Language Learning 46:4, December 1996, pp.

643-679

Vocabulary Learning Strategies and

Language Learning Outcomes

Yongqi Gu Robert Keith Johnson

Hong Kong Institute University of Hong Kong
of Education

We aimed to establish the vocabulary learning strate-

gies used by Chinese university learners of English and
the relationship between their strategies and outcomes in
learning English. We asked 850 sophomore non-English
majors at Beijing Normal University to complete a vocabu-
lary learning questionnaire. We correlated replies to the
questionnaire with results on a vocabulary size test and on
the College English Test (CETBAND2). Participants
reported using a wide variety of vocabulary learning strat-
egies. In a multiple regression analysis, Self-Initiation
and Selective Attention, two metacognitive strategies,
emerged as positive predictors of CETBAND2 scores.
Contextual guessing, skillful use of dictionaries, note-
taking, paying attention to word formation, contextual
encoding, and activation of newly learned words also
positively correlated with the two test scores. However,

Yongqi Gu, Department of English; Robert Keith Johnson, Department of

Education.
An earlier version of this article was presented at the International
Language in Education Conference held in Hong Kong, 14-16 December
1994. We gratefully acknowledge the cooperation of students and teachers
at Beijing Normal University.
Correspondence concerning this article should be addressed to Yongqi
Gu, Room 603, Bonham Campus, Department of English, The Hong Kong
Institute of Education, 2 Hospital Road, Hong Kong. Telephone (852) 2803-
7452. Internet: pegu@boc.ied.edu.hk

643
644 Language Learning Vol. 46, No. 4

visual repetition of new words was the strongest negative

predictor of both vocabulary size and general proficiency.
Furthermore, strategies aiming at vocabulary retention
only related more to vocabulary size than to English
proficiency. We identified 5 approaches to learning. These
strategy combinations, rather than individual strategies,
may have made the difference in these people's learning.

The word vocabulary has long connoted word lists, and

vocabulary learning strategies have been tantamount to tech-
niques that help commit these lists to memory. Most research on
vocabulary learning strategies has therefore explored various
methods of vocabulary presentation and their corresponding effec-
tiveness in retention (Meara, 1980). Hence, most studied are
memory strategies, one of the many aspects of vocabulary learning
strategies,' on the presupposition that strategies good for vocabu-
lary retention will also benefit language learning in general.
Some earlier research focused on rehearsal strategies and
addressed questions such as the number of repetitions needed to
learn a list (Crothers & Suppes, 1967; Lado, Baldwin, & Lobo,
1967), the optimum number of words to be learned at one time
(Crothers & Suppes, 1967), or the timing of repetitions (Anderson
& Jordan, 1928; Seibert, 1927). Overall, rote repetition appears
less efficient than using spaced recall and structured reviews
(Atkinson, 1972; Royer, 1973; Seibert, 1927); silent repetition and
silent writing are less effective than repeating the words aloud
(Gershman, 1970; Seibert, 1927).
Research into mnemonics has continued through the past
two decades, following Atkinson (1975) and Atkinson and Raugh
(1975). The bulk of such interest has centered on the key-word
method, a technique that starts with an acoustic link (i.e., finding
a keyword in L1 that sounds like the foreign word) then links the
keyword and the foreign word by means of an interactive image.
Despite awesome evidence showing its superiority over any other
strategies (extensive reviews in Cohen, 1987; Meara, 1980; Na-
tion, 1982; Paivio & Desrochers, 1981), the keyword method (or
 Gu and Johnson 645

any other mnemonic technique, for that matter) suffers from its
fundamental assumption that vocabulary learning largely means
list learning. As Meara (1980) rightly pointed out, these labora-
tory experiments "completely ignore the complex patterns of
meaning relationships that characterize a proper, fully formed
lexicon" (p. 225). Consequently, even if these memory crutches do
not interfere with retrieval and production, though researchers
have presented little convincing evidence that they do not, they
are unlikely to play a major role in the development of a dynamic
living lexicon in the target language.
Developments in lexical semantics and studies on the mental
lexicon form a different, more recent focus on vocabulary learning.
Componential analysis and the "paradigmatic versus syntagmatic"
conceptions of the mental lexicon, for example, have prompted the
development of the semantic field, semantic network / map, or
semantic grid strategies, which present and organize new words
in terms ofmaps or grids of interrelated lexical meanings (Channell,
1981, 1988; Crow & Quigley, 1985). These semantically based
strategies, though intuitively appealing, tend to be prescriptive.
Although some empirical evidence does suggest their effective-
ness (e.g., Crow & Quigley, 1985), other studies have warned of
the danger of presenting closely related new words at the same
time (Higa, 1963; Nation, 1994; Tinkham, 1993). Researchers
have little idea whether these strategies make vocabulary reten-
tion easier, let alone how much they help develop the active use of
vocabulary thus learned.
Most previous research either ignores or overlooks one of the
crucial characteristics of second language (L2) learners that
makes them fundamentally different from mother-tongue (L1)
learners of vocabulary: Beginning L2 learners most need not
concept-formation but threshold-level L2 skill, without which the
simple retention of word lists is meaningless. In fact, some
evidence (e.g., Gu, 1994) shows that inadequate understanding of
vocabulary is but one aspect of language development, which
must relate to and integrate with other aspects results in serious
consequences.
646 Language Learning Vol. 46, No. 4

Research on vocabulary learning through reading, a direc-

tion that has received rigorous scrutiny particularly in recent
years, has dealt with this issue head-on (e.g., Huckin, Haynes, &
Coady, 1993). Research now has extensively demonstrated that
vocabulary can be acquired through reading (Krashen, 1989;
Parry, 1991; Saragi, Nation, & Meister, 1978) or any "fully
contextualized activities" (p. 240), to use Oxford and Scarcella's
(1994) term. Furthermore, vocabulary words thus acquired re-
tain not just their referential meaning but also the syntactic,
pragmatic, and even emotional information from their context.
Most important, vocabulary is no longer thought of as acquired as
separate items; it is an integral part of discourse and is developed
along with reading strategies such as contextual guessing. That
said, researchers need to remember two points when examining
this promising line of research. First, vocabulary acquisition
through reading presumes a basic reading ability in the L2, a skill
beginning learners possess only to a limited extent. Learning to
read an L2 with totally different orthography—for example,
Chinese students learning English as a foreign language (EFL),
seriously challenges not just the development of reading ability
but also vocabulary learning through reading (Haynes, 1990).
Second, instruction should not overemphasize the incidental /
indirect, or even subliminal, acquisition of words at the expense
of intentional and direct studying of vocabulary (not necessarily
in lists) that has proved so effective among good EFL learners in
"input-poor environments" (Kouraogo, 1993, p. 165), where learn-
ers unluckily have insufficient reading materials at their dis-
posal. These vocabulary learning strategies might, in any con-
text, valuably add to the acquisition of vocabulary through exten-
sive reading; they should lead to increased retention of that new
vocabulary and increased availability of those items for active use.
Thus far, research has largely sought the "best" strategy for
vocabulary retention. In reality, however, learners tend to use a
variety of strategies in combination (Ahmed, 1989; Gu, 1994;
Sanaoui, 1995). Even discussants of "approaches" to vocabulary
learning normally take a stand either on the "direct" side or the
 Gu and Johnson 647

"indirect" side, as if direct and indirect methods were mutually

exclusive. A more balanced and integrated approach is likely to
be the most effective. Students consistently adopt types of
strategies based either on their beliefs about vocabulary and
vocabulary learning (cf. Abraham & Vann, 1987; Horwitz, 1987),
or on other preexisting cognitive or social factors. Although each
strategy contributes to success or failure, consistent employment
of certain types of strategies forms an approach to vocabulary
learning that may considerably influence the outcomes of L2
learning (cf. Sanaoui, 1995). Therefore, how different learners
combine different strategies and how this affects their learning
outcomes warrant studying as much as, perhaps more than, the
effects of individual strategies.
Research questions. Vocabulary study no longer languishes
as the neglected "Cinderella" of applied linguistics (cf. Carter &
McCarthy, 1988; Laufer, 1986; Lord, 1974; Meara, 1980). Yet too
many questions remain unanswered. To begin with, among a
spectrum of vocabulary learning strategies, do any strategies
work better or worse than others? Do all strategies good for
vocabulary retention automatically benefit the development of
general L2 proficiency? Do learners stick to certain types of
strategies and adopt distinctive approaches to vocabulary learn-
ing? If so, how does that influence outcomes? Above all, among a
whole range of vocabulary strategies, from initial handling of a
new word, to contextual guessing, to dictionary use, to note-
taking, to reinforcement strategies, and to the activation and use
of the newly learned word, which do EFL learners tend to employ?
In the context of tertiary students in a major teacher-training
institution in China, we also ask whether Chinese learners em-
ploy more rote learning strategies, the caricature of Asians so
often seen in the literature, than other "better" strategies en-
dorsed by North American researchers (Field, 1984; O'Malley,
Chamot, Stewner-Manzanares, Kupper, & Russo, 1985; Oxford &
Scarcella, 1994; Politzer & McGroarty, 1985)?
648 Language Learning Vol. 46, No. 4

Method

Participants

All second-year non-English majors at Beijing Normal Uni-

versity (BNU), an intact grade totaling 27 classes of university
students,2 participated in this study. By the time of the study,
these learners had all had 6 years of English learning experience
in secondary schools (932 contact hours), and had just completed
1 year (140 hours) at BNU (The State Education Commission,
PRC, 1986). After initial elimination of unusable data, 850 of
these students formed the final pool of participants.

Instruments

Questionnaire. We used a vocabulary learning question-

naire (VLQ Version 3, see Appendix3) to elicit students' beliefs
about vocabulary learning and their self-reported vocabulary
learning strategies. The questionnaire, written in Chinese, re-
flected previous quantitative and qualitative research (e.g., Ahmed,
1989; Gu, 1994; Oxford, 1990; Politzer & McGroarty, 1985) and
item analyses that removed redundant items from two earlier,
pilot versions. It included 3 sections. Section 1, Personal Data,
asked about each respondent's demographic information, nation-
wide college-entrance English score, and score on a university-
wide College English Test (CETBAND2), which the students had
taken at the end of their first year at BNU—about 2 months
previously. Section 2, Beliefs About Vocabulary Learning, in-
cluded 17 statements representing 3 dimensions of beliefs: Vo-
cabulary should be memorized; Vocabulary should be picked up
naturally; and Vocabulary should be studied and used. We asked
participants to rate each statement on a 7-point scale from
Absolutely Disagree (1) to Absolutely Agree (7). Section 3, Vo-
cabulary Learning Strategies, contained 91 vocabulary learning
behaviors divided into two major parts: Metacognitive Regulation
and Cognitive Strategies. We asked respondents to rate each
statement, again on a 7-point scale, ranging from Extremely
 Gu and Johnson 649

Untrue of Me (1) to Extremely True of Me (7). Table 1 outlines the

major dimensions in the questionnaire, the categories under each
dimension, the number of items under each category, acronyms
for the independent variables that will appear in later sections,
and the internal consistency reliability of each category.
Vocabulary size tests. We combined two vocabulary size tests
as our vocabulary size indicator (VOCSIZE). We adapted the first
test from Goulden, Nation, and Read (1990). We asked the
students to provide a Chinese equivalent, a synonym, or para-
phrase showing their understanding of at least one meaning, for
each of the 50 target words and to leave "unknowns" blank. Lest
this test turn out too difficult to enable discrimination among this
group oflearners (see Bird, 1994; Izawa, 1993), we added Nation's
(1990, pp. 266-268) Vocabulary Levels Test at the 3,000-word
level. Scores summed the number of correct responses in both
tests.
Proficiency Measures. The best available English proficiency
measure at the time of study was a composite score (CET Band 2,
85%, 10 quizzes taken throughout the year, 10%, and the teacher's
overall rating, 5%). This composite might better indicate profi-
ciency than a single test; we therefore used it as our English
proficiency measure (referred to hereafter as CETBAND2). The
CET Band 2 test itself, mock-Band 44 in format, comprised
sections on listening comprehension (15%), vocabulary (10%),
structure (10%), reading comprehension (30%), doze (10%), and
sentence translation from Chinese into English (10%).
In addition, we obtained participants' English scores on their
college entrance examinations, a nationwide, standardized test
taken each year by hundreds of thousands of high school leavers
in China. The full score is usually 100, except for a few experimen-
tal cities where a score of either 900 or 150 was also possible. We
converted the scores for the minority of participants with scores of
the latter two types into percentages to make them comparable to
the majority. This variable we coded as PRESCORE.
 Table 1
Dimensions and Categories (VLQ Version 3: 108 items)

Dimensions and Categories No. of Items Variable Labels Reliability

Beliefs About Vocabulary Learning 17

Words Should Be Memorized 8 MEMORIZ a=.68
Words Should Be Acquired in Context: Bottom-Up 4 ACQUIRE a=.46
Words Should Be Studied and Put to Use: Top—Down 5 LEARN oc=.46
Metacognitive Regulation 12
Selective Attention 7 ATTEND a=-.71
Self-Initiation 5 SELFINI a=.71
Guessing Strategies 12
Using Background Knowledge/Wider Context 7 WIDECUE cc=.79
Using Linguistic Cues/Immediate Context 5 LOCOCUE oc-,.68
Dictionary Strategies 17
Dictionary Strategies for Comprehension 4 DICOMPR a=.67
Extended Dictionary Strategies 8 DICEXTN a=.81
Looking-Up Strategies 5 DICLOOK a=.63
Note-Taking Strategies 9
Meaning-Oriented Note-Taking Strategies 5 NOTEMNG a..68
Usage-Oriented Note-Taking Strategies 4 NOTEUSE a=.75
Rehearsal Strategies 12
Using Word Lists 6 VOCLIST a=.73
Oral Repetition 3 ORALREP a=.50
Visual Repetition 3 VISUREP a=.56
Encoding Strategies 24
Association/Elaboration 4 ASSOCIA a..61
Imagery 4 IMAGERY a=.67
Visual Encoding 3 VISUCOD a=.48
Auditory Encoding 3 AUDICOD a=.73
Using Word-Structure 3 WDFORM a=.70
Semantic Encoding 3 SEMANET a=.58
Contextual Encoding 4 CONTEXT a=.72
Activation Strategies 5 ACTIVAT a=.78
652 Language Learning Vol. 46, No. 4

Procedures

We held a briefing about one week before data collection for

all teachers whose students were to participate. We distributed
a guide to administration during the briefing. The teachers then
took the questionnaires to class and administered them, using
about 30 minutes of class time. We then collected them from the
teachers immediately after class and coded them for analysis.
The 2 vocabulary size tests were administered immediately
after the questionnaire was filled in and were collected together
with the questionnaire. We obtained both CETBAND2 and
PRESCORE as part of the Personal Data in Section 1 of the
anonymous questionnaire.

Analyses

We obtained descriptive statistics first to see the overall

patterns of vocabulary learning strategies used by the students.
We then performed correlation analyses between all independent
variables and the 2 dependent variables—vocabulary size and
English proficiency—to see how various strategies related to
vocabulary size and general proficiency. We subsequently did
multiple regression on the 2 dependent variables to identify the
best predictors from all variables considered together. Finally, we
performed a cluster analysis to identify the strategy profiles of
different types of learners.

Results and Discussion

Learning Strategies

Table 2 presents descriptive statistics on each category of

beliefs and strategies. A look at the 3 types of beliefs tells us that,
overall, these learners emphasized the belief that vocabulary
should be memorized (MEMORIZ) (M=3.04, SD=.83) less than the
other 2 belief categories. They predominantly believed that
vocabulary should be carefully studied and put to use (LEARN)
 Gu and Johnson 653

Table 2
How Chinese Learners Learn Vocabulary: Self Reports

Categories and Strategies M SD n

Beliefs
Words Should Be Memorized 3.04 0.83 849
Acquire Vocabulary in Context 4.94 0.78 850
Learn Vocabulary and Put It to Uuse 5.74 0.62 847

Metacognitive Regulation
Selective Attention 4.23 0.86 822
Self-Initiation 4.58 1.00 829

Guessing Strategies
Wider Context 4.60 0.85 824
Immediate Context 4.47 0.84 824

Dictionary Strategies
Comprehension 4.97 1.00 830
Extended Dictionary Strategies 4.82 0.93 820
Looking-Up Strategies 4.55 0.94 823

Note-Taking Strategies
Meaning-Oriented Note-Taking 4.15 0.99 824
Usage-Oriented Note-Taking 4.27 1.14 834

Rehearsal Strategies
Using Word Lists 3.15 0.99 824
Oral Repetition 4.20 1.07 840
Visual Repetition 3.92 1.17 833

Encoding Strategies
Association/Elaboration 3.69 0.97 826
Imagery 3.11 1.00 833
Visual Encoding 4.00 1.08 831
Auditory Encoding 3.69 1.19 832
Using Word-Structure 3.96 1.16 840
Semantic Encoding 3.24 1.03 839
Contextual Encoding 4.11 1.07 829

Activation Strategies 3.80 1.05 827

654 Language Learning Vol. 46, No. 4

(M=5.74, SD=.62), though they also tended to agree that words

can be acquired in context (ACQUIRE) (M=4.94, SD=.78).5
The students were generally more positive with regard to
regulating their own vocabulary learning with Self-Initiation
(SELFINI) ranked higher (M=4.58, SD=1.00) than Selective At-
tention (ATTEND) (M=4.23, SD=.86). The students reported
extensive use of guessing strategies when reading, employing
both local cues (LOCOCUE) (M=4.47, SD=.84) and wider cues
(WIDECUE) (M=4.60, SD=.85). They seemed to use dictionary
strategies widely, both for comprehension (DICOMPR) (M=4.97,
SD=1.00) and for vocabulary learning (DICEXTN) (M=4.82,
SD=.93). They also reported a variety of looking-up strategies
(DICLOOK: e.g., looking up the root if an affixed form of a new
word cannot be found in the dictionary; M=4.55, SD=.94). They
also used note-taking strategies, notes containing either mean-
ing-related (NOTEMNG; M=4.15, SD=.99) or usage-related infor-
mation (NOTEUSE; M=4.27, SD=1.14) or both. They were gener-
ally less likely to use rehearsal (often associated with rote learn-
ing) and encoding strategies than other strategies. Of the re-
hearsal strategies, Oral Repetition ranked highest (ORALREP;
M=4.20, SD=1.07) and the Use of Vocabulary Lists lowest
(VOCLIST; M=3.15, SD=.99). Among encoding strategies, Con-
textual Encoding (arguably the least "rote") ranked highest (CON-
TEXT; M=4.11, SD=1.07), and Imagery, for example, associating
a part of a word with word meaning: two "eyes" in the word look)
the lowest (IMAGERY; M=3.11, SD=1.02). Activation strategies
ranked relatively low (ACTIVAT; M=3.80, SD=1.05), not surpris-
ing given the extent to which such strategies demand the manage-
ment of learning time and effort.
Generally, the participants did not seem to believe in memo-
rization; in accordance with their beliefs, they generally re-
sponded negatively to rote memorization strategies, except for
oral repetition. Neither the mnemonic devices so much valued by
some psychologists nor the semantically based strategies favored
by some linguists enjoyed much popularity among these learners.
What these students did most centered on guessing, dictionary
 Gu and Johnson 655

work, and note-taking (cf. Chern, 1993). These findings do not

indicate whether Chinese learners employ more rote strategies
than do students of other cultural backgrounds. In fact, they may
well utilize more rote strategies compared with students from
western cultures. Results here only suggest that these learners do
not value rote learning as highly as other strategies, and that they
employ a wide range of vocabulary learning strategies.

Learning Strategies and Learning Outcomes

We obtained simple correlations among 24 independent

variables (3 belief variables, 2 metacognitive regulation vari-
ables, 18 cognitive strategy variables covering the whole process
of vocabulary learning, and a time variable representing extra-
curricular time spent weekly on English learning) and the 2
dependent variables (English proficiency and vocabulary size);
these appear in Table 3. Listwise deletion of missing data reduced
the sample size to 548.
Believing in memorization was negatively correlated with
both CETBAND2 (r.—.2339, p<.001) and VOCSIZE (r=—.1273,
p<.01). Visual Repetition (e.g., repeating a new word to oneself by
writing it again and again) also was negatively correlated with the
2 dependent variables (p<.001 for both). The 2 metacognitive
regulation variables were positively correlated with the 2 depen-
dent variables, as were the 2 guessing variables and the 2 note-
taking variables. Among the 3 dictionary variables, only Looking
Up Words for Comprehension did not reveal a significant correla-
tion. Some interesting patterns showed up for the other indepen-
dent variables. Mnemonic devices (e.g., imagery, visual associa-
tions, and auditory associations) probably related more to vocabu-
lary size; their correlations with general English proficiency were
mostly insignificant or even negative. Likewise, semantic encod-
ing strategies and word list learning correlated highly with
vocabulary size, but not with general English proficiency. Contex-
tual encoding, on the other hand, correlated significantly with
both dependent variables. We then examined the relationship
656 Language Learning Vol. 46, No. 4

Table 3
Correlations Among 24 Independent Variables and 2 Dependent
Variables

MEMORIZ ACQUIRE LEARN ATTEND

CETBAND2 -0.23** 0.09 0.06 0.26**

VOCSIZE -0.13* 0.10* 0.07 0.24**
SELFINI WIDECUE LOCOCUE DICOMPR
CETBAND2 0.30** 0.24** 0.23** 0.08
VOCSIZE 0.35** 0.17** 0.19** 0.08
DICEXTN DICLOOK NOTEMNG NOTEUSE
CETBAND2 0.27** 0.23** 0.17** 0.16**
VOCSIZE 0.23** 0.24** 0.23** 0.18**
VOCLIST ORALREP VISUREP ASSOCIA
CETBAND2 0.08 0.15** - 0.24** 0.10*
VOCSIZE 0.16** 0.03** - 0.23** 0.18**
IMAGERY VISUCOD AUDICOD WDFORM
CETBAND2 -0.01 0.05 0.01 0.14**
VOCSIZE 0.06 0.02 0.08 0.15**
SEMANET CONTEXT ACTIVAT TIME
CETBAND2 0.09 0.25** 0.18** 0.11*
VOCSIZE 0.24** 0.22** 0.31** 0.13*
n=548. *p<.01, **p<.001.
MEMORIZ=Memorize words; ACQUIRE=Acquire words in context;
LEARN=Study and put words to use; ATTEND=Selective attention;
SELFINI=Self-initiation; WIDECUE=Wider context; LOCOCUE=Immedi-
ate context; DICOMPR=Dictionary strategies for comprehension;
DICETXN=Extended dictionary strategies; DICLOOK=Looking-up strate-
gies; NOTEMNG=Meaning-oriented note-taking strategies; NOTEUSE=
Usage-oriented note-taking strategies; VOCLIST=Use word lists;
ORALREP=Oral repetition; VISUREP=Visual repetition; ASSOCIA=Asso-
ciationelaboration; VISUCOD=Visual encoding; AUDICOD=Auditory en-
coding; WDFORM=Use word-structure; SEMANET=Semantic encoding;
CONTEXT=Contextual encoding; ACTIVAT=Activation strategies.

between vocabulary size and general English proficiency, and

obtained a reasonably high correlation coefficient (r=.5338,p<.001),
signifying, nevertheless, 28.5% of shared variance only.
 Gu and Johnson 657

Table 4
Multiple Regression: Predictors of CETBAND2

Variables R2
Step Entered Beta t p Change

Block 1 1 LEARN -.04 -0.95 .34 .05

2 MEMORIZ -.08 -2.02 .04
3 ACQUIRE .01 0.26 .80
Block 2 4 ATTEND .17 3.46 .00 .10
5 SELFINI .15 3.28 .00
Block 3 6 LOCOCUE -.11 -1.72 .09 .00
7 WIDECUE .08 1.29 .20
Block 4 8 DICOMPR -.01 -.29 .77 .01
9 DICLOOK .07 1.44 .15
10 DICEXTN .07 1.15 .25
Block 5 11 NOTEUSE -.04 -.68 .49 .00
12 NOTEMNG -.01 -.25 .80
Block 6 13 VISUREP -.18 -4.40 .00 .03
14 ORALREP .11 2.80 .01
15 VOCLIST .00 .02 .98
Block 7 16 IMAGERY -.15 -3.07 .00 .03
17 AUDICOD -.09 -1.80 .07
18 WDFORM -.06 -1.25 .21
19 VISUCOD .03 .59 .56
20 CONTEXT .16 3.40 .00
21 SEMANET .08 1.41 .16
22 ASSOCIA .05 .89 .37
Block 8 23 ACTIVAT -.07 -1.38 .17 .00
Block 9 24 TIME .05 1.23 .22 .00
F=7.25, p‹.001, R2=.22.

We next performed multiple regression analyses to get a

better picture of the relationship between the independent and
dependent variables when considering all independent variables
simultaneously. We entered 24 independent variables in 9 blocks
658 Language Learning Vol. 46, No. 4

Table 5
Multiple Regression: Predictors of VOCSIZE

Variables R2
Step Entered Beta t p Change

Block 1 1 LEARN .01 0.14 .89 .02

2 MEMORIZ .02 0.56 .57
3 ACQUIRE .04 1.09 .28
Block 2 4 ATTEND .08 1.68 .09 .11
5 SELFINI .16 3.51 .00
Block 3 6 LOCOCUE -.04 -0.66 .51 .00
7 WIDECUE -.03 -0.43 .67
Block 4 8 DICOMPR .03 0.59 .56 .01
9 DICLOOK .14 2.65 .01
10 DICEXTN -.05 -.89 .37
Block 5 11 NOTEUSE -.08 -1.48 .14 .00
12 NOTEMNG .06 1.10 .27
Block 6 13 VISUREP -.18 -4.47 .00 .03
14 ORALREP .01 0.31 .76
15 VOCLIST .04 0.82 .41
Block 7 16 IMAGERY -.12 -2.35 .02 .02
17 AUDICOD -.06 -1.16 .25
18 WDFORM -.06 -1.15 .25
19 VISUCOD -.07 -1.48 .14
20 CONTEXT .06 1.13 .26
21 SEMANET .10 1.95 .05
22 ASSOCIA .10 1.66 .10
Block 8 23 ACTIVAT .13 2.28 .02 .01
Block 9 24 TIME .09 2.45 .01 .01
F=6.54, p<.001, R2=.21.

in an order roughly characterizing a normal vocabulary learning

process (Tables 4 & 5). We found 7 variables significantly pre-
dicted CETBAND2 (Table 4). Both Self-Initiation and Selective
Attention, the 2 metacognitive regulation variables, turned out to
best predict overall proficiency in EFL learning. Contextual
 Gu and Johnson 659

Encoding and Oral Repetition also entered the equation as signifi-

cant positive predictors. On the other hand, Visual Repetition,
Imagery mnemonics, and believing in Memorization emerged as
significant but negative predictors of overall proficiency.
We subjected the same group of independent variables to a
multiple regression analysis against VOC SIZE as the dependent
variable (Table 5). Self-Initiation again emerged as the best
predictor, followed by dictionary Looking-Up strategies, extracur-
ricular Time spent on English, and intentional Activation of new
words learned. Semantic Encoding also seemed to play a role in
predicting vocabulary size. Visual Repetition and Imagery encod-
ing again emerged as strong negative predictors. Learners'
vocabulary sizes seem very much related to, among other things,
the learners' self-initiation in learning, their skillful use of a
dictionary, their willingness to spend extracurricular time to
practice newly learned items, and their remembering words in
semantically meaningful groups. Learners should not, on the
other hand, depend on visual repetition and fanciful imagery
techniques to remember the words they might thus spend so much
time on.
But the size of vocabulary, though highly correlated with
language proficiency, forms only part of a living language; strat-
egies highly correlated with vocabulary size may not demonstrate
a similar relationship with overall proficiency. Students would
benefit more if they aimed at learning the language skills rather
than at just remembering English equivalents of all Chinese
words.
Learning a word includes much more than remembering the
orthographic and phonological forms and their corresponding
meanings. A large part of EFL vocabulary learning involves
learning to use words syntactically and pragmatically (Richards,
1976). Vocabulary learning should hence aim toward vocabulary
in action. In this connection, some theorists (e.g., Carter, 1987;
Judd, 1978; McCarthy, 1984) also see vocabulary not as items in
isolation, but as a skill to be developed. In other words, in addition
to remembering the form-meaning association, learning the skill
660 Language Learning Vol. 46, No. 4

of recognizing a word automatically in natural contexts, the skill

of guessing what a word means, and most importantly, the skill of
using a word correctly and appropriately should be the purpose of
vocabulary learning. Teachers should adopt materials, teaching
methods, exercises, and evaluation techniques that encourage
developing learning strategies suitable for skill learning. If it does
not get through to the students that words do not just mean
something in isolation, they could continue, for example, to insert
foreign words in Ll sentences, and in the wrong part of speech.
That said, the previous results and implications should be
understood in the context that the students' vocabulary-related
beliefs and strategies explained only about 20% of the variance in
either vocabulary size or English proficiency. However, this is a
substantial proportion given the number of related factors not
investigated here. Within these limitations, among vocabulary
strategies examined here, strategies that aim only for retaining
vocabulary words did not necessarily lead to the development of
general English proficiency, and learning vocabulary in its natu-
ral contexts positively related to both vocabulary size and general
proficiency.

Types of Learners: Success and Failure

As discussed earlier, learners seldom use one single strategy

in learning vocabulary. Perhaps their choice of strategy combina-
tions, rather than individual strategies, results in learning differ-
ences. We therefore conducted the following cluster analysis to
classify learners by their strategy profiles and learning outcomes.
Again, we used listwise deletion of missing data, which reduced
the sample size to 486.
Besides all independent and dependent variables mentioned
above, we added another variable, college entrance score
(PRESCORE) to this analysis, in comparison with CETBAND2,
which the participants took one year after they entered the
university. To make values comparable between different types
of variables, we converted all original values into z scores; thus,
 Gu and Johnson 661

PRESCORE
MEMORIZ —1.35
ACQUIRE
LEARN
ATTEND
SELFINI
WIDECUE
LOCOCUE —1.89
DICOMPR
DICETXN
on DICLOOK
NOTEMNG —1.15
-1 NOTEUSE
6
VOCLIST
ORALREP —1.85
VISUREP 02
ASSOCIA 1.73
di:j IMAGERY 1.75
VISUCOD —1.75
AUDICOD—1.80
WDFORM
SEMANET
CONTEXT
ACTIVAT
TIME
2.20
VOCSIZE
CETBAND2 1.47
—2.5 —2.0 —1.5 —1.0 —0.5 0 0.5 1.0 1.5 2.0 2.5
z Scores
PRESCORE=College entrance exam scores; MEMORIZ=Memorize words;
ACQUIRE=Acquire words in context; LEARN=Study and put words to use;
ATTEND=Selective attention; SELFINI=Self-initiation; WIDECUE=Wider
context; LOCOCUE=Immediate context; DICOMPR=Dictionary strategies
for comprehension; DICETXN=Extended dictionary strategies; DICLOOK=
Looking-up strategies; NOTEMNG=Meaning-oriented note-taking strate-
gies; NOTEUSE=Usage-oriented note-taking strategies; VOCLIST=Use
word lists; ORALREP=Oral repetition; VISUREP=Visual repetition;
ASSOCIA=Associationelaboration; VISUCOD=Visual encoding; AUDICOD=
Auditory encoding; WDFORM=Use word-structure; SEMANET=Semantic
encoding; CONTEXT=Contextual encoding; ACTIVAT=Activation strate-
gies; VOCSIZE=Vocabulary size; CETBAND2=College English Test.
Figure 1. Readers.
662 Language Learning Vol. 46, No. 4

values shown in Figures 1, 2, and 3 are expressed in terms of

standard deviations. We accepted a five-cluster model because at
this level we discovered significant differences among the clusters
on all variables but PRESCORE in an analysis of variance.
Readers. The best group of students, Cluster 2 (n=3, 0.6%),
obtained vocabulary size scores 2.2 standard deviations above the
mean and CETBAND2 scores 1.5 standard deviations above the
average. They strongly believed that vocabulary should be picked
up through natural exposure (z=0.40) and careful studying (z=1.18),
but not memorization (z=-1.35). Their strong self-initiation
(z=0.56) mostly reflected in their positive response to guessing
and contextual encoding of new vocabulary. In addition, they paid
some attention to word-structure (e.g., affixation). Given the
difficult EFL environment in China, this group represents only a
tiny minority of Chinese who learn EFL and its vocabulary
primarily through reading, guessing, and contextual encoding,
while at the same time sparing some attention to word forms. We
hence labeled this group Readers. The small number of Readers
in our sample population renders any interpretation speculative.
However, unrepresentative as it is, this cluster shows the possi-
bility of acquiring vocabulary through reading in a difficult EFL
context. Moreover, it demonstrates the benefits of doing so. This
group, mostly acquisition-oriented in terms of what they claimed
they did, were 1.18 standard deviations above the mean in
believing in the effortful study and deliberate use of new words.
This suggests not so much a conflict between the two seemingly
contrasting approaches to learning, but rather an integration of
the two, in that these learners might have regarded reading as the
best way of using new words. Furthermore, it also shows how
much Chinese culture values effort, a fact backed up by the figures
(M=5.74, SD=.62) for the LEARN variable in Table 2. Although
it is theoretically plausible for students to read to learn, and
although teachers should encourage students to read as much as
they can, finding enough suitable English materials to read in
China is not easy; teachers should make students aware of other
means to success.
 Gu and Johnson 663

PRESCORE -0.0 0:15-

MEMORIZ -0.16
ACQUIRE 0.04 0.49
LEARN - efr. 9
ATTENITI-96 1.27
SELFINI -1.30 .98
WIDECUE 2.12 .09
LOCOCUE G3 1.15
DICOMPR
DICETXN -1.82 .03
DICLOOK 1.90 1.15
NoTEmNG 1.73 .06
NOTEUSE .46 1.08
VOCLIST 1.18 0.92
ORALREP .09 0.66
VISUREP
ASSOCIA 1.26
IMAGERY 0.95
Pa VISUCOD -1.58 0.60
AUDICOD -1.33 0.82
WDFORM_L78 1.10
SEMANET 1.50 1.03
CONTEXT -1.53 1.03
ACTIVAT -1.67 1.28
TIME 0.3E
) 0.47
VOCSIZE
CETBAND2 -0.93,-- -0.410.94 1.19
ASU
- PSU
-2.5 -2.0 -1.5 -1.0 -0.5 0 0.5 1.0 1.5 2.0 2.5
z Scores
PSU=Passive Strategy Users; ASU+Active Strategy Users; PRESCORE=College
entrance exam scores; MEMORIZ=Memorize words; ACQUIRE=Acquire words in
context; LEARN=Study and put words to use; ATTEND=Selective attention;
SELFINI=Self-initiation; WIDECUE=Wider context; LOCOCUE=Immediate con-
text; DICOMPR=Dictionary strategies for comprehension; DICETXN=Extended dic-
tionary strategies; DICLOOK=Looking-up strategies; NOTEMNG=Meaning-oriented
note-taking strategies; NOTEUSE=Usage-oriented note-taking strategies;
VOCLIST=Use word lists; ORALREP=Oral repetition; VISUREP=Visual repetition;
ASSOCIA=Associationielaboration;VISUCOD=Visual encoding; AUDICOD=Auditory
encoding; WDFORM=Use word-structure; SEMANET=Semantic encoding; CON-
TEXT=Contextual encoding; ACTIVAT=Acti-vation strategies; VOCSIZE= Vocabu-
lary size; CETBAND2=College English Test.
Figure 2. Passive Strategy Users and Active Strategy Users.
664 Language Learning Vol. 46, No. 4

Active Strategy Users. Cluster 3 (n=48, 9.9%), the second

best group in terms of vocabulary size and general proficiency,
achieved success through a very different approach. Although
they also believed in natural acquisition as well as careful study
and use of new words, they did not disagree as drastically as the
Readers did with the memorization of words. They spent a lot of
extracurricular time on English learning (z=0.47) and employed
almost every strategy 0.29 to 1.28 standard deviations more than
average, except for visual repetition (z=-0.1). One could readily
interpret this as proving that using more strategies is better than
using fewer. However, the success of this group rests more on
their self-initiation and high flexibility in strategy use. They
probably more willingly tried new strategies and therefore had a
wider spectrum of strategies in stock. We can best characterize
these learners as hard-working and motivated. They guessed
more, used more dictionary strategies to learn vocabulary, took
more notes, did more memorization, and activated more newly
learned words than their peers (cf. Sanaoui, 1995). What paid off
eventually was the time and effort they exerted. In other words,
these students might have succeeded despite using certain strat-
egies. That a majority of the good learners in this study used
certain strategies (e.g., visual repetition) shown to be least condu-
cive to success in the previous section does not contradict the
results there.
Passive Strategy Users. Cluster 5 (n=9, 1.9%) shows an
almost mirror image of the Active Strategy Users. This group,
whose vocabulary size and CETBAND2 scores recorded 0.41 and
0.93 standard deviations below average, respectively, strongly
believed in memorization (z=1.18). They also believed in the
effortful studying of new words. Nevertheless, with the sole
exception of Visual Repetition (z=-0.03), their strategy use was
0.35 to 2.13 standard deviations below average. They spent only
as much time on English learning, albeit not much in absolute
terms (z=-0.35), as did the Readers, who reported the least
extracurricular time on learning English among the five groups
(z=-0.40). For whatever reasons, this group, about 2% of the
 Gu and Johnson 665

PRESCORE
MEMORIZ -0.16 0.02
ACQUIRE /L0.08 I 10
LEARN -0.11 0.14
ATTEND -0. 0.04
SELFINI -0.1 0.00
WIDECUE -0.14 0.27
LOCOCUE -0.17 0.24
DICOMPR -0.14 0.12
Q) DICETXN / -0.02
DICLOOK 0.02
NOTEMNG .00 Encoders
NOTEUSE -0.15
cf) Non-Encoders
ro VOCLIST -0.15
ORALREP -0.12
VISUREP
-0.47
ASSOCIA 49
Q)
Ra IMAGERY_0.46 0.54
VISUCOD 51
AUDICOD 0.56
WDFORM -0.26
0.29
SEMANET -0.40< 0.51
CONTEXT 0.36
ACTIVAT -0.36- 0.37
TIME -0.1 -0.03
VOC SIZE -0.17 .11
CETBAND2 -0.13 0.08
-0.5 -0.25 0 0.25 0.50
z Scores
PRESCORE=College entrance exam scores; MEMORIZ=Memorize words;
ACQUIRE=Acquire words in context; LEARN=Study and put words to use;
ATTEND=Selective attention; SELFINI=Self-initiation; WIDECUE=Wider
context; LOCOCUE=Immediate context; DICOMPR=Dictionary strategies
for comprehension; DICETXN=Extended dictionary strategies; DICLOOK=
Looking-up strategies; NOTEMNG=Meaning-oriented note-taking strate-
gies; NOTEUSE=Us age-oriented note-taking strategies; VOCLIST=Use word
lists; ORALREP=Oral repetition; VISUREP=Visual repetition; ASSOCIA=As-
sociationielaboration; VISUCOD=Visual encoding; AUDICOD=Auditory en-
coding; WDFORM=Use word-structure; SEMANET=Semantic encoding; CON-
TEXT=Contextual encoding; ACTIVAT=Activation strategies; VOCSIZE=
Vocabulary size; CETBAND2=College English Test.
Figure 3. Encoders and Non-Encoders.
666 Language Learning Vol. 46, No. 4

participants, did not learn much after spending 7 years learning

English as a school subject. Apparently, they had not developed
the basic idea of what a language is and how it should be learned,
and relied most heavily on visual repetition—the kind of strategy
they might have used in primary school to memorize Chinese
characters. That the z score of the strategy they used most, Visual
Repetition, barely reached the mean indicates their lack of moti-
vation in learning EFL. In other words, unlike the poor learner
reported in Gu (1994), they could attribute their failure not to their
inappropriate beliefs (though they strongly believed in memoriza-
tion) nor to their ineffective strategies (visual repetition), but to
their lack of effort. Regrettably, the anonymity of data collection
procedures prevented locating these learners, or for that matter,
the good learners as well: exactly where quantitative methods fall
short, qualitative in-depth alternatives start to reveal.
Encoders and Non-Encoders. Clusters 1 (n=157, 32.3%) and
4 (n=269, 55.3%) clustered around the mean and were almost
indistinguishable from each other except for their use of encoding
strategies, hence the labels. Cluster 1, the Encoders, used encod-
ing strategies 0.29 to 0.56 standard deviations above average;
both their vocabulary size and proficiency scores were below
average and lower than those of the Non-Encoders, who used the
same encoding strategies 0.26 to 0.47 standard deviations lower
than the mean.
Overall, then, in terms of VOCSIZE and CETBAND2 scores,
the most successful learners were the Readers, followed in order
by Active Strategy Users, Non-Encoders, Encoders, and Passive
Strategy Users. When we examined college entrance English
scores (PRESCORE) in the light of these different strategy pro-
files, the five clusters did not vary significantly (F=0.379, p=.824),
suggesting that differences in vocabulary size and general En-
glish proficiency were largely a function of the beliefs and strate-
gies studied here and not of the students' initial proficiency level
in English on entering the university.
The Readers and the Active Strategy Users, the two types of
high achievers, constituted less than 11 % of the whole population;
 Gu and Johnson 667

the Passive Strategy Users, the "underachievers," represented

less than 2%. The great majority (more than 87%) of the partici-
pants here had similar belief and strategy profiles (the Encoders
and the Non-Encoders). This suggests that, at least for the
population of this one university, (a) most Chinese non-English
majors hold similar beliefs and use similar strategies in vocabu-
lary learning; and (b) although using a bit of every strategy may
have saved them from total failure, it did not help most learners
perform as well as their high-achieving classmates.
The strategy profiles of these clusters of learners more
probably result from social and educational than from cognitive
factors, though differences may stem from cognitive style too. The
successful Readers, for example, represent the exceptional few
who not only believe in actively seeking opportunities to use
English outside and beyond the classroom environment but are
successful in doing so. Perhaps these students are more resource-
ful that others, or socially advantaged in ways that make such
opportunities more possible for them. The Active Strategy Users
represent more the traditional, hard-working Chinese learners
who believe in effort regardless of approach. Passive Strategy
Users may see little value in learning EFL; it is a requirement in
their courses but their main interest lies elsewhere; or perhaps
they are generally less motivated toward their studies. The
Encoders seem to value vocabulary learning and to find attractive
the shortcuts to vocabulary acquisition offered by mnemonics. A
number of books available in China recommend such approaches
to learning and are popular among some language learners. The
Non-Encoders typify learners who do not believe in quick fixes
and, like the Passive Strategy Users, show little self-initiation in
(or perhaps motivation for) learning English. These explanations
are speculative; more work is required to determine not only what
makes one approach more effective than another but also what
contributes to more or less effective vocabulary learning within
the different approaches.
668 Language Learning Vol. 46, No. 4

Summary and Implications

This study profiled the beliefs and strategies of adult Chinese

learners for learning EFL vocabulary. Contrary to popular beliefs
about Asian learners, the participants generally did not dwell on
memorization, and reported using more meaning-oriented strat-
egies than rote strategies in learning vocabulary. Second, we
examined a wide range of vocabulary learning beliefs and strate-
gies in relation to both vocabulary size and general English
proficiency. Self-Initiation and Selective Attention, the 2
metacognitive strategies studied here, emerged as positive pre-
dictors of general proficiency. At the cognitive level, contextual
guessing, skillful use of dictionaries for learning purposes (as
opposed to looking up for comprehension only), note-taking, pay-
ing attention to word formation, contextual encoding, and inten-
tional activation of new words all positively correlated with the 2
dependent variables. In addition, oral repetition positively corre-
lated with general proficiency, a finding in line with previous
research (Hill, 1994; Kelly, 1992; Seibert, 1927). On the other
hand, visual repetition of new words was the strongest negative
predictor of both vocabulary size and general proficiency. Strat-
egies aiming at vocabulary retention only correlated more with
vocabulary size than with English proficiency in general. Finally,
a cluster analysis identified five approaches to learning. One
small group excelled primarily through extensive reading; an-
other group by actively employing a wide range of strategies. This
suggests that both direct and indirect approaches to vocabulary
learning can be useful, and that both can, with disinterested and
well-informed advice, foster "situated cognition" (Brown, Collins,
& Duguid, 1989). The weakest group strongly believed in memo-
rization and placed the greatest emphasis on visual repetition of
word lists.
This was an exploratory study: Therefore, correlational re-
sults suggest only strong or weak, positive or negative links
between the beliefs and strategies studied and the 2 dependent
variables. Furthermore, as with any similar studies, one can ask
 Gu and Johnson 669

how much self-reports reflect reality. In answer, the participants

had not been exposed to similar studies before; therefore, they had
not received much influence from existing theoretical and empiri-
cal work on learning strategies. In addition, the anonymity of the
questionnaire considerably reduced the possibility of false re-
ports.
With these limitations in mind, the study obtained enough
evidence to issue the following warning to EFL learners: Vocabu-
lary knowledge, to be of real use, must become integrated into
discourse. Therefore, a large part of EFL vocabulary learning
necessarily involves skill learning. Pure retention of
decontextualized words without a threshold level of L2 skill offers
limited value no matter what "deep" processing strategies learn-
ers use to achieve this purpose. Learners should use memory
strategies that aim for retaining word-meaning pairs with cau-
tion, if at all, and should complement them with other fully
contextualized strategies. Taking care of words individually will
not necessarily take care of other aspects of the language, espe-
cially for foreign language learners in input-poor environments.
Revised version accepted 18 June 1996

Notes

'We distinguish here between memory strategies and vocabulary learning

strategies. The former refers to most vocabulary strategies, which aim only
to commit form-meaning pairs to memory. We contend that these constitute
but one type of vocabulary learning strategy and that committing words to
memory is far from an end in itself in foreign language learning. The latter
refers to a wide spectrum of strategies used as part of an on-going process of
vocabulary learning (see Ahmed, 1989; Gu, 1994; Sanaoui, 1995; Schmitt &
Schmitt, 1995). From the initial handling of a new word (guessing, postpon-
ing, or abandoning), to finding out the meanings, usages, and examples of
the word and taking down notes about it, if necessary, to committing the
word to memory, all the way to putting the word to use, learners differ in
almost every step they take in learning vocabulary. These strategies other
than memory strategies warrant much more attention.
2 We excluded a class of "absolute beginners" in this grade, who had either

learned other foreign languages or no foreign language at all in their

secondary schools and who only started to learn English after their enroll-
ment at BNU.
670 Language Learning Vol. 46, No. 4

3We administered the questionnaire in Chinese. This appendix shows the

English version only. In addition, questionnaire items in the appendix are
organized under category headings. These items were randomly ordered
during administration.
4 College English Test (CET) Band 4 is a standardized English proficiency
test taken each year by thousands of university non-English majors across
China at the end of their second year. Passing CET Band 4 is the minimum
requirement for students in key universities. The majority of students stop
the formal classroom learning of English after passing CET Band 4, whereas
some others continue and finish Band 6 (The State Education Commission,
PRC, 1986).
5The distinction between memorization and learning is often less clear-cut
for Chinese than for Western students. The Chinese tend to see memoriza-
tion as a part, though not the whole, of the process of learning. In this study,
we distinguish between learning and rote-memorization; the students in
this study generally rejected the latter, often attributed to Chinese learners
as a stereotype.

References

Abraham, R. G., & Vann, R. J. (1987). Strategies of two language learners:

A case study. In A. Wenden & J. Rubin (Eds.), Learner strategies in
language learning (pp. 85-102). Englewood Cliffs, NJ: Prentice-Hall.
Ahmed, M. 0. (1989). Vocabulary learning strategies. In P. Meara (Ed.),
British studies in applied linguistics: Vol. 4. Beyond words (pp. 3-14).
London: British Association of Applied Linguistics/Centre for Language
Teaching.
Anderson, J. P., & Jordan, A. M. (1928). Learning and retention of Latin
words and phrases. Journal of Educational Psychology, 19, 485-496.
Atkinson, R. C. (1972). Optimizing the learning of a second-language
vocabulary. Journal of Experimental Psychology, 96, 124-129.
Atkinson, R. C. (1975). Mnemotechnics in second-language learning. Ameri-
can Psychologist, 30, 821-828.
Atkinson, R. C., & Raugh, M. R. (1975). An application of the mnemonic
keyword method to the acquisition of a Russian vocabulary. Journal of
Experimental Psychology, 104, 126-133.
Bird, N. (1994). Investigating lexis beyond the most frequent words: Part 2.
In N. Bird, P. Falvey, A. B. M. Tsui, D. M. Allison, A. McNeill, & R. J. Webb
(Eds.), Language and learning (pp. 274-287). Hong Kong: Hong Kong
Institute of Language in Education, Education Department.
Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the
culture of learning. Educational Researcher, 18, 32-42.
Carter, R. (1987). Vocabulary: Applied linguistic perspectives. London: Allen
& Unwin.
 Gu and Johnson 671

Carter, R., & McCarthy, M. (Eds.). (1988). Vocabulary and language teach-
ing. London: Longman.
Channell, J. M. (1981). Applying semantic theory to vocabulary teaching.
ELT Journal, 35, 115-122.
Channell, J. M. (1988). Psycholinguistic considerations in the study of L2
vocabulary acquisition. In R. Carter & M. McCarthy (Eds.), Vocabulary
and language teaching (pp. 83-96). London: Longman.
Chern, C.-L. (1993). Chinese students' word-solving strategies in reading in
English. In T. Huckin, M. Haynes, & J. Coady (Eds.), Second language
reading and vocabulary learning (pp. 67-85). Norwood, NJ: Ablex.
Cohen, A. D. (1987). The use of verbal and imagery mnemonics in second-
language vocabulary learning. Studies in Second Language Acquisition,
9, 43-61.
Crothers, E., & Suppes, P. C. (1967). Experiments in second-language
learning. New York: Academic Press.
Crow, J. T., & Quigley, J. R. (1985). A semantic field approach to passive
vocabulary acquisition for reading comprehension. TESOL Quarterly,
19, 497-513.
Field, M. L. (1984). A psycholinguistic model of the Chinese ESL reader. In
P. Larson, E. L. Judd, & D. S. Messerschmitt (Eds.), On TESOL '84: A
brave new world for TESOL (pp. 171-182). Washington, DC: TESOL.
Gershman, S. J. (1970). Foreign language vocabulary learning under seven
conditions. Unpublished doctoral dissertation, Columbia University,
New York.
Goulden, R., Nation, I. S. P., & Read, J. A. S. (1990). How large can a
receptive vocabulary be? Applied Linguistics, 11, 341-363.
Gu, Y. (1994, March). Vocabulary learning strategies of good and poor
Chinese EFL learners. Paper presented at TESOL '94, Baltimore. (ERIC
Document Reproduction Service No. ED 370 411)
Haynes, M. (1990). Examining the impact of Ll literacy on reading success
in a second writing system (WS2). In H. Burmeister & P. L. Rounds (Eds.),
Variability in second language acquisition: Proceedings of the tenth
meeting of the second language research forum (pp. 381-398). Eugene,
OR: University of Oregon, Department of Linguistics.
Higa, M. (1963). Interference effects of intralist word relationships in verbal
learning. Journal of Verbal Learning and Verbal Behavior, 2, 170-175.
Hill, M. (1994). A word in your ear: To what extent does hearing a new word
help learners to remember it? In N. Bird, P. Falvey, A. B. M. Tsui, D. M.
Allison, A. McNeill, & R. J. Webb (Eds.), Language and learning (pp. 447-
462). Hong Kong: Hong Kong: Hong Kong Institute of Language in
Education, Education Department.
Horwitz, E. K. (1987). Surveying student beliefs about language learning. In
672 Language Learning Vol. 46, No. 4

A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp.

119-129). Englewood Cliffs, NJ: Prentice-Hall.
Huckin, T., Haynes, M., & Coady, J. (1993). Second language reading and
vocabulary learning. Norwood, NJ: Ablex.
Izawa, H. (1993). The English vocabulary of 21 Japanese adults on a high
proficiency level. JALT Journal, 15, 63-78.
Judd, E. L. (1978). Vocabulary teaching and TESOL: A need for re-evalua-
tion of existing assumptions. TESOL Quarterly, 12, 71-76.
Kelly, P. (1992). Does the ear assist the eye in the long-term retention of
lexis? International Review of Applied Linguistics in Language Teaching,
30, 137-145.
Kouraogo, P. (1993). Language learning strategies in input-poor environ-
ments. System, 21, 165-173.
Krashen, S. D. (1989). We acquire vocabulary and spelling by reading:
Additional evidence for the Input Hypothesis. Modern Language Jour-
nal, 73,440-464.
Lado, R., Baldwin, B., & Lobo, F. (1967). Massive vocabulary expansion in a
foreign language beyond the basic course: The effects of stimuli, timing
and order of presentation (Project No. 5-1095). Washington, DC: U.S.
Department of Health, Education, and Welfare.
Laufer, B. (1986). Possible changes in attitude towards vocabulary acquisi-
tion research. International Review of Applied Linguistics in Language
Teaching, 24, 73-79.
Lord, R. (1974). Learning vocabulary. International Review of Applied
Linguistics in Language Teaching, 12, 239-247.
McCarthy, M. J. (1984). A new look at vocabulary in EFL. Applied Linguis-
tics, 5, 12-22.
Meara, P. (1980). Vocabulary acquisition: A neglected aspect of language
learning. Language Teaching and Linguistics Abstracts, 13, 221-246.
Nation, I. S. P. (1982). Beginning to learn foreign vocabulary: A review of the
research. RELC Journal, 13, 14-36.
Nation, I. S. P. (1990). Teaching and learning vocabulary. Boston: Heinle &
Heinle.
Nation, I. S. P. (1994). [Review of books Working with words, Teaching and
learning vocabulary, Vocabulary in action, and Vocabulary]. System, 22,
283-287.
O'Malley, J. M., Chamot, A. U., Stewner-Manzanares, G., Kupper, L., &
Russo, R. P. (1985). Learning strategy applications with students of
English as a second language. TESOL Quarterly, 19, 557-584.
Oxford, R. L. (1990). Language learning strategies: What every teacher
should know. New York: Newbury House.
Oxford, R. L., & Scarcella, R. C. (1994). Second language vocabulary
 Gu and Johnson 673

learning among adults: State of the art in vocabulary instruction. System,

22, 231-243.
Paivio, A., & Desrochers, A. (1981). Mnemonic techniques in second-lan-
guage learning. Journal of Educational Psychology, 73,780-795.
Parry, K. (1991). Building a vocabulary through academic reading. TESOL
Quarterly, 25, 629-653.
Politzer, R. L., & McGroarty, M. (1985). An exploratory study of learning
behaviors and their relationship to gains in linguistic and communicative
competence. TESOL Quarterly, 19, 103-123.
Richards, J. C. (1976). The role of vocabulary teaching. TESOL Quarterly,
10, 77-89.
Royer, J. M. (1973). Memory effects for test-like-events during acquisition of
foreign language vocabulary. Psychological Reports, 32, 195-198.
Sanaoui, R. (1995). Adult learners' approaches to learning vocabulary in
second languages. Modern Language Journal, 79, 15-28.
Saragi, T., Nation, I. S. P., & Meister, G. F. (1978). Vocabulary learning and
reading. System, 6, 72-78.
Schmitt, N., & Schmitt, D. (1995). Vocabulary notebooks: Theoretical under-
pinnings and practical suggestions. ELT Journal, 49, 133-143.
Seibert, L. C. (1927). An experiment in learning French vocabulary. Journal
of Educational Psychology, 18, 294-309.
The State Education Commission, PRC. (1986). Daxue yingyu jiaoxue
dagang [College English teaching syllabus for arts and science students
in tertiary institutions]. Shanghai, China: Shanghai Foreign Languages
Education Press.
Tinkham, T. N. (1993). The effect of semantic clustering on the learning of
second language vocabulary. System, 21, 371-380.

Appendix

Vocabulary Learning Questionnaire (VLQ Version 3)

Dimensions, Variables, and Items

Beliefs About Vocabulary Learning: (17 items)

1. Words should be memorized (8)

Once the English equivalents of all Chinese words have been
remembered, English is learned.
The best way to remember words is to memorize word lists or
dictionaries.
Remembering the meanings of a word is an end in itself.
674 Language Learning Vol. 46, No. 4

English words have fixed meanings.

It is only necessary to remember one dictionary definition.
A good memory is all you need to learn a foreign language well.
Repetition is the best way to remember words.
You can only acquire a large vocabulary by memory of individual
words.
2. Words should be acquired in context: bottom-up (4)
The meanings of a considerable amount of words can be picked up
through reading.
One can expand his vocabulary simply through reading a lot.
Guessing words in context is one of the best ways to learn
vocabulary.
When you come across a word several times in different contexts,
you will know what it means.
3. Words should be studied and put to use: top-down (5)
One should pay attention to set phrases and collocations that go
with a word.
Words studied should be put to use before they are finally learned.
Using the language (listening, speaking, reading, and writing) is
more important than memorizing words.
The least a learner should know about a word is its form, its
meaning, and its basic usage.
Words are learned after you use them.

Metacognitive Regulation: (12)

1. Selective attention (7)

I know when a new word or phrase is essential for adequate
comprehension of a passage.
I know which words are important for me to learn.
I have a sense of which word I can guess and which word I can't.
I look up words that I'm interested in.
When I meet a new word or phrase, I have a clear sense of whether
I need to remember it.
I know what cues I should use in guessing the meaning of a
particular word.
I make a note of words that seem important to me.
 Gu and Johnson 675

2. Self-initiation (5)
Besides textbooks, I look for other readings that fall under my
interest.
I wouldn't learn what my English teacher doesn't tell us to learn.
(Reversed value)
I only focus on things that are directly related to examinations.
(Reversed value)
I wouldn't care much about vocabulary items that my teacher does
not explain in class. (Reversed value)
I use various means to make clear vocabulary items that I am not
quite clear of.

Guessing Strategies (12 items)

1. Using background knowledge/wider context (7)

I use alternative cues and try again if I fail to guess the meaning
of a word.
I make use of the logical development in the context (e.g., cause
and effect) when guessing the meaning of a word.
I make use of my common sense and knowledge of the world when
guessing the meaning of a word.
I check my guessed meaning against the wider context to see if it
fits in.
I make use of my knowledge of the topic when guessing the
meaning of a word.
I look for other words or expressions in the passage that support
my guess about the meaning of a new word.
I look for any definitions or paraphrases in the passage that
support my guess about the meaning of a word.
2. Using linguistic cues/immediate context (5)
I make use of the grammatical structure of a sentence when
guessing the meaning of a new word.
I look for any examples provided in the context when guessing the
meaning of a new word.
I make use of the part of speech of a new word when guessing its
meaning.
I check my guessed meaning against the immediate context to see
if it fits in.
I analyze the word structure (prefix, root, and suffix) when
guessing the meaning of a word.
676 Language Learning Vol. 46, No. 4

Dictionary Strategies (17 items)

1. Dictionary strategies for comprehension (4)

When I see an unfamiliar word again and again, I look it up.
When I want to confirm my guess about a word, I look it up.
When not knowing a word prevents me from understanding a
whole sentence or even a whole paragraph, I look it up.
I look up words that are crucial to the understanding of the
sentence or paragraph in which it appears.
2. Extended dictionary strategies (8)
I pay attention to the examples of use when I look up a word in a
dictionary.
I look for phrases or set expressions that go with the word I look
up.
I consult a dictionary to find out about the subtle differences in the
meanings of English words.
When I want to know more about a word that I already have some
knowledge of, I look it up.
When I don't know the usage of a word I already have some
knowledge of, I look it up.
I make a note when I want to help myself distinguish between the
meanings of two or more words.
When looking up a word in the dictionary, I read sample sentences
illustrating various meanings of the word.
When I get interested in another new word in the definitions of the
word I look up, I look up this word as well.
3. Looking-up strategies (5)
If the new word is inflected, I remove the inflections to recover the
form to look up (e.g., for created, look for create).
If the new word I try to look up seems to have a prefix or suffix, I
will try the entry for the stem.
If the unknown appears to be an irregularly inflected form or a
spelling variant, I will scan nearby entries.
If there are multiple senses or homographic entries, I use various
information (e.g., part of speech, pronunciation, style, colloca-
tion, meaning, etc.) to reduce them by elimination.
I try to integrate dictionary definitions into the context where the
unknown was met and arrive at a contextual meaning by adjust-
ing for complementation and collocation, part of speech, and
breadth of meaning.
 Gu and Johnson 677

Note-Taking Strategies (9 items)

1. Meaning-oriented note-taking strategies (5)

I make a note of the meaning of a new word when I think the word
I'm looking up is commonly used.
I make a note when I think the word I'm looking up is relevant to
my personal interest.
I put synonyms or antonyms together in my notebook.
I write down the English synonym(s) or explanations of the word
I look up.
I write down both the Chinese equivalent and the English syn-
onyms of the word I look up.
2. Usage-oriented note-taking strategies (4)
I make a note when I see a useful expression or phrase.
I take down the collocations of the word I look up.
I take down grammatical information about a word when I look it
up.
I note down examples showing the usages of the word I look up.

Memory Strategies: Rehearsal: (12 items)

1. Using word lists (6)

I make vocabulary lists of new words that I meet.
I write the new words on one side of a card and their explanations
on the other side.
I keep the vocabulary lists of new words that I make.
I go through my vocabulary list several times until I am sure that
I do not have any words on that list that I still don't understand.
I make vocabulary cards and take them with me wherever I go.
I make regular and structured reviews of new words I have
memorized.
2. Oral repetition (3)
When I try to remember a word, I repeat it aloud to myself.
Repeating the sound of a new word to myself would be enough for
me to remember the word.
When I try to remember a word, I repeat its pronunciation in my
mind.
3. Visual repetition (3)
When I try to remember a word, I write it repeatedly.
I memorize the spelling of a word letter by letter.
678 Language Learning Vol. 46, No. 4

I write both the new words and their Chinese equivalents repeat-
edly in order to remember them.

Memory Strategies: Encoding (24 items)

I. Association/elaboration (4)
I remember a group of new words that share a similar part in
spelling.
I associate a group of new words that share a similar part in
spelling with a known word that looks or sounds similar to the
shared part.
I create a sentence in Chinese when I link a new word to a known
word.
I attach physical sensations to certain words (e.g., stinking) when
I try to remember them.
2. Imagery (4)
I act out a word in order to remember it better.
I create a mental image of the new word to help me remember it.
I associate one or more letters in a word with the word meaning
to help me remember it (look has two "eyes" in the middle).
I create mental images of association when I link a new word to a
known word.
3. Visual encoding (3)
I visualize the new word to help me remember it.
I associate a new word to a known English word that looks similar.
I remember the spelling of a word by breaking it into several visual
parts.
4. Auditory encoding (3)
I remember together words that sound similar.
I remember together words that are spelled similarly.
I associate a new word with a known English word that sounds
similar.
5. Word-structure (3)
I analyze words in terms of prefixes, stems, and suffixes.
I deliberately study word-formation rules in order to remember
more words.
I memorize the commonly used stems and prefixes.
 Gu and Johnson 679

6. Semantic encoding (3)

I try to create semantic networks in my mind and remember words
in meaningful groups.
When I meet a new word, I search in my memory and see if I have
any synonyms and antonyms in my vocabulary stock.
I group words into categories (e.g., animals, utensils, vegetables,
etc.).
7. Contextual encoding (4)
When I try to remember a word, I remember the sentence in which
the word is used.
I deliberately read books in my areas of interest so that I can find
out and remember the special terminology that I know in Chi-
nese.
I remember the new word together with the context where the new
word occurs.
I learn words better when I put them in contexts (e.g., phrases,
sentences, etc.).

Activation Strategies (5 items)

I try to read as much as possible so that I can make use of the words
I tried to remember.
I make up my own sentences using the words I just learned.
I try to use the newly learned words as much as possible in speech
and writing.
I try to use newly learned words in real situations.
I try to use newly learned words in imaginary situations in my
mind.