VIETNAM NATIONAL UNIVERSITY, HANOI
UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES
FACULTY OF POST-GRADUATE STUDIES
NGUYỄN HÀ SÂM

AN ACTION RESEARCH ON IMPROVING 10
th
GRADE STUDENTS’
READING COMPREHENSION THROUGH THE TEACHING OF
LEXICAL INFERENCE STRATEGY
AT HIGH SCHOOL FOR GIFTED STUDENTS,
HANOI NATIONAL UNIVERSITY OF EDUCATION

(NGHIÊN CỨU HÀNH ĐỘNG VỀ VIỆC NÂNG CAO KHẢ NĂNG
ĐỌC HIỂU CỦA HỌC SINH THÔNG QUA VIỆC GIẢNG DẠY
CHIẾN LƯỢC ĐOÁN NGHĨA CỦA TỪ VỰNG
TẠI TRƯỜNG TRUNG HỌC PHỔ THÔNG CHUYÊN,
ĐẠI HỌC SƯ PHẠM HÀ NỘI)

M.A. MINOR PROGRAMME THESIS

Field: English Teaching Methodology
Code: 6014.0111 Hanoi, 2014
i
DECLARATION

I hereby certify that the minor thesis entitled “An Action Research on
Improving 10
th
Grade Students’ Reading Comprehension through the Teaching of
Lexical Inference Strategy at High School for Gifted Students, Hanoi National
University of Education” is the result of my own work in partial fulfillment of the
requirements for the degree of Master of Arts in English Teaching Methodology at
Faculty of Post - Graduate Studies, Hanoi University of Languages and International
Studies, Vietnam National University. The research has not been submitted to any other
university or institution wholly and partially. Hanoi, 2014

Nguyen Ha Sam iii
ABSTRACT
This study was intended to investigate the impact of lexical inference strategy on
10
th
grade students‟ reading comprehension ability. More specifically, the researcher
strived to (1) examine whether students‟ lexical knowledge had any correlation to their
reading ability, (2) whether the teaching of lexical inferencing had a positive effect to
enhance tenth grade students‟ reading comprehension, (3) investigate the students‟
attitudes towards the instruction, and (4) propose some recommendations for instructing
this strategy to other teachers.
To achieve those abovementioned aims, an action research was conducted with
the participation of 46 tenth grade students from a class at High School for Gifted
Students, Hanoi National University of Education. The teaching program aimed at
instructing the students with lexical inferencing to help them enhance their reading
comprehension ability. The study included: (1) pretest and a questionnaire (2) the
instruction of lexical inferencing, (3) posttest and a questionnaire, (4) teacher‟s notes and
students‟ learning logs. Later, (5) interviews were included to support the findings.
The main findings of the research were as follows: (1) Lexical competence served

Table 15: Results of Question 4 in the Questionnaire After the Posttest 34
Table 16: Results of Question 5 in the Questionnaire After the Posttest 34
Abbreviations
T: Teacher
S1: Subject 1 in the interview
S2: Subject 2 in the interview
S3: Subject 3 in the interview
HNUE: Hanoi National University of Education
L2: second language v

TABLE OF CONTENT

DECLARATION……………………………………………………………………… i
ACKNOWLEDGEMENT………………………………………………………… ii
ABSTRACT………………………………………………………………………… iii
LIST OF ABBREVIATIONS, TABLES AND FIGURES………………………………iv
TABLE OF CONTENT…………………………………………………………… … V
PART A: INTRODUCTION 1
1.1. Rationale for the Study 1
1.2. Aims of the Research 2
1.3. Objectives of the Research 2
1.4. Research Questions 3
1.5. Scope of the Research 3
1.6. Significance of the Research 3
1.7. Structural Organization of the Thesis 3
PART B: DEVELOPMENT 5
Chapter 1: Literature Review 5

3.1. Preliminary Investigation 28
3.1.1. Findings 28
3.1.2. Discussion 29
3.2. Evaluation 30
3.2.1. Research Question 1 30
3.2.2. Research Question 2 31
3.2.2.1. Results from Pretest and Posttest 31
3.2.2.2. Discussion 31
3.2.3. Research Question 3 32
3.2.3.1. Results from Questionnaire After the Posttest 33
3.2.3.2. Discussion 34
3.2.3.3. Results of the Interview 35
3.2.3.4. Discussion 37
3.3. Summary 37
PART C: CONCLUSION 39
3.1. Recapitulation 39
3.2. Conclusions 39
3.3. Limitations of the Research 40
3.4. Suggestions for Future Research 41
REFERENCES 43
APPENDICES……………………………………………………………………………I

1


2

strategies. Teachers normally stress on the production of reading comprehension rather
than the reading process. Therefore, a better way of teaching and learning reading
skills, as many researchers have suggested, is to equip students with the strategy of
guessing word meaning from context and to help them become independent readers.
Particularly, 10
th
grade students may have not been taught any reading strategy before.
They have acquired a range of vocabulary and grammar after five to seven years of
learning English. Now it is urged that they be taught learning strategies to become
better and more efficient learners.
With a view to gaining some insight into reading strategies and reading strategy
instruction, I chose to study how to improve students‟ reading comprehension through
the teaching of lexical inference strategy. The rationale for my focus is that lexical
inference strategy plays a critical role in successful reading (Sullivan, 1978). Although
the importance of lexical inferencing has been highlighted in a number of empirical
studies, much remains to be learned about it.
1.2. Aims of the Research
The study aimed at improving reading comprehension ability of 10
th
grade students at
High School for Gifted Students (HNUE) through lexical inference strategy instruction,
and giving recommendations on how to teach students to use lexical inference strategies
effectively to enhance their reading abilities.
1.3. Objectives of the Research
The specific objectives of the research were as follows:
 Examine whether there is a correlation between lexical knowledge and reading
comprehension.
 Apply the teaching of lexical inference strategy to improve reading

1.7. Structural Organization of the Thesis
The thesis is organized as follows:
Part A - Introduction – provides an overview of the study.
Part B - Development - consists of three chapters:
In Chapter One – Literature Review, a review of relevant literature on this issue is
presented.
4

In Chapter Two – Research Methodology, a description of research methods is
discussed in detail, including a description of selection of the participants, data
collection instruments and data collection procedures.
In Chapter Three – Findings and Discussions, the results of the study are presented.
In Part C - Conclusion, the researcher provides a summary on main points raised in
the study, the major conclusion drawn from conducting this study, and a discussion of
the study‟s limitations, and suggestions for further research.

seems that the reader tries to reconstruct the writer‟s meaning by recognizing letters and
words. The top-down model is viewed as reader-driven. It emphasizes both using
learners‟ prior knowledge and constructing meaning from written text by predicting,
6

sampling, confirming, and correcting in reading comprehension. It seems that learners
with more or less reading proficiency have the ability to make predictions, and their
general knowledge becomes more important for their reading comprehension. As for
the interactive model, it stresses reader ability to switch between the top-down and
bottom-up models during reading comprehension. The reading strategy of this model is
more flexible and can be employed according to learner competence.
When the three reading models are applied to second language reading, some
problems still exist. For example, foreign language readers are not usually fully
proficient about the phonological system when reading. Further, for foreign readers, the
differences and distance in terms of culture, and the distance inherent in any target
language text can also make reading comprehension more difficult. Nevertheless,
Barnett (1989) still claimed that the three reading models have their own advantages in
terms of developing learners‟ reading ability. In addition, the goal of each reading
model is to make foreign language readers succeed in reading target texts. Since
reading is a “psychological guessing game” (Goodman, 1982), by guessing through
reading, readers can “reduce their dependence on the print and phonics of the texts”
(Barnett, 1989). However, what can readers do when they encounter unknown words?
How can readers go through reading without any interruption? As mentioned
previously, the most useful strategy for coping with vocabulary is guessing words from
context or lexical inferencing (Nation, 1990; Nassaji, 2003). Therefore, there indeed
exists a strong relationship between coping with vocabulary and fluent reading.
1.1.2. Reading Comprehension Strategies
Researchers point out that effective readers spontaneously use reading strategies in
the reading process, and the use of appropriate reading strategies may improve reading
comprehension (Oxford, 1990).

attention Transfer
Deduction
Question for
classification
Functional
planning Inferencing
Recombination
Self-talk
Selective
attention Summarizing
Translation

Self-
management
Key word
method

Table 1: Classification of Reading Strategies (O’Malley & Chamot, 1990)

Freebody and Anderson (1983) found that readers skipped unfamiliar words and even
parts of sentences containing them if they did not seem to be necessary to follow the
theme of the text or to do the task they were set. Putting a time limit on such reading
will also make ignoring some unknown words an advantage (British Council Teachers,
1980).
2. Giving the meaning quickly: In some situations, translation will be the most
effective way. Usually the teacher will give the meaning that works in the context. If
the word has useful parts, quick analysis of the word will not take much time and will
have useful effects.
3. Pre-teaching vocabulary: One way of stopping vocabulary work from interrupting
reading is to pre-teach vocabulary in the text. This can be a useful procedure but there
are several difficulties associated with it (Nation & Coady, 1988). First, the research
has not shown the gains in comprehension as a result of pre-teaching vocabulary.
Second, pre-teaching takes away the opportunity for learners to use their guessing skill.
9

Third, vocabulary teaching will be more meaningful for learners after they have met it
in the text because then they have some experience to attach the teaching to.
4. Glossing: One way of dealing unknown words in a text is to provide a short
definition somewhere near the text. Glossing helps reading by means of providing the
meanings of words that are not easily guessed by using contextual clues. Learners
should be encouraged to use glosses as a way of confirming guessing from context.
They should not look up the meaning of a word without first having a guess at its
meaning. For this reason, glosses are best situated at the end of the text or in the back of
the book.
5. Referring to a dictionary: Dictionaries are used primarily to check meaning. The
next most frequent uses are to check spelling and pronunciation (Bejoint, 1981;
MacFarquhar & Richard, 1983). In guessing words from context, reference to a
dictionary is one of the steps used to check if a guess is correct. Research on dictionary
use and comprehension has shown that allowing learners to use dictionary while

Step 2: Look at the clause or sentence containing the unknown word. In this
step, learners can ask themselves some questions such as “Who does what to whom?”
or “What does what?”. Learners can find clues in the immediate grammar. In other
words, learners use local clues to guess the meaning of unknown word. For example,
“Typhoon Nari destroyed the city.” Here “destroyed” is an unknown word and learners
can guess that the typhoon did something to the city. With the help of the surrounding
clue, learners can guess that the word may have a negative meaning.
Step 3: Look at the relationship between the clause and sentence containing the
unknown word and other sentences or paragraphs. It means that l earners can use global
clues to guess the word meaning. Pay attention to transitional words such as “but”,
“because”, or “in other words”. Therefore, it is important to find if sentence pattern
belongs to the one such as cause and effect, condition, contrast, clarification, or
exemplification.
Step 4: Use the knowledge learners have gained from Steps 1-3 to guess the
meaning of the word. It means that learners combine the knowledge they gained by
following the first three steps and then make a guess of the word meaning.
Step 5: Check the guess is correct.
a. Check the speech of the word learners guess and the unknown word.
11

b. Replace the unknown word with the guess.
c. Use prefixes and roots as a guide.
d. Consult the dictionary.
The procedure proposed by Clarke and Nation (1980) is based on language clues
and does not focus on background content knowledge (Nation, 2001). The reason for
this is that linguistic clues can be found in every text, but background knowledge
cannot. Background knowledge is strongly related to both the textual aspects and the
learners themselves (Nation, 2001). Further, by using this procedure to guess words
from context, learners can pay more attention to vocabulary learning than focusing on
background knowledge.

learners‟ knowledge of words and texts.
As mentioned previously, lexical inferencing is a very important way of incidental
vocabulary learning. However, to achieve successful guessing, in addition to contextual
and learner-related factors proposed by Paribakht (2005), some more detailed factors
should also be discussed. Schmitt (2000) proposed seven factors that affect inferencing
success. These factors are described as follows.
1. The context must be rich enough to offer adequate clues to guess a word‟s meaning.
2. Readers are better able to use local clues in proximity to an unknown word than more
global clues that are located further away.
3. Learners may mistake an unknown word for one they already know with a similar
orthographic (written form) or phonological form (spoken form).
4. Cognates can help guessing from context if they are used prudently.
5. Background knowledge about the topic and the culture being discussed aids
inferencing.
7. Learners need to be skilled in guessing.
8. Guessing a word from context does not mean that it will be remembered.
1.2. Review of Previous Studies Related to the Research Area of the Thesis
Nagy et al. (1985) attempted to determine whether students could really acquire
unfamiliar word knowledge while reading authentic texts. In their study, eighth-grade
students of average or above average reading ability read either an expository or a
narrative text of about 1,000 words in length. After reading, those students received two
13

vocabulary assessment tasks on 15 target words selected from each text. An individual
interview and multiple-choice test were also conducted. After the experiment, the
results showed that learning from context did take place and vocabulary gain was small
but statistically reliable. Further, the amount of learning from the narrative was the
same as that from the exposition. Thus, it seems that word learning from context is not
confined to any specific genre. This study shows that here is a close relationship
between vocabulary learning and reading. After the proof of the existence of

in vocabulary knowledge contributed to better reading comprehension. Besides, the
participants of the experimental group who were instructed in lexical inferencing
performed better in vocabulary and reading comprehension tests than those of the
control one. That is, the instruction of the strategy – lexical inferencing had a positive
effect. Finally, the participants of the experimental group could make use of more types
of knowledge sources and contextual cues in the process of inferencing than those of
the control one. From the findings of Lo‟s study (2004), it is suggested that teachers
should help students to practice the inference strategy.
Hsu (2004) focused on an investigation of how to enhance incidental vocabulary
acquisition through guessing words from context. The participants in her study
consisted of four classes with a combined total of 186 junior college students in
Taiwan. The aim of the study was to instruct the students in lexical inferencing to
observe if the strategy could help them enhance incidental vocabulary acquisition
through reading. The participants were divided into two groups, an experimental group
and a control group, and later each group was further divided into three different levels
based on English proficiency. Afterwards, the experimental group received the
instruction in lexical inferencing as proposed by Clarke and Nation (1980), while the
control group received none. The results of the study are as follows. First, the
instruction of learning vocabulary strategies helped students to enhance their
vocabulary acquisition through reading. The incidental vocabulary gains increase in
them experimental group was significantly superior to that in the control group. This
implies that the students in the experimental group made a lot of incidental
improvement in vocabulary acquisition after they received the instruction. Second, the
researcher al so found those students' vocabulary gains did not absolutely correlate to
their reading comprehension level. Finally, based on the collected data of the
15

questionnaire, most of the students in the experimental group had a positive attitude
toward the strategy of lexical inferencing. Therefore, Hsu (2004) suggested that the
strategy of lexical inferencing be implemented in students‟ daily class.

into different majors: Mathematics, Physics, Chemistry, Biology, Informatics,
Literature, English, and two other non-majored classes: A1 and A2. As the students
come from different parts in Vietnam, many of them come from the countryside, so in
general, the English proficiency of the students is not very high. Reading is a difficult
skill for them, mostly due to their lack of vocabulary. In addition, they tend to be word-
by-word readers, and frequently tried to translate the reading text into Vietnamese.
The teaching staff is composed of 12 teachers of English, all of whom graduated
from University of Languages and International Studies, Vietnam National University,
Hanoi. Their ages vary from 23 to 40. The researcher is also a teacher of English at this
high school. She is 25 years old and has about nearly 3 years of teaching experience.
The textbooks used at this school are Tieng Anh 10, Tieng Anh 11, and Tieng Anh 12
– standard syllabus for non English majored classes, and advanced one for English
specialized classes. However, from the researcher‟s observations and from the students‟
reflection, the reading texts in this series of textbook are not rich enough in context to
motivate students to practice reading strategies.
2.2. Subjects
The subjects of this thesis study were composed of 46 students from class 10A1 at
High School for Gifted Students (HNUE), which the researcher is in charge of.
According to the result of Oxford Placement Test given by the researcher at the very
beginning of the school year, most of the students in this class are at pre-intermediate
level. The table below summarizes the background information of the participants
including the gender, age, and their English reading proficiency assessed by the teacher. 17

Total
number of
participants
Gender

instructions on
students‟ reading comprehension ability. To achieve the research
goal, the study
combined both quantitative and qualitative data. Quantitative data were
collected based on the results of the questionnaires, pretest and posttest. Further,
qualitative data were gathered using interviews, teacher‟s notes and students‟ learning
logs. In this way, the study could get more complete results.
2.4. Research Approaches
This study was carried out using action research approach. Action research is first and
foremost situational, being concerned with the identification and solution of problems
in a specific context. The aim of action research is to improve the current state of affairs
within the educational context in which the research is being carried out. To put it
another word, action research is a “small-scale intervention in the functioning of the