are
word-focused, six of which are multiple choice questions and three in
question-and-answer form. Given the nature of the study, the scores of these
nine questions were analyzed, rather than the total score of all 13 questions.
The other reason to adopt a word-focused approach is to ensure a higher
incidence of incidental vocabulary acquisition. As reported by previous re-
searchers (Hulstijn 1992, Cho and Krashen 1994, Parikbakht and Wesche
1997, Zahar et al. 2001, Horst 2005), vocabulary gains through reading without
any enhancement tasks tend to be extremely small, ranging from one to seven
words per text (which is up to 7,000 words). Empirical evidence supports the
claim that in instructed L2 context, word-focused activities play a more im-
portant role than reading alone in building the learner’s lexical knowledge

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

(Paribakht and Wesche 1997, Laufer 2001, 2003, Hill and Laufer 2003,
Peters 2007). The author’s previous experience with a failed study also
showed that reading without a word-focused task could barely lead to any
incidental vocabulary learning. In that study, the author divided subjects in
two groups, namely a dictionary group and a non-dictionary group, who were
told to read a text which included some target words and then do a compre-
hension task. The target words were not marked or distinguished in any way
and the questions were mostly based on the overall text comprehension. It
turned out that the incidental vocabulary gains of both groups were so small
that it made no sense to make any comparison. Since that reading task did not
directly involve the comprehension of individual target words which, to make
matters worse, were not marked for salience in the text, most subjects simply
did not pay any attention to them, let alone do any contextual guessing or
dictionary consultation. That experience, together with previous researchers’
findings, leads the author to believe that the inclusion of target items in test
questions can effectively enhance vocabulary retention.
With regard to the issue of test format, in addition to the multiple choice test
which is preferred by a majority of previous studies, a generative test form is
also adopted, i.e. the subjects were required to give their own interpretation for
the questions asked instead of choosing passively from given options. It is
believed that the combination of two test forms can reveal a fuller picture of
subjects’ comprehension of target items.

3.2.3 Design of the vocabulary retention tests. Like most incidental learning re-
search, vocabulary retention in this study is measured by checking whether
the subjects can recall the meaning of the target items.2 The study included
two retention tests of the same contents, one was conducted immediately after
the reading task, the other one week later. Hulstijn (2003: 372) argues that
experiments comparing different methods of cognitive processing of new lexical
material need only immediate post-tests, for it would not be possible to
 Dictionary Use and Vocabulary Learning in the Context of Reading 225

differentiate the extent to which performance on delayed post-tests is affected

by processes during the experimental learning session or by processes after that
session. However, the author believes that with certain control devices, as will
be explained in Section 3.4, the delayed test could also yield some valuable
decay data to provide a clearer picture of the actual learning that has taken
place, as hoped by Waring and Nation (2004: 107).

3.2.4 Selection of the BLDs and solution to the problem of dictionary underuse. As the
study also involves a comparison of effectiveness between a PBLD and an
EBLD, to strictly control the variable of dictionary form, the author operatio-
nalized the electronic-paper opposition by using a desktop dictionary and its

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

printouts so that factors related to dictionary information, lexicographical
presentation, layout, font, and line spacing etc. are balanced off, allowing to
isolate the on-screen versus paper presentation. Many dictionary surveys report
that Oxford Advanced Learner’s English-Chinese Dictionary (henceforth
OALECD) is by far the most popular paper dictionary with Chinese EFL
learners, yet the electronic version of the current dictionary edition is very
hard to find, except in a few high-profiled pocket electronic dictionaries such
as CASIO, which, unfortunately, are owned by a relatively small number of
students. The only electronic version of OALECD available on the Internet is
its fourth edition rather than the sixth or seventh, therefore, the author had
to settle for OALECD4. Actually, e-OALECD4 was downloaded from http://
www.lingoes.cn/zh/dictionary/ and installed in Lingoes 2.7.1.0, a powerful
multi-lingual software program released as a freeware translation utility.
The author copied all the relevant on-screen dictionary information and
made a paper minidictionary which contains only the entries of ten target
items. In other words, the EBLD used for the study is OALECD4 installed
inside Lingoes and the PBLD is a minidictionary with exactly the same
information copied from the EBLD. It should be mentioned that, unlike the
EBLD, the PBLD does not use, due to financial reasons, colors to represent
different categories of information except bold type. Figures 1 and 2 are
example screenshots of rapt in different BLD versions.
However, the availability of a dictionary can not guarantee that subjects
would use it. The problem of dictionary underuse has been identified by a
number of studies (e.g. Krantz 1991, Hulstijn et al. 1996, Atkins and
Varantola 1998) which registered disappointingly low rates of dictionary use
during experimental treatment. This poses a fundamental methodological prob-
lem to dictionary use research, because the failure to consult dictionaries masks
any potential effects of dictionary consultation (Lew 2004: 59). The author’s
failure experience of the study mentioned in Section 3.2.2 also made her aware
that a more strict control for dictionary use in the treatment groups might be
needed. Therefore, several steps were taken in this study to minimize dictionary
226 Yuzhen Chen

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Figure 1: Screenshot of dictionary entry for rapt in the EBLD. This figure
appears in colour in the online version of the International Journal of
Lexicography.

Figure 2: Dictionary entry for rapt in the PBLD. This figure appears in
colour in the online version of the International Journal of Lexicography.

underuse effect. First, all the target lexical items in the reading text were marked
in bold type so as to obtain a higher degree of salience. Secondly, all of them were
included in the question items, in bold type again, and subjects had to know their
meanings when completing the task. Thirdly, the dictionary groups were encour-
aged to consult the target lexical items and told to underline in the reading text
any word they consulted during task completion. Those who failed to consult the
target items were excluded from the final data analysis. Fourthly, the study was
done in subjects’ regular class sections and under the supervision of their re-
spective teachers, which promoted a higher degree of willingness on the part of
students to cooperate and follow test instructions. All these measures proved
useful to handle the problem of dictionary underuse in experimental treatment.

3.2.5 SelectionoftheVocabulary LevelsTest. Generally, there are two areas of inter-

est in L2 vocabulary testing: estimating vocabulary size (or breadth of vocabu-
lary knowledge) and assessing quality of word knowledge (or depth of
knowledge) (Read 1997). It is argued that except for certain research purposes,
 Dictionary Use and Vocabulary Learning in the Context of Reading 227

there is little point in eliciting all that learners may know about a particular set
of words (Read 2007: 113). In fact, learners’ word knowledge naturally deepens
as vocabulary size increases, so that good size measures may be all that are
required (Vermeer 2001). Furthermore, there are so many more aspects of word
knowledge that could potentially be assessed and no consensus has emerged as
to which are the most significant ones (Read 2004). Therefore, the author
decided on a two-section Vocabulary Levels Test (henceforth VLT), adapted
from Nation (2001), Schmitt et al. (2001) and Laufer and Nation (1999). The
first section is taken from Schmitt et al.’s version of VLT (2001) in which
subjects are requested to match words with their synonyms or short definitions.
The following is an example.

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

1 business
2 clock _part of a house
3 horse _animal with four legs
4 pencil _something used for writing
5 shoe
6 wall

Considering the estimated general vocabulary levels of the participants, only

three levels were chosen, i.e. the 3000 word level, the 5000 word level and the
Academic Vocabulary. An informal random pilot test done by the present
author on junior students showed that a majority of them knew nearly all the
words in the 2000 word level and barely any word in the 10,000 word level, which
led to the exclusion of these two. For practical reasons of test administration,
only seven clusters were chosen out of the original ten clusters at each word level
and each cluster contains three tested items, leading to a total of 63 items.
The second section is abridged from Laufer and Nation’s A Levels Test of
Productive Vocabulary (1999) which requires subjects to complete the word
form based on sentence context. An example is given below.

He was riding a bicycle.

For the same reasons mentioned above, only three word levels, i.e. the 3000
word level, the 5000 word level and the University Word List level were
adopted and except for the University Word level which includes 13 items,
each level contains 12 sentences with 12 tested items, thus reaching a total of
37 items. Altogether, there are 100 tested items in the VLT.

3.3 Participants

Participants for this study included three intact teaching classes of English
seniors from Putian University and another four classes of juniors from
228 Yuzhen Chen

Xiamen University and Fujian Agriculture and Forestry University, two

classes from each. Initially, more than 210 students participated in the first
phase of the study, yet unfortunately, due to some reasons that will be ex-
plained in the next subsection, the number of qualified participants was
reduced to 176. These students come from similar linguistic backgrounds and
have learned English for at least eight years. Nearly all of them have passed the
Test for English Majors (TEM, Band 4).

3.4 Testing and scoring procedures

The study was conducted during two regular class sections on two consecutive
weeks under the supervision of the participants’ respective teachers. In the case

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

of Putian University, the author monitored the testing processes in her own
classes while at the other two universities, the teachers in charge were trained
beforehand as to how to administer the tests. In the first week, the study was
conducted in students’ language labs which are equipped with computers. All
classes were randomly divided into three groups, each assigned to one of three
reading conditions: to read with access to an EBLD, or a PBLD, or with no
access to any dictionary. As mentioned in Section 2.2.4, the EBLD was
OALECD4 installed in Lingoes, a multifunctional translation software pro-
gram which had been downloaded and installed on the computer desktop
prior to the study. Lingoes was also set up in such a way that only
OALECD4 was available. In other words, students had to use OALECD4 as
it was the only dictionary contained in Lingoes after resetup. And other irrele-
vant functions of Lingoes were also forbidden. A few minutes had been given
to familiarize the EBLD group with the dictionary before the study began.
Except for the EBLD group, all computers in the seats for the other two
groups had been shut off.
Before the study began, the participants were informed of the purpose of the
study, i.e. to compare the effects between different reading conditions, but they
were not forewarned of the upcoming vocabulary retention tests. Both the
reading task and the retention tests were done in paper and pencil. To minimize
the time-upon-task effect, a 25-minute time constraint was imposed on all
participants to finish the reading task. As introduced in Sections 3.2.1 and
3.2.2, the reading text contains ten target items marked in bold type and the
reading task includes questions related to the comprehension of individual
target lexical items. During the reading task, from time to time, the dictionary
groups were required to use the designated dictionaries and underline in the
text every word they looked up while the non-dictionary group (henceforth the
ND group) were encouraged to do contextual guessing. After all students had
handed in the reading task material, with all computers shut down and mini-
dictionaries returned, an unexpected immediate retention test (henceforth RT1)
was administered in which students were asked to recall the meaning of the
 Dictionary Use and Vocabulary Learning in the Context of Reading 229

target items in either L1 or L2. The RT1 was to be finished within six to seven
minutes and without the assistance of any dictionary. At the end of the RT1
paper, students were also asked to identify among the target lexical items the
word(s), if any, that they had known prior to the study. No mention was made
of the other retention test that would follow in the next week, but students were
told in a delicate way not to do anything more with the target lexical items
after the RT1. In addition, they were instructed not to tell their fellow students
in other classes anything about the study so as to prevent collaboration. This
first phase of study, including the reading task and the RT1, took about
45 minutes.
Seven days later, the participants were given a delayed retention test
(henceforth RT2) in the same class and under the supervision of the same

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

teachers. The RT2 was exactly the same as the RT1 except for a rearranged
order of the target lexical items as an effort to avoid the carryover effect from
the previous test. After the RT2, students were told to do the VLT to assess
their vocabulary levels. They had no access to any dictionary and there was
no time limit, but most students finished it within 45 minutes. This second
phase, including the RT2 and the VLT, took about 50 minutes. The rea-
son why the VLT was done in the second week is out of practical consider-
ations of test administration. After all, the VLT does not include any target
item in the reading task, so it is unlikely to be affected by the experimental
treatment.
Altogether, four kinds of data were collected from the study: the score of the
reading task, the RT1 score, the RT2 score, and the VLT score. Nearly 40
students were excluded from final data processing. Some of them took part in
only one of the study phases or failed to produce all the data required. For
example, several students turned in only three kinds of data instead of four,
perhaps for fear that their poor performance would make an unfavorable im-
pression on their teacher. Despite the instruction to use the dictionary, some
students in the dictionary groups still did not use it or used it only for some
target lexical items, as no words or not all target items were underlined in the
reading text. This happened more frequently in the PBLD group than in the
EBLD group. In addition, some students had known two or more target items
before the study. It should be mentioned that although a few students identified
one or more target lexical items as previously known, a crosscheck with their
answers in the RT1 showed that they either wrote the wrong meaning or
did not provide any answer at all. In such cases, they were not eliminated
from the study as evidence indicated that they had not known the identified
items prior to or even after the reading task. Furthermore, some students
scored higher in the RT2 than in the RT1 which means they either further
processed the target items after the RT1 or used the dictionary sneakily dur-
ing the RT2 despite the teacher’s instruction. Some of them may be highly
motivated students, eager to learn vocabulary at every opportunity; some
230 Yuzhen Chen

may have known from other students that there would be another retention
test. The exclusion of all these students from the study left the final number
at 176.
The author alone undertook all the scoring work. For the reading task, the
maximum score is 18 for the nine word-focused questions. Each correct answer
for the multiple choice question yielded two points. The scoring of question
answering was based on the semantic and pragmatic criteria. If each question
was answered correctly and appropriately, it would get two points. If the
answer fit only one of the criteria, then one point. For example, when asked
to compose a sentence with a dab hand, many students wrote sentences like I’m
a dab hand in cooking, she is a dab hand with her papers or he is a dab hand in
farming, instead of using the idiom with a more frequent preposition at, which

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

is shown in the examples in the dictionary entry. In such a case, only one point
was given. The two retention tests were scored in the same way: one point for
correct responses while half point for partially correct ones. The total score is
10. In the case of polysemous target lexical items, if students wrote other cor-
rect meanings instead of the one used in the reading text, they could also get
points. For example, several students wrote , (make sb. angry)
for wind up though in the reading text it meant ... (end up). There were
chances that when students used the dictionary, they would incidentally acquire
more meanings of a word than the one involved in reading and this did happen
in this study, though at a low frequency. The total score for the VLT is 100 with
one point for each tested word. The second section of the VLT involved the
spelling of word forms. Any minor spelling errors, say, sooth for soothe, or
adequaet for adequate were ignored, as was the case with the use of verb tense.
For example, if students used scared instead of scare, they could also get a
point.
After the scoring procedures, all data were further processed by SPSS 16.0.
Results and findings are discussed at length in the next section.

3.5 Results

3.5.1 Vocabulary comprehension. As mentioned above, the reading task was

viewed as a measure of vocabulary comprehension. A one-way ANOVA
(Analysis of Variance, with post hoc tests) was performed to compare the re-
sults of different reading conditions. Table 1 shows that the PBLD group ob-
tained the best scores, followed by the EBLD group and then the ND group.
There is a strongly significant mean difference between the three groups [F(2,
173) = 36.04, p < 0.001] (see Table 2). Post-hoc tests (Games-Howell) revealed
that the mean score of the ND group differed significantly from those of the
EBLD group and the PBLD group (p < 0.001) while the latter two were not
significantly different from each other (p = 0.088).
 Dictionary Use and Vocabulary Learning in the Context of Reading 231

Table 1: Descriptive statistics of vocabulary comprehension scores

(Max = 18.00)

Group N Mean Std. Std. Error Minimum Maximum

Deviation

EBLD 46 13.11 2.07 0.30 8.00 18.00

PBLD 43 14.09 2.26 0.34 6.00 18.00
ND 87 10.34 2.96 0.32 2.00 17.00
Total 176 11.9 3.06 0.23 2.00 18.00

Table 2: ANOVA between vocabulary comprehension scores

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Sum of Squares df Mean Square F Sig.

Between Groups 483.21 2 241.61 36.04 0.000

Within Groups 1159.74 173 6.70
Total 1642.95 175

Table 3: Descriptive statistics of VLT scores (Max = 100.00)

N Mean Std. Std. Error Minimum Maximum

Deviation

EBLD 46 74.30 7.19 1.06 62.00 92.00

PBLD 43 77.16 8.73 1.33 51.00 94.00
ND 87 76.89 10.72 1.15 20.00 95.00
Total 176 76.28 9.46 0.71 20.00 95.00

A one-way ANOVA was also conducted to compare participants’ VLT

scores which verified the non-significant difference between these three
groups in terms of vocabulary proficiency [F(2, 173) = 1.38, p = 0.254; see
Table 3). This means the significant difference in students’ comprehension
scores did not result from the different vocabulary proficiency levels between
these groups. Very probably, it was the experimental treatment that led to the
significant difference in group performance. Apparently, the use of dictionaries
facilitated vocabulary comprehension much more than contextual guessing.

3.5.2 Vocabulary retention. As shown in Table 4, on both retention tests, the

EBLD group obtained the highest scores and the ND group the lowest.
Retention scores were submitted to 3  2 repeated measures ANOVA with
232 Yuzhen Chen

Table 4: Descriptive statistics of retention scores (Max = 10.00)

Group Mean Std. Deviation N

RT1 EBLD 6.95 1.85 46

PBLD 6.22 1.90 43
ND 4.20 1.95 87
Total 5.41 2.26 176
RT2 EBLD 4.77 1.79 46
PBLD 4.02 1.92 43
ND 2.74 1.94 87
Total 3.59 2.08 176

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Table 5: Test of within-subject effects (retention scores)

Source Type III Sum df Mean F Sig. Partial

of Squares Square Eta Squared

RT 300.39 1 300.39 237.03 0.000 0.578

RT * Group 11.77 2 5.89 4.64 0.011 0.051
Error (RT) 219.25 173 1.27

Table 6: Test of between-subjects effects (retention scores)

Source Type III Sum df Mean Square F Sig. Partial Eta

of Squares Squared

Intercept 3698.85 1 3698.85 1.235E3 0.000 0.877

Group 193.12 2 96.56 32.23 0.000 0.271
Error 518.26 173 2.99

reading Groups (the EBLD, the PBLD and the ND groups) as the
between-subjects factor and Time (the RT1 and the RT2) the within-subjects
factor. The measure of effect size is Z2, expressing explained variance. Results
in Tables 5 and 6 indicate a significant main effect for both Time
[F(1, 173) = 237.03; p < 0.001; Z2 = 0.578] and Groups [F(2, 173) = 32.23;
p < 0.001; Z2 = 0.271] and a significant time  groups interaction as well
[F(2, 173) = 4.64; p = 0.011; Z2 = 0.051]. As revealed by multiple comparisons
(Scheffe), a significant difference occurred between the ND and the other two
groups (p < 0.001) while the latter two did not differ substantially from each
other (p = 0.137).
 Dictionary Use and Vocabulary Learning in the Context of Reading 233

Table 7: Two-way ANOVA for RT1 scores as a function of reading condi-

tions and vocabulary levels

Variable and source df Mean F Sig Partial Eta

square Squared

RT1
Vocabulary level 1 42.99 14.35 0.000 0.140
Reading condition 2 40.52 13.52 0.000 0.235
Vocabulary level *reading condition 2 11.27 3.76 0.027 0.079
Error 88 2.99

R Squared = 0.403 (Adjusted R Squared = 0.369).

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

3.5.3 Vocabularylevels and reading conditions. Based on their VLT scores, the stu-
dents were divided into three vocabulary levels: higher, medium and lower. The
mean score of the whole sample is 76.2. Forty-seven students who ranked at the
top with scores ranging from 83 to 95 belong to the higher level group, taking
up 26.7% of the entire sample. The lower level group included another 47
students whose scores were at the bottom, between 20 and 70, also accounting
for 26.7% of the whole sample. Any comparison done in this study only
involved the higher and the lower level students. An Independent Samples
T-test confirmed that there was a highly significant difference between these
two groups in their VLT scores (p < 0.001).
Two two-way ANOVAs were performed with two levels of vocabulary pro-
ficiency (higher and lower) and three levels of reading conditions (EBLD,
PBLD and ND) as the between-factor variables and vocabulary retention
scores as the dependent measure, one for the RT1, the other the RT2.
Table 7 demonstrates that the interaction between the effects of the
between-factor variables on students’ RT1 scores was significant
[F(2, 88) = 3.76, p = 0.027, Z2 = 0.079]. Therefore, the ‘significant’ main
effect of vocabulary levels on RT1 scores [F(1, 88) = 14.35, p < 0.001,
Z2 = 0.140] and the same ‘significant’ results in the case of reading conditions
[F(2, 88) = 13.52, p < 0.001, Z2 = 0.235] (see also Table 7) should be treated
with caution. Table 8 and Figure 3 indicate that the two BLD conditions
produced different effects on students’ immediate vocabulary retention for
the higher and the lower level groups. Simple effect analyses (Post hoc tests)
were performed by computing a new variable with each of the original cells as a
level and found that there was a significant effect difference between cell codes
[F(5, 88) = 11.89, p < 0.001, Z2 = 0.403]. Results of multiple comparisons in
Table 9 suggest that in terms of the RT1, for the higher level group, those who
used the PBLD fared significantly better than those with no access to the
dictionary (mean difference = 1.99, p = 0.008); for the lower level group,
234 Yuzhen Chen

Table 8: Descriptive statistics of RT1 scores as a function of reading condi-

tions and vocabulary levels

EBLD PBLD ND

N mean S.D. n mean S.D. n mean S.D.

Higher 6 6.67 2.16 15 7.20 1.42 26 5.21 1.65

Lower 16 6.75 1.82 10 4.30 1.27 21 3.48 1.99
Total 22 6.28 1.86 25 6.04 1.97 47 4.44 1.99

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Figure 3: Profile plot: RT1 scores for vocabulary levels. This figure appears
in colour in the online version of the International Journal of Lexicography.
those who used the EBLD scored significantly higher than those who did not
use any dictionary (mean difference = 3.27, p < 0.001), and those with access to
the PBLD (mean difference = 2.45, p = 0.009).
Results of two-way ANOVA in Table 10 show that there was no significant
interaction between the effects of vocabulary levels and reading conditions on
students’ RT2 scores [F(2, 88) = 1.15, p = 0.322, Z2 = 0.025]. The main effect
of vocabulary levels on students’ long-term vocabulary retention was statistic-
ally significant [F(1, 88) = 13.30, p < 0.001, Z2 = 0.131], so was the case with
 Dictionary Use and Vocabulary Learning in the Context of Reading 235

Table 9: Post hoc tests for six new cell codes: multiple comparisons
Dependent variable: RT1 scores (Tukey HSD)

(I) six new cell codes (J) six new cell codes Mean Std. Sig.
Difference Error
(I-J)

Higher and EBLD Higher and PBLD 0.53 0.84 0.988

Higher and ND 1.46 0.78 0.436
Lower and EBLD Lower and PBLD 2.45* 0.70 0.009
Lower and ND 3.27* 0.57 0.000
Higher and PBLD Higher and ND 1.99* 0.56 0.008

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Lower and PBLD Lower and ND 0.82 0.67 0.817

Based on observed means. The error term is Mean Square (Error) = 2.996.
*The mean difference is significant at the 0.05 level.

Table 10: Two-way ANOVA for RT2 scores as a function of reading condi-
tions and vocabulary levels

Variable and source df Mean F Sig Partial Eta

square Squared

RT2
Vocabulary level 1 52.52 13.30 0.000 0.131
Reading condition 2 26.57 6.73 0.002 0.133
Vocabulary level *reading condition 2 4.54 1.15 0.322 0.025
Error 88 2.99

R Squared = 0.270 (Adjusted R Squared = 0.229).

Table 11: Means, standard deviations, and n for RT2 scores as a function of
reading conditions and vocabulary levels

EBLD PBLD ND

N mean S.D. n mean S.D. n mean S.D.

Higher 6 5.33 1.78 15 5.30 2.12 26 3.87 2.04

Lower 16 4.63 2.19 10 2.70 1.60 21 2.14 1.87
Total 22 4.82 2.07 25 4.26 2.30 47 3.10 2.13
236 Yuzhen Chen

reading conditions [F(2, 88) = 6.73, p = 0.002, Z2 = 0.133]. Multiple compari-

sons (Tukey HSD), together with the figures in Table 11, indicated that stu-
dents at both levels of vocabulary proficiency using the EBLD obtained
significantly better scores than the ND group (mean difference = 1.72,
p = 0.003).

3.6 Interpretation and discussion

3.6.1 BLD use vs. contextualguessing. The study yielded clear evidence to support
Hypothesis 1 proposed in Section 3.1, i.e. students using the PBLD or
the EBLD achieve significantly better results of vocabulary learning than
those without access to the dictionary. Students who used the BLDs fared

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

much better on the vocabulary comprehension test than those who didn’t,
suggesting that BLD use is a more effective strategy for vocabulary compre-
hension than contextual guessing. This result accords with the findings of
some previous studies, such as Summers (1988), Tono (1989), Knight (1994),
Koga (1995), Bogaards (2002), Hayati and Mohanmmadi (2005), and Zucchi
(2010). Although this study differs from previous research in several aspects
such as involving the use of BLDs, a type of dictionary rarely investigated
before, and using English majors as subjects, it does confirm the effectiveness
of dictionary use strategy for vocabulary comprehension.
Compared with guessing from the context, the use of dictionary can guar-
antee an exposure to richer and more accurate lexical information, provided
students have proper dictionary skills. For example, when asked to explain the
meaning of wand, many students in the ND group gave wrong answers such as
a magic trick/performance/phenomenon, or a towel/handkerchief/clothing used
by magicians; some figured out a more approximate meaning such as a tool
or something used by magicians, while a majority of students in the BLD groups
could write the correct meaning of this target word, though they were warned
in advance not to copy the exact definition from the BLD entry. Another
example is the use of a dab hand. Over 70% of the ND group wrote sentences
such as She is a dab hand with/in/of cooking while most students of the BLD
groups could use a more appropriate preposition at, as shown in the dictionary
examples. Some students in the BLD groups also made mistakes by using with/
in, probably because they did not pay attention to the dictionary examples. It
should be mentioned here that, conventionally speaking, the use of target lex-
ical items belongs to the dimension of language production rather than com-
prehension, but of course comprehension precedes production as students have
to decode before they can encode (Rundell 1999: 35). Therefore, here the
author prefers to include such cases (actually only two) in a more general
dimension of vocabulary comprehension.
It is interesting to note that despite the significantly better performance ob-
tained by the BLD groups, a close scrutiny of students’ responses to each
 Dictionary Use and Vocabulary Learning in the Context of Reading 237

Table 12: Correct answer rates for each multiple choice question

Question Question Question Question Question Question

5 6 7 8 9 10

ND group 74.4% 77.0% 48.3% 77.0% 82.8% 43.7%

BLD groups 80.9% 100% 39.3% 94.4% 75.3% 36.0%

question revealed that dictionary use did not necessarily produce better results
than non-dictionary use in every case. Table 12 shows the correct answer rates
for each question between the ND group and the two BLD groups combined.

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

For the sake of convenience, the author only analyzed the six multiple choice
questions.
It can be seen that for three out of six questions, the ND group did better
than the BLD groups. Through careful examination of the responses to these
three questions, two reasons surfaced to explain the poorer performance by the
BLD groups. One is related to students’ inadequate dictionary skills. For ex-
ample, in Question 10, students were asked to choose a sentence which con-
tained the same sense of play up as used in the text sentence, Well, yes, and
Kennedy shamelessly plays up to it. Instead of choosing the right answer He
played up the old lady’s vanity to get her support, most students in the BLD
groups chose She’s always playing up to the boss. Although the BLD did pro-
vide explanations for both play sth up and play up to sb, most students simply
ignored the information that when meaning flatter sb. in order to win favor, play
up to is usually followed by sb. instead of sth. They seemed to take it for
granted that play up to in the original sentence meant the same thing as play
up to sb. in the BLD entry instead of a ‘variant’ form play sth up. Question 9
reflects another problem with dictionary use. It concerns the meaning of aug-
ment in He augments his income by . . . Two of the four options contained the
correct meaning of this word, one is . . . so that he could increase his income (in
Option A), the other . . . as an additional means of money-making (in Option C).
However, these two options differed in the remaining part of the contents.
Many students in the BLD groups chose the wrong answer A instead of the
correct C, probably because they found the identical word increase in both
Option A and the BLD entry, thus making a hasty decision without further
reading. Had they taken time and read all the four options, they would have
chosen the correct answer quite easily. Actually, as far as 82.8% of the ND
group got it right through contextual guessing, the highest rate among all six
multiple choice questions for this group. The other source of the negative re-
sults of BLD use concerns the dictionary itself. There were occasions when the
BLD failed to provide the needed information, presented the lexical informa-
tion in a way hard to benefit its users, or gave awkward or unidiomatic L1
238 Yuzhen Chen

translation. Take Question 7 for example. The dictionary entry of wind up is as

follows:

wind up (infml ) (of a person) arrive finally in a place; end up

, : . . . . . . wind (sth) up finish (a speech, etc)
: . . . . . . wind sb up cause sb to reach a high level of
excitement or agitation : . . . . . . wind sth up
settle the affairs of and finally close (a business, company, etc)
:... ...

In this case arrive finally in a place and end up are treated as two distinct senses
of wind up. Yet the Chinese translation covers only one of the senses, omitting

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

the translation for end up, which is the exact meaning tested in Question 7.
Those who tend to read L1 translations rather than L2 definitions would very
likely ignore the definition end up and turn to other parts of the entry, thus
coming up with a wrong answer to this question. Fortunately, this entry was
improved in OALECD6 like the following:

wind up (informal) (of a person) to find yourself in a particular place or

situation ... ... ...

3.6.2 BLD use and incidental vocabulary acquisition. As revealed by the study, stu-
dents who had access to the dictionary, be it the PBLD or the EBLD, achieved
significantly better results on both vocabulary retention tests than those who
did not, thus confirming the benefit of BLD use for incidental vocabulary
learning. As can be calculated from Tables 5 and 6, there was a large effect
size of both time (Z = 0.76) and reading conditions (Z = 0.52) on vocabulary
retention scores. Since the RT2 was done seven days later than the RT1, the
factor of time surely played a major role, for a loss of lexical knowledge is
bound to occur when students have no further exposure to target lexical items.
The more noteworthy finding is the large effect size (Z = 0.52) of reading
conditions upon retention scores, suggesting that the former exerted a powerful
impact on the latter.
The advantage of dictionary use for vocabulary retention identified by the
study echoes what was found in Luppescu and Day (1993), Knight (1994), and
Hulstijn et al. (1996). It can be explained in terms of the Involvement Load
Hypothesis proposed by Laufer and Hulstijn (2001). This task-induced con-
struct involves motivational and cognitive dimensions: need, search and evalu-
ation, which can be absent or present during word processing in a natural or
artificially designed task. Compared with the ND group, the task-induced in-
volvement load of the BLD groups is higher, because the component of search
was present: students had to consult the dictionary when dealing with the task.
 Dictionary Use and Vocabulary Learning in the Context of Reading 239

According to the hypothesis, retention of hitherto unfamiliar words is condi-

tional upon the amount of involvement during word processing. The higher
involvement load imposed on the BLD groups naturally led to better retention
results than those without access to the dictionary.
As shown in Table 4, the average vocabulary retention rate of the two BLD
groups is 65.9% measured immediately after the reading task and 43.9% one
week later. Even for the ND group, the retention rate reached 42.0% on the
RT1 and 27.4% on the RT2. Compared with the results of previous studies,
these figures are quite high. One of the possible reasons is the large extent to
which the reading task contributed to incidental vocabulary acquisition. In
Knight (1994), the dictionary group retained about 20.6% on a definition-
supply retention test and the figure dropped to 14.0% after two weeks. For

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

those who did not use the dictionary, the retention rate on definition-supply
test was as low as 7.2%, but it unexpectedly rose to 9.6% after two weeks. The
reading task involved in Knight (1994) was a proposition recall, i.e. subjects
were asked to write down whatever they remembered about the contents of
reading texts. Besides, the target words in the reading texts were not high-
lighted or distinguished in any manner. In Laufer and Hill (2000) all target
words were highlighted and students were encouraged to consult them in a
specially designed dictionary program, but the retention test was performed
immediately after text reading and followed by a comprehension exercise about
the text. In other words, there was no enhancement task prior to the retention
test. Nevertheless, it was found that 47.7% of the target words were remem-
bered, much higher than those in Knight (1994). Three reasons may explain the
discrepancy: the salience of target words in reading texts, the length of reading
texts and the number of target words. Laufer and Hill used a short text of 120
words and tested 12 target words while Knight used two texts with a total of
500 words and tested 24 target words. The even higher rate of vocabulary
retention identified by the present study than that of Laufer and Hill’s may
be attributed to the word-focused reading task: students had to understand the
meaning of target lexical items while doing the reading task. In other words,
the present study proved that word-focused activities can significantly enhance
word retention, a finding supported by Paribakht and Wesche (1997), Laufer
(2001, 2003), Hill and Laufer (2003) and Peters (2007). Actually, the retention
rates yielded by the present study are quite similar to those of Hill and Laufer
(2003). Two of the three tasks involved in their study were form-oriented, one
for comprehension, the other for production. They found that the average
retention rate induced by the two form-oriented tasks was 67.1% on the im-
mediate test and 41.6% on the delayed test seven days later.

3.6.3 Comparison between BLDsin paperand electronic form. The study showed that
there was no significant difference between the PBLD and the EBLD groups in
240 Yuzhen Chen

either comprehension scores or retention scores, suggesting that the effect of

BLD use on vocabulary learning is not dependent on the form of the dictionary
when students’ vocabulary proficiency is not considered. Since the two BLDs
are exactly the same except for their form and there was no significant differ-
ence in vocabulary proficiency levels between groups, the outcomes of vocabu-
lary learning can be safely regarded as indicators of the effects between
different dictionary forms. Therefore, Hypothesis 2 postulating non-significant
difference between the PBLD and the EBLD in terms of dictionary effective-
ness for vocabulary learning is verified.
This finding corroborates the result found by the author’s previous research,
i.e. Chen 2010, though the two studies compared PBLDs with different forms
of EBLDs: one with pocket EBLDs, the other with the computer desktop BLD.

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

It is also in line with the results found other studies comparing paper diction-
aries and pocket dictionaries (e.g. Koyama and Takeuchi 2003, Iso and Osaki
2004, and Kobayashi 2007). More interestingly, the present study is supported
by the recent findings of Dziemianko (2011, in press) that there was no signifi-
cant difference between the on-line and paper versions of LDOCE5 and
OALDCE7 in terms of dictionary usefulness for language production, compre-
hension and learning. In Dziemianko (2010), COBUIL6 online was found to be
significantly more useful for language tasks and retention than its paper form,
yet its replication study, Dziemianko (2011), which yielded different findings
with LDOCE5 revealed that the better effects of e-COBUILD6 than
e-LDOCE5 are to a large extent due to its clearer and more neatly organized
layout of web-pages. Accumulating evidence seems to suggest that dictionary
effectiveness is not dependent on dictionary form. In other words, the differ-
ence in the form, paper vs. electronic, may not be the major factor accounting
for dictionary usefulness. As far as BLDs are concerned, the present study,
together with Chen (2010), confirms that the form of BLDs is not a very
important issue for the assessment of BLD usefulness in vocabulary learning.
Of course, other factors such as extra functions, the speed of use, and
portability of the dictionary etc. should also be taken into consideration for
the overall evaluation of dictionaries of different forms, yet these are not the
focus of the present study.
Interestingly, despite the overall non-significant difference between PBLD
and EBLD use for vocabulary learning, the former yielded higher comprehen-
sion scores than the latter while on retention tests, the result was reversed. The
EBLD group retained 68.63% of the vocabulary gained a week before while the
PBLD kept 64.63%. In other words, the PBLD seemed to be slightly better
than the EBLD for vocabulary comprehension while the EBLD proved to be a
little more advantageous for vocabulary retention. Actually, the PBLD group
got a higher mean score of the VLT than the EBLD group did (77.16 vs. 74.30
against a total of 100), which may account for their better performance at
vocabulary comprehension. However, this small advantage of the PBLD
 Dictionary Use and Vocabulary Learning in the Context of Reading 241

group in vocabulary proficiency did not lead to better vocabulary

retention. The slight superiority of the EBLD over the PBLD for vocabulary
retention may be attributed to their different access route. In the EBLD,
any keyboard input of headword would bring the entry information on
the screen in an instant while PBLD users have to spend more time
looking through the headwords for the needed one. Another reason may be
related to the different degree of salience of entry information, for, unlike the
EBLD, the PBLD did not use different colors to indicate different categories
of entry information except bold type, although these two are identical in
contents, font and line spacing. Furthermore, unlike the EBLD, the PBLD
used in the study is not a regular and complete dictionary which prevents
users from following up on some other entries besides the target ones. This

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

might also affect negatively on users’ performance.

3.6.4 The effects of vocabulary levels and reading conditions on incidental vocabulary
acquisition. The study revealed a significant interaction, with a medium effect
size (Z = 0.28), between the effects of vocabulary levels and reading conditions
on students’ RT1 scores. In particular, the effect size of reading conditions was
larger than that of vocabulary levels (0.48 vs. 0.37), suggesting that the former
exerted a more powerful influence on vocabulary retention than the latter.
What is more noteworthy is the finding about the effects of different BLDs
on immediate vocabulary retention for students at different vocabulary levels.
For the higher level group, PBLD use led to significantly better result than
non-dictionary use whereas for the lower level group, EBLD users fared sig-
nificantly better than the PBLD and the ND groups. In other words, students
at the higher vocabulary level were at an advantage when using the PBLD
while the EBLD proved to be more beneficial for those at the lower vocabulary
level.
As to the long-term retention measured by the RT2, the interaction between
the effects of vocabulary levels and reading conditions was not statistically
significant, though the main effect of both factors reached the 0.05 significance
level. The effect size of both factors dropped from large in the case of the RT1
to medium in the RT2 and was very close to each other in the latter case.
Obviously, the impact of reading conditions and vocabulary levels both
decreased with the lapse of time. Students at the higher vocabulary level
achieved substantially better scores on both retention tests than those at the
lower level. In other words, more proficient students remembered more words
than less proficient ones. For students at both levels of vocabulary proficiency,
those using the EBLD fared significantly better than those without access to
the dictionary. Interestingly, for the higher level group, the advantage of PBLD
use over non-dictionary use diminished to a considerable degree from the RT1
242 Yuzhen Chen

to the RT2 while students at the lower level group were still at an advantage
when using the EBLD.
The discussion above substantiates Hypothesis 3, i.e. students varying on
vocabulary proficiency levels and reading conditions fare differently on inci-
dental vocabulary acquisition.

4. Concluding remarks

The study revealed that compared with non-dictionary use, BLD use can
effectively facilitate vocabulary comprehension, indicating that dictionary use
is a more effective strategy of vocabulary learning than contextual guessing.
Useful as it is, contextual guessing might be a process prone to incomplete or

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

wrong inference, in particular for those students at the low level of L2 profi-
ciency, whereas the use of dictionary means exposure to the richer and more
accurate lexical information which is vital for correct comprehension of words.
With adequate dictionary use skills, students having access to good dictionaries
will be much more likely to achieve successful word comprehension than those
who simply rely on contextual guessing. Therefore, as a strategy of vocabulary
learning, dictionary use deserves more attention in L2 vocabulary research and
pedagogy.
It was also found that BLD use is conducive to incidental vocabulary learn-
ing as students with the aid of BLDs obtained substantially better results of
vocabulary retention than those reading in non-dictionary condition. What is
more noteworthy is that the study yielded higher vocabulary retention rates
than most of previous related studies. One of the reasons is the word-focus
approach adopted by the present author: students were given tasks involving
the comprehension of individual lexical items prior to retention tests. This is
illuminating for L2 vocabulary pedagogy. Reading alone might not be a good
solution to incidental vocabulary learning. Instead, a combination of diction-
ary use and word-focus vocabulary tasks, together with reading, can effectively
enhance vocabulary retention. Considering the limited amount of language
input students are exposed to in L2 learning context, such an approach
should be advocated in vocabulary learning.
There was no indication of significant difference in dictionary effectiveness
between the PBLD and the EBLD, suggesting that success in vocabulary
learning is not dependent on the form of the dictionary. However, despite
the relatively lower level of vocabulary proficiency of the EBLD group, they
still achieved better vocabulary retention than those using the PBLD.
Apparently, the electronic dictionary has some advantage for vocabulary
retention. Therefore, this type of dictionary should not be subject to prejudice
and is worth recommending to students. Teachers should first of all have an
enlightened attitude towards dictionaries of different media and then help their
students make an informed choice among different dictionaries.
 Dictionary Use and Vocabulary Learning in the Context of Reading 243

Despite the general advantage of BLD use over non-dictionary use for vo-
cabulary learning, some problems with dictionary use were also identified by
the author. Some of them are related to the dictionary itself, but more are
concerned with students’ inadequate dictionary use skills. For example, some
students were unable to distinguish variant forms of phrases or collocations;
some did not pay sufficient attention to dictionary examples; and some were
too careless or overconfident to make judicious choice. These problems might
not be specific to BLD use and might also be found with students using other
types of dictionaries. Therefore, teachers should pay sufficient attention to
students’ dictionary use skills and provide necessary training to help them
make the best use of the dictionary.

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Supplementary material

Supplementary material is available at International Journal of Lexicography

online.

Acknowledgments

I wish to express my heartfelt gratitude to the two anonymous reviewers of this

article for their insightful comments and constructive suggestions. I’m also
indebted to Professor Jan Hulstijn, Dr. Robert Lew and Dr. Anna
Dziemianko for their generous help. My sincere thanks also go to Dr. Paul
Bogaards for his careful editing of the article. Last but not least, I would like to
thank all the teachers who helped me supervise the study and all the students
who participated in my study.

Notes

1 In this article, the term L2 refers to second or foreign language acquired after one’s
native language (L1). The acquisition-learning distinction as proposed by S. Krashen
(1981) is not distinguished here; therefore, the terms vocabulary learning and vocabu-
lary acquisition are used interchangeably throughout the article.
2 It is acknowledged that such a simple format allows only a crude measurement of
the word knowledge elaborated by Nation (2001). However, since the purpose of the
study is not to investigate how well subjects mastered new lexical items but how much
they retained after different experimental treatment, there is reason to claim that such a
test is as adequate as any other.

References
Aizawa, K. 1999. A Study of Incidental Vocabulary Learning Through Reading by
Japanese EFL Learners. Tokyo: Tokyo Gakugei University.
Albus, D., J. Bielinski, M. Thurlow and K. Liu. 2001. The Effect of a Simplified English
Language Dictionary on a Reading Test (LEP Project Report). Minneapolis, MN:
244 Yuzhen Chen
University of Minnesota. Available at: http://education.umn.edu/NCEO/
OnlinePubs/LEP1.html.
Atkins, B. T. S. and K. Varantola. 1998. ‘Language Learners Using Dictionaries:
The Final Report on the EURALEX/AILA Research Project on Dictionary Use’.
In B. T. S. Atkins (ed.), Using Dictionaries. Studies of Dictionary Use by Language
Learners and Translators. Tübingen: Niemeyer, 21–81.
Bensoussan, M., D. Sim and R. Weiss. 1984. ‘The Effect of Dictionary Usage of EFL
Test Performance Compared with Student and Teacher Attitudes and Expectations’.
Reading in a Foreign Language, 2.2: 262–276.
Bogaards, P. 2002. ‘The Use of the DE GRUYTER WÖRTERBUCH DEUTSCH
ALS FREMDSPRACHE for receptive purposes’. In H. E. Wiegand (ed.),
Perspektiven der pädagogischen Lexikographie des Deutschen. Untersuchungen
anhand des’de Gruyter Wörterbuchs Deutsch als Fremdsprache. Tübingen: Niemeyer,

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

647–660.
Chang, S. M. 2002. The Roles of Marginal Glosses and Pocket Electronic Dictionaries in
EFL Incidental Vocabulary Learning. MA Dissertation: National Taiwan Normal
University, Taiwan.
Chen, Y. Z. 2010. ‘Dictionary Use and EFL Learning. A Contrastive Study of Pocket
Electronic Dictionaries and Paper Dictionaries’. International Journal of
Lexicography, 23.3: 275–306.
Cho, K. S. and S. D. Krashen. 1994. ‘Acquisition of Vocabulary from the Sweet Valley
Kids Series: Adult ESL Acquisition’. Journal of Reading, 37.8: 662–667.
Conceição, M. P. 2004. Vocabulário e consulta ao dicionário: analisando as relações entre
experieˆncias, crenças e ações na aprendizagem de LE. PhD Thesis, Universidade
Federal de Minas Gerais.
Dziemianko, A. 2010. ‘Paper or Electronic? The Role of Dictionary Form in Language
Reception, Production and the Retention of Meaning and Collocations’. International
Journal of Lexicography, 23.3: 257–274.
Dziemianko, A. 2011. ‘Does Dictionary Form really Matter?’. In Kaoru Akasu and
Satoru Uchida (eds), Asialex 2011 Proceedings Lexicography: Theoretical and
Practical Perspectives. Kyoto: Asian Association for Lexicography, 92–101.
Dziemianko, A. in press. ‘Why One and Two Do Not Make Three – Dictionary Form
Revisited’. eLexicography 2011 Proceedings. Ljubljana: Trojina.
Fraser, C. A. 1999. ‘Lexical Processing Strategy Use and Vocabulary Learning Through
Reading’. SSAL, 21.2: 225–241.
Gu, Y. Q. 2003. ‘Vocabulary Learning in a Second Language: Person, Task, Context
and Strategies’. TESL-EJ 7.2. Available at http://www-writing.berkeley.edu/TESL-
EJ/ej26/a4.html.
Gu, Y. Q. and R. K. Johnson. 1996. ‘Vocabulary Learning Strategies and Language
Learning Outcomes’. Language Learning, 46.4: 643–679.
Hayati, A. and M. Pour-Mohammadi. 2005. ‘A Comparative Study of Using Bilingual
and Monolingual Dictionaries in Reading Comprehension of Intermediate EFL
Students’. The Reading Matrix, 5.2: 61–66. Available at: http://www.readingmatrix
.com/articles/hayati_mohammadi/article.pdf.
Heijnen, C. 2000. ‘Vocabulary-Dictionaries–Reading: The Effects of Dictionary Use on
Reading Comprehension for MAVO Pupils’. TTWiA, 63.1: 73–82.
Hill, M. and B. Laufer. 2003. ‘Type of Task, Time-on-task and Electronic Dictionaries
in Incidental Vocabulary Acquisition’. International Review of Applied Linguistics,
41.2: 87–106.
 Dictionary Use and Vocabulary Learning in the Context of Reading 245

Horst, M. 2005. ‘Learning L2 Vocabulary Through Extensive Reading: a Measurement

Study’. The Canadian Modern Language Review, 61.3: 355–382.
Hu, M. and P. Nation. 2000. ‘Unknown Vocabulary Density and Reading
Comprehension’. Reading in a Foreign Language, 13.1: 403–430.
Hulstijn, J. H. 1992. ‘Retention of Inferred and Given Word Meanings: Experiments in
Incidental Vocabulary Learning’. In P. J. L. Arnaud and H. Béjoint (eds), Vocabulary
and applied linguistics. London: MacMillan, 113–125.
Hulstijn, J. H. 2001. ‘Intentional and Incidental Second-language Vocabulary Learning:
a Reappraisal of Elaboration, Rehearsal and Automaticity’. In P. Robinson (ed.),
Cognition and Second Language Instruction. Cambridge: Cambridge University Press,
258–286.
Hulstijn, J. H. 2003. ‘Incidental and Intentional Learning’. In C. J. Doughty and M.
H. Long (eds), The Handbook of Second Language Acquisition. Malden, MA:

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Blackwell, 349–381.
Hulstijn, J. H. 2005. ‘Theoretical and Empirical Issues in the Study of Implicit and
Explicit Second-language Learning’. SSLA, 27: 129–140.
Hulstijn, J. H. in press. ‘Incidental Learning in Second Language Acquisition’.
In C.A. Chapelle (ed.), The Encyclopedia of Applied Linguistics. Wiley-Blackwell
online.
Hulstijn, J. H., M. Hollander and T. Greidanus. 1996. ‘Incidental Vocabulary Learning
by Advanced Foreign Language Students: the Influence of Marginal Glosses,
Dictionary Use, and Reoccurrence of Unknown Words’. Modern language journal,
80.3: 327–339.
Iso, T. and S. Osaki. 2004. Denshi jisho to insatsu jisho ni miru eibun dokkai, goi
kensaku, hoji no sai. Paper presented at a conference. Tokyo, Japan.
Ji, X. H. 2009. Study of Contextual Guessing and Dictionary Consulting in Incidental
Vocabulary Learning Through Reading. MA Dissertation: China University of
Petroleum, Shangdong.
Knight, S. 1994. ‘Dictionary: the Tool of Last Resort in Foreign Language Reading? A
New Perspective’. Modern language journal, 78.3: 285–299.
Kobayashi, C. 2006. The Use of Pocket Electronic Dictionaries as Compared with
Printed Dictionaries by Japanese Learners of English. PhD Thesis, the Ohio State
University.
Kobayashi, C. 2007. ‘Comparing Electronic and Printed Dictionaries: Their Effects on
Lexical Processing Strategy Use, Word Retention, and Reading Comprehension’.
In K. B. Watts (ed.), JALT 2006 Conference Proceedings. Tokyo: JALT, 657–671.
Koga, Y. 1995. ‘The Effectiveness of Using an Electronic Dictionary in
Second Language Reading’. Bulletin of the Liberal Arts of Hiroshima University,
44: 239–244.
Koyama, T. and O. Takeuchi. 2003. ‘Printed Dictionaries vs. Electronic Dictionaries: a
Pilot Study on How Japanese EFL Learners Differ in Using Dictionaries’. Language
Education and Technology, 40: 61–79.
Krantz, G. 1991. Learning Vocabulary in a Foreign Language: A Study of Reading
Strategies. Gothenburg: Acta Universitatis Gothburgensis.
Krashen, S. 1981. Second Language Acquisition and Second Language Learning. Oxford:
Pergamon.
Laufer, B. 2000. ‘Electronic Dictionaries and Incidental Vocabulary Acquisition: does
Technology Make a Difference?’. In U. Heid (ed.), Proceedings of EURALEX 2000.
Stuttgart: IMS, Universität Stuttgart, 849–854.
246 Yuzhen Chen

Laufer, B. 2001. ‘Reading, Word-focused Activities and Incidental Vocabulary

Acquisition in a Second Language’. Prospect, 16.3: 44–54.
Laufer, B. 2003. ‘Vocabulary Acquisition in a Second Language: do Learners really
Acquire most Vocabulary by Reading? Some Empirical Evidence’. The Canadian
Modern Language Review, 59.4: 567–587.
Laufer, B. 2011. ‘The Contribution of Dictionary Use to the Production and Retention
of Collocations in a Second Language’. International Journal of Lexicography, 24.1:
29–49.
Laufer, B. and M. Hill. 2000. ‘What Lexical Information do L2 Learners Select in a
CALL Dictionary and How Does it Affect Word Recall?’. Language Learning and
Technology, 3.2: 58–76.
Laufer, B. and J. Hulstijn. 2001. ‘Incidental Vocabulary Acquisition in a Second
Language: the Construct of Task-Induced Involvement’. Applied Linguistics, 22.1:

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

1–26.
Laufer, B. and P. Nation. 1999. ‘A Vocabulary Size Test of Controlled Productive
Ability’. Language Testing, 16: 36–55.
Lew, R. 2004. Which Dictionary for Whom? Receptive Use of Bilingual and Semi-bilingual
Dictionaries by Polish Learners of English. Poznań: Motivex.
Luppescu, S. and R. Day. 1993. ‘Reading, Dictionaries, and Vocabulary Learning’.
Language Learning, 43.2: 263–287.
Nation, P. 1990. Teaching and Learning Vocabulary. New York: Newbury House
Publishers.
Nation, P. 2001. Learning Vocabulary in Another Language. Cambridge: Cambridge
University Press.
Nation, P. and P. Meara. 2010. ‘Vocabulary’. In N. Schmitt (ed.), An Introduction to
Applied Linguistics. London: Edward Arnold, 34–52.
Nesi, H. 2000. The Use and Abuse of EFL Dictionaries. Tübingen: Niemeyer.
Nesi, H. and P. Meara. 1991. ‘How Using Dictionaries Affects Performance in
Multiple-choice EFL Tests’. Reading in a Foreign Language, 8.1: 631–643.
Neubach, A. and A. Cohen. 1988. ‘Processing Strategies and Problems Encountered in
the Use of Dictionaries’. Dictionaries, 10: 10–19.
Padron, Y. N. and H. C. Waxman. 1988. ‘The Effect of ESL Students’ Perceptions of
Their Cognitive Strategies on Reading Achievement’. TESOL Quarterly, 22.1:
146–150.
Paribakht, T. S. and M. Wesche. 1993. ‘Reading Comprehension and Second Language
Development in a Comprehension-based ESL Program’. TESL Canada Journal, 11.1:
9–29.
Paribakht, T. S. and M. Wesche. 1997. ‘Vocabulary Enhancement Activities and
Reading for Meaning in Second Language Vocabulary Acquisition’. In J. Coady
and T. Huckin (eds), Second Language Vocabulary Acquisition. A Rationale for
Pedagogy. Cambridge: Cambridge University Press, 174–200.
Peters, E. 2007. ‘Manipulating L2 Learners’ Online Dictionary Use and its Effect on L2
Word Retention’. Language learning and Technology, 11.2: 36–58.
Read, J. 1997. ‘Vocabulary and Testing’. In N. Schmitt and M. McCarty (eds),
Vocabulary: Description, Acquisition and Pedagogy. Cambridge: Cambridge
University Press, 303–320.
Read, J. 2004. ‘Research in Teaching Vocabulary’. Annual Review of Applied Linguistics,
24.1: 146–161.
 Dictionary Use and Vocabulary Learning in the Context of Reading 247

Read, J. 2007. ‘Second Language Vocabulary Assessment: Current Practices and New
Directions’. International Journal of English Studies, 7.2: 105–125.
Ronald, J. 2003. ‘A Review of Research into Vocabulary Acquisition through
Dictionary Use. Part 1: Intentional Vocabulary Learning through Dictionary Use’.
Studies in the Humanities and Sciences, 44.1: 285–307.
Rundell, M. 1999. ‘Dictionary Use in Production’. International Journal of
Lexicography, 12.1: 35–53.
Schmidt, R. 1994. ‘Deconstructing Consciousness: in Search of Useful Definitions for
Applied Linguistics’. AILA Review, 11: 11–26.
Schmidt, R. 2001. ‘Attention’. In P. Robinson (ed.), Cognition and Second Language
Instruction. Cambridge: Cambridge University Press, 3–32.
Schmitt, N., D. Schmitt and C. Clapham. 2001. ‘Developing and Exploring the Behavior
of Two New Versions of the Vocabulary Levels Test’. Language Testing, 18.1: 55–88.

Downloaded from http://ijl.oxfordjournals.org/ at Leiden University on July 21, 2014

Scholfield, P. 1997. ‘Vocabulary Reference Works in Foreign Language Learning’.
In N. Schmitt and M. McCarty (eds), Vocabulary: Description, Acquisition and peda-
gogy. Cambridge: Cambridge University Press, 279–302.
Shi, Z. L. 2008. ‘A Comparative Study of Three Learning Strategies in EFL Students’
Incidental Vocabulary Acquisition’. CELEA Journal, 31.6: 91–101.
Summers, D. 1988. ‘The Role of Dictionaries in Language Learning’. In R. Carter and
M. McCarthy (eds), Vocabulary and Language Teaching. London: Longman,
111–125.
Szczepaniak, R. 2006. The Role of Dictionary Use in the Comprehension of Idiom
Variants. Tübingen: Niemeyer.
Tono, Y. 1989. ‘Can a Dictionary Help One Read Better? On the Relationship between
E.F.L. Learners’ Dictionary Reference Skills and Reading Comprehension’.
In G. James (ed.), Lexicographers and Their Works. Exeter: University of Exeter
Press, 192–200.
Tono, Y. 2001. Research on Dictionary Use in the Context of Foreign Language Learning:
Focus on Reading Comprehension. Tübingen: Niemeyer.
Vermeer, A. 2001. ‘Breadth and Depth of Vocabulary in Relation to L1/L2 Acquisition
and Frequency of Input’. Applied Psycholinguistics, 22.2: 217–234.
Waring, R. and P. Nation. 2004. ‘Second Language Reading and Incidental Vocabulary
Learning’. Angles on the English Speaking World, 4: 97–110.
Welker, H. A. 2010. Dictionary Use: a General Survey of Empirical Studies. Brasilia:
Author’s Edition. Available at www.let.unb.br/hawelker/images/stories/professores/
documentos/dictionary_use_research.pdf.
Wu, J. S, J. G. Lang and Q. Dang. 2007. ‘Incidental Vocabulary Learning and the
Involvement Load Hypothesis’. Foreign Language Teaching and Research, 5: 360–366.
Zahar, R., T. Cobb and N. Spada. 2001. ‘Acquiring Vocabulary Through Reading:
Effects of Frequency and Contextual Richness’. The Canadian Modern Language
Review, 57.4: 541–572.
Zhang, L. 2007. Study of Incidental Vocabulary Learning of Chinese Intermediate English
Learners. MA Dissertation: Shanghai Jiao Tong University, Shanghai.
Zucchi, A. M. T. 2010. O dicionário nos estudos de lı´nguas estrangeiras: os efeitos de seu
uso na compreensão escrita em italiano. PhD Thesis, Universidade de São Paulo, São
Paulo.