A STUDY ON THE LEXICAL RICHNESS IN THE WRITTEN
WORK OF THIRD YEAR STUDENTS OF ENGLISH
LANGUAGE EDUCATION STUDY PROGRAM OF SANATA
DHARMA UNIVERSITY
A ThesisPresented as Partial Fulfillment of the Requirements to Obtain the Sarjana Pendidikan Degree
in English Language Education
By
ANDREAS DIMAS ARDITYA Student number: 021214038
ENGLISH LANGUAGE EDUCATION STUDY PROGRAM DEPARTMENT OF LANGUAGE AND ARTS EDUCATION FACULTY OF TEACHERS TRAINING AND EDUCATION
SANATA DHARMA UNIVERSITY YOGYAKARTA
STATEMENT OF WORK ORIGINALITY
I honestly declare that this thesis which I wrote does not contain the works or part
of the works of other people, except those cited in the quotations and
bibliography, as a scientific paper should.
Yogyakarta, 9 February 2007
The Writer
ACKNOWLEDGEMENTS
My gratitude goes to my major sponsor, Dr. F.X. Mukarto, M.S. I am also
indebted to my cosponsor, C. Tutyandari, S.Pd, M.Pd.
I wish to thank the third year students of English Language Education
Study Program of Sanata Dharma University who have volunteered themselves
participating in this study.
My appreciation also goes to my beloved ones who have supported me in
my ups and downs. I wish to acknowledge all my colleagues and fellows, whose
list of names is too voluminous to be written here. I am grateful to those who have
directly or indirectly helped and supported me in writing this thesis.
TABLE OF CONTENTS
Page
TITLE PAGE ... i
PAGE OF APPROVAL ... ii
PAGE OF ACCEPTANCE ... iii
PAGE OF DEDICATION ... iv
STATEMENT OF WORK ORIGINALITY ... v
ACKNOWLEDGEMENTS ... vi
TABLE OF CONTENTS ... vii
LIST OF TABLES ... x
LIST OF FIGURES ... ABSTRACT ... ABSTRAK ... CHAPTER 1 INTRODUCTION 1.1 Background ... 1.2 Problem Identification ………..……….. xi xii xiii 1 3 1.3 Problem Limitation ... 1.4 Problem Formulation ... 1.5 Research Objective ... 1.6 Benefits of the Study ... 1.7 Definition of Terms ... CHAPTER 2 LITERATURE REVIEW 4 5 5 6 6 2.1 Theoretical Description ... 9
2.1.2 Knowledge of Vocabulary ...
2.1.3 Writing Process and Writing Feature ...
2.1.4 Vocabulary in Writing Process ...
2.1.5 Measurement of Productive Vocabulary in Composittions ...
2.2 Theoretical Framework of the Study ...
CHAPTER 3 METHODOLOGY
11
14
17
18
20
3.1 Method ………...
3.2 Research Participants ………...
3.3 Data Source and Nature of Data ...
3.4 Data Collection ...
3.5 Data Analysis ... 22
23
24
25
25
CHAPTER 4 RESEARCH RESULTS AND DISCUSSION
4.1 Data Presentation of Written Works ...
4.2 Analysis Results and Discussion of the Lexical Richness ...
4.2.1 Degree of Lexical Variation ...
4.2.2 Degree of Lexical Sophistication ..………...
4.2.3 Degree of Lexical Density ...
CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS
5.1 Conclusions ...
5.1.1 The Third Year Students’ Lexical Variation Degree in Their
Written Work ...
5.1.2 The Third Year Students’ Lexical Sophistication Degree in Their
Written Work ………...
5.1.3 The Third Year Students’ Lexical Density Degree in Their
28
30
30
33
35
39
39
Written Work ...
5.2 Recommendations ...
5.2.1 to the Lecturers ...
5.2.2 to the English Learners ...
5.2.1 to Other Researchers ...
BIBLIOGRAPHY ...
APPENDICES ...
Appendix A. Table of Number of Words, Word Families, Low Frequency
Word Families, and Content Words Counts ...
Appendix B. Tables of Scores of Lexical Variation, Lexical
Sophistication, and Lexical Density Measurements ...
Appendix C. List of Words Used in the Participants’ Composition and
Their Frequency ...
Appendix D. Samples of Third Year Students Written Works ...
Appendix E. Text Fragment of Published Research Report ... 40
41
41
41
42
43
45
46
48
52
64
LIST OF TABLES
Page
Table 4.1 The Table of Descriptive Statistics of the Written Works ………
Table 4.2 The Table of Descriptive Statistics of Lexical Variation
Measurement ………....………
Table 4.3 The Table of Descriptive Statistics of Lexical Variation
Measurement ………
Table 4.4 The Table of Descriptive Statistics of Lexical Density
Measurement ………...
Table 4.5 Top Ten Most Used Words in the Students’ Written Works …... 28
31
33
36
LIST OF FIGURES
Page
Figure 2.1 Chapelle’s vocabulary ability construct ...………...………
Figure 2.2 Hayes’ Writing Model General Organization …...……….. 12
ABSTRACT
Arditya, Andreas D. 2007. A Study on the Lexical Richness in the Written Work of Third Year Students of English Language Education Study Program of Sanata Dharma University. Yogyakarta: English Language Education Study Program, Sanata Dharma University.
Vocabulary is the bridge between what message is meant to be delivered and what message is actually delivered. There seems to be an agreement among researchers that lexical ability is given more emphasis in writing skill than it is in the other skills. Writing possesses greater lexical density: it is densely packed with information, less redundant and more fully formulated to fulfill the needs of a distant reader and to avoid ambiguity; suggesting that written work draws on a large supply of words.
This study investigated the lexical richness in the written work of third year students of English as Foreign Language in Indonesian university. The study was meant to find out three measurements of lexical richness: (1) the degree of the lexical variation, (2) the degree of the lexical sophistication, and (3) the degree of the lexical density in students’ written work.
The method used was document analysis on students’ written works. Descriptive statistics was used to determine how accurately inductive reasoning can be employed to infer that what was observed on the sample in this study would be observed on the third year students’ population. Cluster sampling was used to represent the third year students of English Education Study Program of Sanata Dharma University. The students’ written works were uploaded into a computer and Simple Concordance Program 4.07 was used to analyze the lexical statistics of the data.
ABSTRAK
Arditya, Andreas D. 2007. A Study on the Lexical Richness in the Written Work of Third Year Students of English Language Education Study Program of Sanata Dharma University. Yogyakarta: Program Studi Pendidikan Bahasa Inggris, Universitas Sanata Dharma.
Kosakata adalah jembatan antara pesan yang dikirimkan dan pesan yang diterima dalam komunikasi bahasa. Dikatakan bahwa kemampuan kosakata mendapat penekanan lebih dalam keahlian menulis, dibanding dalam keahlian keahlian bahasa lain. Karya tulis memiliki kepadatan leksikal lebih besar: informasi lebih padat, penggunaan bahasa lebih efektif, dan dibuat untuk memenuhi kebutuhan pembaca dan dengan tingkat ambiguitas rendah; menunjukkan bahwa karya tulis bersandar pada kemampuan kosakata.
Studi ini meneliti kekayaan leksikal pada karya tulis mahasiswa Bahasa Inggris tahun ketiga di Indonesia. Penelitian kekayaan leksikal dilakukan untuk mencari: (1) tingkat keragaman kosakata, (2) tingkat penggunaan katakata sulit, dan (3) tingkat kepadatan leksikal dalam karya tulis mahasiswa.
Metode yang digunakan dalam studi ini adalah analisis dokumen pada karya tulis mahasiswa. Statistika deskriptif digunakan untuk menentukan apakah hasil observasi pada sampel dapat digunakan untuk mewakili seluruh popuplasi mahasiswa tahun ketiga. 49 sampel digunakan untuk mewakil mahasiswa tahun ketiga Progam Studi Bahasa Inggris Universitas Sanata Dharma. Karyatulis mahasiswa dipindai ke dalam komputer dan program komputer Simple Concordance Program 4.07 digunakan untuk membantu analisa data statistik kosakata.
Chapter 1
INTRODUCTION
This chapter places the current research among previous research,
especially those dealing with lexical characters and/or abilities of students of
English as a Second or Foreign Language. This chapter consists of background of
the study, problem identification, problem limitation, problem formulation,
research objectives, benefit of the study, and definition of terms.
1.1 Background
Research in vocabulary has been various in kinds. According to Johnson
(2000, p. 177) these last decades were period of strong interest in vocabulary.
During the time, educators became convinced that most words probably are
learned from oral and written context—that is, through listening and reading. The
emphasis of the period, as during the earlier period, was predominantly on
reading and learning words as way to improve comprehension.
Although much less attention has been paid to lexicon than other parts of
language, lexicon is an important factor in second language acquisition. This
importance is noted by Gass (1988b, as cited in Gass and Selinker, 2001, p. 372)
who observed that utterances with grammatical mistakes can be generally
understood, but lexical mistakes lead to interference of the communication. This
is because lexis corresponds closely with the meaning. Bloom (2000, as cited in
the intended meaning cannot be produced and, thus, is not comprehended, then the
communication is a failure. Lexicon is the bridge of what message is meant to be
delivered and what message is actually delivered. As Levelt (1989, as cited by
Gass and Selinker (2001), p. 373) put it, “…lexicon is an essential mediator
between conceptualization and grammatical and phonological encoding...”
In writing skill, the lexical ability is given more emphasis than it is in
other skills. Composition process makes written work more concise, formulated
and explicit. Halliday (1985, cited in Nunan, 1991) pointed out that writing
possesses greater lexical density. Written work is densely packed with
information, it is less redundant. Written work is also more fully formulated to
fulfill the needs of a distant reader and to avoid ambiguity. These suggest that
written work draws on a large supply of words.
Hayes and Flower (1980, as cited in Hayes, 2000) articulated a cognitive
process theory of writing. They refer to three writing processes: planning—which
includes goal setting and organizing, translating, and reviewing. Word selection
comes into play during any of these three writing processes, and each word chosen is important. Each word that comes into a growing text gives impact to the
text; any words or choices of words coming after will be determined and limited
by the word.
Analyzing students’ written work can reveal their lexical characteristics.
The description of students’ vocabulary ability is available on the text they have
written. Laufer (1991) used free compositions written by English learners as
source of lexical richness data on determining their second language vocabulary
namely: lexical variation, lexical density, lexical originality and lexical
sophistication.
In the local context, Susilo (2001) and Saputro (2005) studied vocabulary
of English students in university in Indonesia. Susilo investigated the controlled
active vocabulary size, using Laufer and Nation’s controlled active vocabulary
test. His study, which was crosssectional, aimed to find out whether differences
exist on students at different levels. Saputro’s investigation was also cross
sectional. He used students’ impromptu written compositions using The Passport
to IELTS to measure lexical density and lexical profile of the students.
The current study is a follow up to Saputro’s study. The study measures
the vocabulary characteristics of English students of the English Department in
university in Indonesia using their written works which are part of task they have
to do in their Writing V class.
1.2 Problem Identification
One way to analyze the students’ vocabulary in their written production is
by calculating various statistics that reflect their vocabulary knowledge. Read
(2000) summarized a set of statistical analyses toward written production. The
general term that is used for the vocabulary characteristics measured by these
statistics is lexical richness. Lexical richness analyzes four lexical features of a
written work, namely: lexical variation, lexical sophistication, lexical density and
number of errors.
Lexical variation, or also known as typetoken ratio, is the use of a variety
compares the number of different lexical words used in the text with the total
number of the running lexical words used in the text.
Lexical sophistication is the use of a selection lowfrequency words that
are appropriate to the topic and style of the writing, rather than general everyday
vocabulary. The measurement compares the number of the sophisticated word
families used in the text with the total number of word families in the text.
Lexical density is the use of lexical words in a text. The measurement
compares the number of lexical or content words, which consists of nouns, full
verbs, adjectives and adverbs derived from adjectives, with the total number of
words in the text
Number of errors is the occurrence of lexical errors in a text. The
classification of what are included as lexical error can be varied including minor
spelling mistakes, major spelling mistakes, derivation mistakes, deceptive
cognates, interference from another language learning, confusion between two
items, etc.; depending on what is expected to be revealed of the counting.
1.3 Problem Limitation
The main emphasis of this study is on analyzing three out of four aspects
of the lexical richness, namely, the lexical variation, lexical sophistication, and
lexical density of the written work of third year students of English Education
Department of Sanata Dharma University. Measurement on number of error is
not included in this study because the measurement was the most subjective of all
measurement occurs because researchers are likely to have different identification
of lexical errors one another; their findings may be too different for them to have
a high level of agreement or interrater reliability. Read also added another
limitation for the errors measurement, which is the generalization of seriousness
errors: the measurement does not take into account the relative seriousness of
different errors.
1.4 Problem Formulation
The intention of this study is to find out the Lexical Richness of the third
year students of English Education Study Program of Sanata Dharma University.
As has been explained above, although lexical richness in its respect is composed
of four elements—namely lexical variation, lexical sophistication, lexical density
and number of lexical errors—with regard to students’ written work of English
Education Department, this study strives to answer three questions:
1. What is the degree of the Lexical Variation in students’ written work?
2. What is the degree of the Lexical Sophistication in students’ written
work?
3. What is the degree of the Lexical Density in students’ written work?
1.5 Research Objective
This study aims to find out the Lexical Richness of EED students, which
is interpreted from their written works’ statistical scores of Lexical Variation,
1.6 Benefits of the Study
To give contribution to English Language Teaching (ELT), This study is
hopefully able to: (1) offer insight for ELT teachers in perceiving vocabulary
knowledge usage—use of word types, specific vocabulary, and text’s vocabulary
density—especially in written composition of third year university students, (2)
provide ELT teachers and readers with understanding of how vocabulary
characteristics can be inferred from written work, and (3) give description of
productive vocabulary characteristics in written composition of EEDUSD third
year students.
1.7 Definition of Terms
Some terms are used in this study; therefore, to avoid misunderstanding, it is necessary to explain the terms based on relevant sources.
1.7.1 Word Family
This study analyzed documentary data of written text; then for practical
reason lexical item—that is, a word—was basically defined as
orthographic words, that is “…any sequence of letters (and a limited
number of other characteristic such as hyphen and apostrophe) bounded
on either side by a space or punctuation mark” (Carter, 1998). However,
since a lexical item can have various forms, in this study lexical item is
family, which is a fundamental unit underlying different grammatical
variant or wordforms. For example, ‘makes’, ‘made’, ‘making’ etc. are
underlain by the capitalized lexeme ‘MAKE’ and are counted as
‘MAKE’.
1.7.2 LowFrequency Item
Laufer (1991) defined lowfrequency item as “advanced word” and what
is included as this kind of word would depend on the level of the learner.
In this study “advanced word” is taken to be words that are not in the top
two bands of Collins Cobuild’s (2001) most frequent English word. The
words in the top two band account for approximately 75% of all English
usage. There are 1720 words in the top two bands list. Words that are not
included in the bands are topic specific and advanced words.
1.7.3 Content Word
Content word is “…word that has meaning in isolation and serve more to
provide links within sentences.”(Read, 2000, p.18). In this study content
words are nouns, full verbs, adjectives and adverb derived from adjectives
occurred in the students’ compositions
1.7.4 Lexical Richness
The term is a generic term Read (2000, p.200) used to subordinate four
sophistication, lexical density and number of errors. In this study lexical
richness refers to the first three.
1.7.5 Written Work
Written work is intellectual composition produced as readable text. In this
study written work are the written compositions of the EED students that
consist of no less than 300 running words and no more than 500 running
words. The texts are of argumentation type and are in their draft form,
that is, the texts have gone through process of planning, drafting; but not
revision, redrafting and final drafting. Further description of the nature of
the written work used in this study is explained in Section 3.3.
1.7.6 Third Year Students of English Language Education Department The students study English as a Foreign Language. They are trained to
master English and to qualify as English teachers. The students may have
different traditional, ethnics or L1 background, but they are speakers of
Bahasa Indonesia as a national language and they all use Bahasa
Indonesia as lingua franca; therefore in the current study their
Chapter 2
LITERATURE REVIEW
There are several theories that eventually lead to ascertaining students’
lexical characteristic through statistical lexical analysis on their written work. This
chapter is dedicated to give a deeper topic understanding of the study. A number
of components on related points of the topic will therefore be clarified.
2.1 Theoretical Description
This section discusses how a word is defined; and how vocabulary is
described as a part of language competence, and used in the written process. This
section ends in discussion on, as the focus of this study, analyzing vocabulary in
written production.
2.1.1 Definition of Word
Carter (1998, p.4) discussed several definitions and their related problems
of a ‘word’. When we want to define the meaning of ‘word’, we tend to directly
think of it as a sequence of letters (and a certain added characteristics such as
apostrophe, hyphen, etc.) which is stringed together between spaces or
punctuation marks. This kind of word definition is an orthographical one.
Generally this kind of definition has considerable practical validity, for instance in
counting words or making wordlist. However, this definition has its problems. For
example, should we take swim, swims, swimming, and swam as separate items?
railway line, fishing line and straight line represents different items in one
orthographic word. Other polysemous words can have more extended meanings
and grammatical categories.
A more complex (and maybe more accurate) definition of ‘word’ is a
minimum meaningful unit. But again, questions rise. Do we considerdining room,
airport tax, and cannon ball as single item? How about ambiguous compound
nouns and phrasal verbs? And what meaning do my, if, by, but, could, because,
indeed, etc. represent? The last few items represent words with less semantical
sense and more grammatical one. Another definition may better serve this kind of
words, that is, ‘word’ is a minimal free form.
This definition, which was originally Bloomfield’s, derives that a word
should have ‘positional mobility’ and ‘internal stability’. A word should be able to
move from its particular point in a sentence. Thus, for example, the sentence I
walked across the room quietly can be reordered as I quietly walked across the
room or quietly I walked across the room or the room I walked across quietly.
Also, the morphemes in a word have relatively consistent sequence to one
another; making the morphemic constituents of quietly, walked and across are
fixed, that is, not possibly permutated into *leiuqty, *tuqeily, *kwaled, *leawkd,
*croass, *rosarcs, etc. Singleton (1999, p.14) argued that the grammatical
definition is the least problematic, because it gives a word stability which prevents
further division or reduction, and an ability to stand on its own. However,
although possible it is very rarely do we see could or if occur on their own. And
fixed items which may lose meaning if reduced, can be substituted by a single
word and still can stand on their own. As Carter (1998, p.6) exemplified it:
Q: Is it raining hard?
A: Cats and dogs.
A further definition that a word will not have more than on stressed syllable does
not add any satisfaction.
One theoretical notion that can be used to overcome these problem
definitions is lexeme. A lexeme is the abstract unit which underlies some related
word variants. This notion shares essential view with what ‘word family’ enclose,
which is a set of word form sharing a common meaning. Carter (1998, p.7) used
upper case form to denote lexeme. Thus upper cased LEAK is the lexeme which
underlies grammatical forms of leaks, leaked, and leaking. Lexeme can also be
used for items consisting of more than a word form. For example, A PIECE OF
CAKE is a single lexeme because it underlies a definite meaning. Polysemous
words will also be represented by several lexemes although they use the same
word form. For example, we will have lexemes LAP 1 (noun and verb as in race),
LAP 2 (verb as inthe cat laps the milk), and LAP 3 (noun as insit on my lap).
2.1.2 Knowledge of Vocabulary
Read (2000, p.2835) discussed definition of vocabulary ability by
Chapelle (1994) which was based on Bachman’s general construct of language
ability. The definition includes ‘both knowledge of language and the ability to put
has three components: the context of vocabulary use; vocabulary knowledge and
fundamental processes; and metacognitive strategies for vocabulary use.
Figure 2.1 Chappelle’s Vocabulary Ability Construct
From the point of view of traditional vocabulary testing, context is the whole text that a testee draws on to interpret individual items within. However,
context should be seen more than just a linguistic phenomenon. There is also pragmatic knowledge which affects the vocabulary ability. Thus, the social and
cultural situations influences the meaning of the vocabulary used. For example, in Bahasa Indonesia, ‘iya banget’ would have no meaning whatsoever in a formal
situation like trial court; but in the situation of casual speech especially in the
younger generation the item will signify the user keen agreement on something said. Another example in casual youth American English is ‘no shit’; used as a
response of something said it expresses disbelieve or surprise, but used otherwise it would only send the harsh meaning of ‘shit’. Clearly then, pragmatic knowledge
For the vocabulary knowledge and fundamental processes component of
vocabulary ability, Chappelle (1994, cited in Read, 2000, p.3133) outlines four
dimensions. The first is vocabulary size, which refers to the number of words that
a person knows. Following Chapelle’s logic of a communicative approach to
vocabulary ability, Read (2000, p.32) pointed out that we should seek to measure
vocabulary size not only in absolute sense, but also in relation to particular
context of use; thus distinguishing, for example the learner’s vocabulary ability in
writing an essay from his/her vocabulary ability in discussing a football match or
reading an international newspaper. The second dimension is knowledge of word
characteristic. Some words are used with better knowledge of it than others; each
known words has its own range of understanding, from vague to more precise. For
example, in writing a composition a learner may miswrite effect in intention of
affect because he/she may know more about the latter one. Like the previous
dimension, the extent to which a learner knows a word varies according to the
context it is used. The third one is lexicon organization, which concerns the way
in which words and other lexical items are stored in the brain. The last dimension
is fundamental processes. These last two are processes that a user applies to gain
access to the knowledge of vocabulary, both for understanding and producing
vocabulary in speaking and writing.
The metacognitive strategies for vocabulary use are the strategies
employed by language user to manage the ways of using vocabulary knowledge in
communication. Most of the time these strategies are operated unconsciously, it is
only when communicating task become unfamiliar or cognitively demanding that
these strategies become more conscious. Learners have a particular need to use
function effectively in communication situation. A basic strategy used when
learners are attempting to produce vocabulary is simplification or avoidance. They
may avoid using a vocabulary item because not knowing or not confident about
producing it in its correct form. Some other general strategies includes:
paraphrasing (using telephone you can carry anywhere for cellphone) , language
switch (can I borrow your PENSIL?), and use of superordinate terms (saying tool
to replace hammer).
Richards (1976, cited in Meara 1996) gave a guideline in describing
vocabulary competence. He proposed that knowing a word means: a) knowing the
probability of encountering the word in speech or print; b) knowing the limitation
of using the word according to function and situation; c) knowing the syntactic
behavior associated with the word; d) knowing the base form of the word and its
possible derivations; e) knowing the associations between the word and other
words in the language; f) knowing the semantic value of the word; and g)
knowing many of different meanings associated with the word. Points b) through
g) are related to the productive aspect of vocabulary competence, on which this
study will focus, particularly in the written production.
2.1.3 Writing Process and Writing Feature
As part of language skills, writing is the last one to develop. In the course
of history human invented writing system long after they began to speak, while,
Although writing and speech comprise the productive skill of language
competence, they are essentially different. Kress (1994: 1734) argue that speech
and writing have a) distinctive grammatical and textual structure and organization;
b) different distinctive unit—the distinctive unit in writing is the sentence; c)
distinct social setting; d) different demands and e) distinct syntactic and textual
structure.
Hayes (2000) proposed a new writing model based on the 1980 Hayes
Flower model. Figure 2.2 depicts Hayes’ Writing Model general organization. In
his writing model writing process consisted of two components: the individual
and the environment. Hayes believes that writing depends on an appropriate
combination of cognitive, affective, social, and physical conditions if it is to
happen at all. First and foremost, writing is an intellectual activity requiring
cognitive process and memory, which subordinates, among others: phonological
memory, semantic memory, audience knowledge, topic knowledge, genre
knowledge and linguistic knowledge.
According to Silva and Matsuda (2002) in order for the writing process to
begin, the writer has to assess the rhetorical situation—i.e. a complex web of
relationships among the elements of writing: the writer, the reader, the text and
reality– and identify the primary purpose of writing, with a stress on one of the
elements of writing. Writer starts with the question “what is most important,
topically, to me in this text I am about to write.” What is paramount in writing is
the cohesive and continuous development of a topic, making the development
idiomatic, syntactic, morphological and lexical knowledge (Silva and Matsuda, 2002).
Figure 2.2 Hayes’ Writing Model General Organization
One variable which characterized writing is the physical absence of the audience or addressee—the reader. The language of writing is not generated in interaction. Although the audience may be known but the writer will not have
The Task Environment
The Social Environment The Physical Environment
Collaborators
The Audience The text so far
The composing medium
The Individual
Motivation/Affect Cognitive Processes
Goals
Predispositions
Belief and Attitudes
Cost/Benefit Estimates
Text Interpretation
Reflection
Text Production Working Memory
Phonological Memory
Visual/Spatial Sketchpad
Semantic Memory
LongTerm Memory
Task Schemas Topic Knowledge
Audience Knowledge Linguistic Knowledge
control who may see the text or under what circumstance is the text received. A written work is either filled with all necessary information for adequate interpretation and received well, or it is not and communication fails. Consequently, writing tends to have greater explicitness and elaboration.
2.1.4 Vocabulary in Writing Process
Being able to understand a word requires different approach from using
the word in speech or writing. To make it explicit, the ability to use a word
requires extended knowledge beyond what is needed to understand it. Unlike the
situation in reception, higher level of knowledge is involved in production. Brown
and Payne (1994, cited in Muncie, 2002) argued that converting receptive
vocabulary into productive vocabulary is the final stage of vocabulary learning,
and composing a written work would be the place for this to happen. Writing
allows greater chances of experimentation and resources (e.g. time, articles,
dictionaries, etc.) for learners, which would enable them to use less frequent but
more appropriate words.
While writing gives advantages in vocabulary development, at the same
time vocabulary is one of the most important features in writing. In writing,
communication between writer and reader is done through words and patterns of
words. For communication to happen, the distant reader demands great
explicitness of words and clarity of the written text. This makes knowledge of
words is critical in writing, because it heavily affect the success of written
communication. Muncie (2002) opined that, based on earlier studies, a lack of
vocabulary proficiency is perhaps the best indicator of overall composition
quality.
2.1.5 Measurement of Productive Vocabulary in Compositions
Productive vocabulary use can be measured by calculating various
statistics of written production. Analyzing students’ written work can reveal their
lexical characteristics. The description of students’ vocabulary ability is available
on the text they have written. Read (2000, p.200) summarized four statistical
measurement of productive vocabulary; namely, lexical variation, lexical
sophistication, lexical density and number of errors. The general term for all these
measurements is lexical richness. Lexical richness measures the lexical features of
written productions.
Lexical variation (LV), or also known as typetoken ratio, is the use of a
variety of different words rather than a limited words used repetitively. Lexical
variation measures the number of different lexical items used in the text compared
with the total number of the running lexical items used in the text.
Lexical sophistication (LS) is the use of a selection lowfrequency words
that the writers used in appropriateness to the topic and style of the writing, rather
than general everyday vocabulary. It includes the use of technical terms and
jargon as well as the kind of uncommon words in expression of meanings in
precise and sophisticated manner. Lexical sophistication measures the number of number of words family / type
the low frequency word families used in the text compared with the total number
of word families in the text.
Lexical density (LD) is the use of lexical words in a text. Lexical density
measures the number of content words, which consists of nouns, full verbs,
adjectives and adverbs derived from adjectives, compared with the total number
of lexical items in the text.
Number of errors is the occurrence of lexical errors in a text. The
classification of what are included as lexical error can be varied including minor
spelling mistakes, major spelling mistakes, derivation mistakes, deceptive
cognates, interference from another language learning, and confusion between two
items.
However, before calculation can be done and in order to obtain reliable
statistics; several processes should be drawn first. Among them are manually
checking and deciding some words classified as lexical items or low frequency
ones respectively. This process requires human judgment even though computer
program can be used as instrument. Another important factor is that the variety of
length of texts may affect the figures obtained, which is why it is best in these
kind of statistical measure to do limitation of text length. number of lowfrequency word family
LS = x 100%
number of word family
number of content words
LD = x 100%
2.2 Theoretical Framework of the Study
Vocabulary ability has three main components: the context of vocabulary
use; vocabulary knowledge and fundamental processes; and metacognitive
strategies for vocabulary use. The first dimension of vocabulary knowledge and
fundamental processes is vocabulary size, which refers to the number of words
that a person knows. Vocabulary size measurement should not only be done in
absolute sense, but also in relation to particular context of use. The second
dimension is knowledge of word characteristic. Some words are used with better
knowledge of it than others; each known word has its own range of
understanding, from vague to more precise. Like the previous dimension, the
extent to which a learner knows a word varies according to the context it is used.
The productive aspect of vocabulary competence, particularly in the
written production is related to the ability of : a) knowing the limitation of using
the word according to function and situation; b) knowing the syntactic behavior
associated with the word; c) knowing the base form of the word and its possible
derivations; d) knowing the associations between the word and other words in the
language; e) knowing the semantic value of the word; and f) knowing many of
the different meanings associated with the word.
One main characteristic of written production is the physical absence of
the reader. A written work should be filled with all necessary information, or
communication fails. Consequently, written work tends to have greater
explicitness and elaboration, is densely packed with information, and is less
Statistical measurements of written production are used to asses the
degree of productive vocabulary ability. This study will measure lexical variation,
lexical sophistication, and lexical density of students’ written work to asses their
Chapter 3
METHODOLOGY
This chapter presents the methods of data gathering and analysis. This
chapter provides explanations for the kind of the data, data gathering instruments,
data collection, and data analysis and interpretations required for the study.
3.1 Method
The study was basically descriptivequantitative study, which referred to
investigation using already existing data and concerned with the collection and
analysis of data in numeric form. This kind of study made ready a general
question in mind about certain phenomenon and then with specific question and
specific focus; which made the research focused on certain aspect of the possible
data available in the context being described (Seliger and Shohamy, 1989).
Furthermore, this study could be classified as documentary analysis research,
which analyzed collected document data from several units or individuals that had
already formed or existed in natural context at a given time. This research
collected students’ written work responses to their teachers’ assignment.
The descriptive statistics was used in analyzing the numerical data
interpreted from the written work. This research studied the individuals’
phenomena as group phenomena and treated the characteristics of individuals that
occurred in the measurement as the group’s characteristic. The descriptive
statistics was used to determine how accurately inductive reasoning can be
observed in the whole (the third year students). Thus, the study was concerned
with the generalized statistics, in which data were abstracted from a number of
individual cases.
Given the descriptive nature of the study, statistical analysis was done on
the numerical data. To decide the typical value of the lexical characteristics of the
group, the central tendency measures was applied the mean, and the median. The
central tendency was used as generalization of the groups’ lexical characteristics.
To decide how the central tendency would best represents the lexical
characteristics of the group, that is, whether it was appropriate to use the central
tendency as generalization of the group, information on data variability (standard
deviations) was also obtained.
3.2 Research Participants
To be able to have unbiased representation of data this study used
probability sampling, in which samples were drawn randomly from the
population. Because probability sampling sought representativeness of the
population, probability sample would have less risk of bias.
The subject of the study was 49 written works of third year students of
English Education Program of Sanata Dharma University. Initially the target
subject of this study was 50 (10 from each Writing V class), but after analyses
were done one subject failed to fulfill the criteria of minimum word number. The
students were generally between 1920 years of age. They have gone through the
learning in the university. Their knowledge of English as second language was
categorized as advanced.
3.3 Data Source and Nature of Data
From each of the participants a written work was collected. The
participants composed the written work as the first draft of a writing task given by
their lecturers in Writing V course. The written work was of argumentation type
or genre of writing, in which they were to take a position of on a certain issue and
defend their stand. First draft form meant that the participants made a structure
plan and drafted their written works; but the written works have not gone through
revision (either by selfcorrection or by lectures consultation), redrafting and re
revision before they were submitted to the lectures. The students worked on
written composition for two to three weeks period; from the time the students
were given assignments to the submission task date. The draft form represented
the students’ real productive vocabulary knowledge on the words they used
without being affected, and therefore disturbed, by other authorities.
The written works were collected from lecturers at different times from 1 –
15 September 2006. From each class 10 works were chosen randomly without
regard to name, student numbers, sex, and topic.
Only 300 to 500 first words of each student’s written work were used as
data, this is due to lexical richness known instability and sensitivity when
confronted with various text lengths. The 300 to 500 firstwords limitation was
done in order to reduce the impact of text length on the index of lexical richness
The analyzed words were the English words used in students’ written works; all
other words and illustrations that were found e.g. Indonesian words, names of
person and numbers were considered irrelevant and not counted.
3.4 Data Collection
The compiled 50 students’ written works were uploaded to computer, and
using the Simple Concordance Program (henceforth SCP) computer software a list
of words used in the written work and their frequency of occurrences were made
for each work. These lists were the base data for calculating the lexical variation,
lexical sophistication, and lexical density measurement of each written work.
During this process one of the written works did not reach the minimum threshold
of word number and thus dismissed.
3.5 Data Analysis
Data analysis was begun after frequency of words occurrences list and
used words list was made for each written work. Each written work was later
analyzed using the three lexical richness measurements. The first is lexical
variation measurement. To measure the degree of lexical variation of a written
work is to divide the number of different lexical items used in the work by the
total number of lexical items in the work. For each written work the SCP had
provided the typetoken ratio, which by character is the same measure as lexical
variation, but the SCP was not able to list the words in word family, e.g. made,
manually reexamined and the recounted. 49 figures of lexical variation were later
tabulated.
Almost similar procedures were done for lexical sophistication and lexical
density. For lexical sophistication the number of lowfrequency word family used
in the work was divided by the total number of lexical items in the work. The low
frequency items of the written work were any word that are not in the Band 5 and
Band 4 of Cobuild’s frequency bands. The words in the two Bands are the most
frequent English words used. There are approximately 1720 words in these bands
and they account for about 75% of all English usage (Collins Cobuild, 2001).
Each of the 49 wordlists was examined manually using the “top two” list. 49
figures of lexical sophistication were later tabulated.
To measure lexical density the number of content words used in the work
was divided by the total number of lexical items in the work. Content words are
nouns, full verbs, adjectives and adverbs derived from adjectives. Again the 49
wordlists were examined. Noncontent words or otherwise known as function
words numbers were low in type but high in token, that is, limited but frequent.
Because it was easier to count words which were not content words, such was
done and then to have the number of content words the total number of words was
subtracted by the number of function words. 49 figures of lexical density were
later tabulated. For the sake of data presentation and interpretation convenience,
the output figures of measurements were presented in percentage. This process
gathered 147 figures—three figures for each of 49 written work.
To answer the first problem formulation of the study, that is to acquire the
lexical variations was used. For the mean, the lexical variation figures from each
work were summed up and then divided by 49. The standard deviation was later
calculated to measure the variability within group as to decide whether the mean
figure was an appropriate representation of the groups’ lexical variation degree.
The same was done also for the second and the third problem formulation, their
within group’s means and standard deviation were calculated for measurement of
lexical sophistication and lexical density of the group. The figures of means and
standard deviations, with additional information of ranges, min and max figures,
median figures, were tabulated. To give a better view on the analysis results of
the lexical measurements, in the discussion the participants’ degrees of lexical
richness was compared with that of a benchmark. The benchmark was a 400
word fragment of a research report which was published in ResearchNotes in
Chapter 4
RESEARCH RESULTS AND DISCUSSION
This chapter consists of two main sections. The first section is data
presentation. It deals with the numerical scores of the written works i.e. number of
words, word families, low frequency items and content words. The results of data
analyses and discussion on the findings are presented in the second section, which
covers the answers to the study’s research questions: to find out the third year
students’ degree of lexical variation, lexical sophistication and lexical density.
4.1 Data Presentation of the Written Works
The study researched a group of students regarding its lexical richness of
written work. With accordance to the measurements in lexical richness, the results
of data collection were presented scores by scores.
Table 4.1 presents the mean, median, minimum score, maximum score,
range, and standard deviation scores of the participants’ written work. The table of
participants’ scores of total words, word families, low frequency items, and
content words on their written works’ is available in Appendix A.
Table 4.1 The Table of Descriptive Statistics of the Participants’ Written Works.
Variable (N = 49) Words Word Families LowFrequency Word Families Content Words
Lowest 300 115 17 108
Highest 500 243 91 301
Range 200 128 74 193
Median 493 178 43 228
The table shows that the number of participants’ produced words, with the
range of 200 words, for the lowest was 300 and the highest 500. As explained in
section 3.3, the limitation on the length of written works used in measurement is
important to stabilize measurement results. Central tendency scores of produced
words in the participants’ written works show that the average number was about
450 words and the median 490 words. The number of participants’ produced
words standard deviation 62.43 indicates that the dispersion of the written works
length is quite low. Furthermore, more than 70% (35 out of 49) of participants’
written works consisted of 450500 words. This low dispersion of numbers of
words would help to eliminate the flaws of lexical richness measurements due to
instability and sensitivity to various text lengths.
Participants’ produced word family number, with range of 128, was 115 at
the lowest and 243 at the highest. The extreme number 115 occurred in the word
family numbers, differing 21 points from the second lowest number 136.
However, this atypicality should not affect the participants’ average number. The
average number of the participants’ word family was about 180, while the median
was 178. With the standard deviation for the participants’ word family numbers
30.01, more than 60% (31 out of 49) participants’ produced 160220 word
families in their written works.
The number on lowfrequency word families was at lowest 17 and the
highest 91, ranging at 74. In average the number of participants’ lowfrequency
items was about 44, and not far from that the median was 43. With the
more than 65% (33 out of 49) participants used 3060 low frequency word
families in their written works.
Participants’ content words number varied between 108 and 301, ranging
by 193. The average number was about 220 and the standard deviation was 37.69.
These numbers showed that participants used 200250 content words in their
written works.
4.2 Analysis Results and Discussion of the Lexical Richness
In order to answer the research problem formulations and to present a
detailed discussion of the data analysis, this section is divided into three parts: (1)
analysis and discussion on the third year students’ lexical variation, (2) the
analysis and discussion on the third year students’ lexical sophistication, and (3)
the analysis and discussion on the third year students’ lexical density. The whole
computations were done by using SCP and Microsoft Excel computer programs.
4.2.1 Degree of Lexical Variation
Lexical variation is the tendency of using word family repeatedly. The
assumption is that the more repetition of word family there is in his/her written
work, the smaller quantity of lexical repertoire the participant has. The
measurement of lexical variation was done by dividing the number of word
families with the number of words used.
The lexical variation analysis results of the written works of the third year
students of English Education Program of Sanata Dharma University: the mean,
Table 4.2 and the table of entire results of lexical variation is available in
Appendix B.
Table 4.2 The Table of Descriptive Statistics of Lexical Variation Measurement. Score (%)
Lowest 30.85
Highest 52.37
Range 21.52
Median 39.63
Mean/Average 40.26 Standard Deviation 5.45
Table 4.3 shows that the lexical variation found in students’ compositions
although varied are quite stable with mean and median scores are close (.63
difference) and with 5.45 standard deviation in a 21.52 range. As presented by the
table of entire results in Appendix B, the scores are closely gradual. The
participants have no extreme anomalies whether at the lowest or highest scores.
This indicates that the participants’ lexical variation scores produced a continuum
where no individual was totally removed from the group.
The scores represented a variety of number of words. If the computation of
lexical variation is reversed by entering scores of lexical variation and a fixed
number of 400 words, it would say that the participant with the lowest score 30.85
produced approximately 123 word families. The participant with median score
39.63 produced 159 word families and participant with the highest score 52.37
produced 210 word families in a 400 words composition. This would give 87
word families difference between the lowest and the highest participants. The
mean 40.25 score would represent the third year students’ productive ability of
Before going to the discussion of the students’ lexical variation degree let
us keep in mind that students of English as Second Language are called upon not
only to 'know' vocabulary items, but also to use them in production, which in our
case, in the written production. Usage of these items in actual linguistic situations
is superior to mere understanding or recognition. Furthermore, correct use of
items would mean that the lexical items are not only part of the lexical repertoire
of a learner, but that they can be activated at will, and even better, flexibly. Active
knowledge of a large lexical repertoire may have more influence on a learner's
vocabulary production; such influence which may not observable in the
performance of a learner with smaller lexical repertoire.
The third year students’ inclination to repeat the same words in the
compositions—i.e., their lexical variation—was 40.26%. The 5.45 standard
deviation indicates that there is low dispersion of lexical variation scores among
the students. Despite the existing variance within the group, the students share an
average lexical variation characteristic. This percentage means that in their
composition the students used each word family roughly 2.5 times (from 100%
divided by 40.26%). To make a comparison, a 400word published research report
fragment was uploaded to the SCP computer program and its degree of lexical
variation was counted. The result was that the fragment’s lexical variation was
50% (2 times repetition per word family). Assuming that this fragment was
produced by a highly proficient English user and taking the general principle that
more proficient users use a wider range of vocabulary than less proficient ones
(Read, 2005), the third year students’ 40.26% lexical variation would be a quite
active word repertoire was likely not far below that of the highly proficient.
Furthermore, the students seemed to have in their hands a fair readiness to draw
on an accessible and retrievable lexicon as needs demands.
4.2.2 Degree of Lexical Sophistication
Lexical sophistication is the ability to use specific and advanced word
family, instead of using general and basic word family. If a learner uses more
advanced word family in their production, he/she is said to have better quality of
lexical repertoire. Lexical sophistication was measured by dividing the number of
low frequency word family with the number of word family.
The lexical sophistication analysis results of the written works of the third
year students of English Education Program of Sanata Dharma University: the
mean, median, lowest score, highest score, range and standard deviation are
presented in Table 4.3 and the table of entire results of the participants’ lexical
sophistication is available in Appendix B.
Table 4.3 The Table of Descriptive Statistics of Lexical Sophistication Measurement.
Score (%)
Lowest 6.55
Highest 27.54
Range 20.99
Median 15.23
Mean/Average 15.22 Standard Deviation 4.82
The participants scored at variety from as low as 6.55 to as high as 27.54,
ranging the scores by 20.99. Table of entire results of participants’ lexical
were used at more than 1 at each 4 word families rate, which made the particular
participant’s written composition was highly specific. But the table also shows
opposite extreme that in two written works only less than 9% of the word families
were low frequency items. However, with extreme scores occurred at both ends,
variability remains low as indicated by standard deviation 4.82. Again, as it was in
the lexical variation scores, variance remains low within the group’s lexical
sophistication scores. In average the third year students scored 15.23 on lexical
sophistication. This means that more than 15% of the total words families they
used in their written works were Collins Cobuild’s low frequency word families.
The third year students have shown that in terms of specific and advanced
word family they have managed well in their vocabulary choices. 15% of the total
word families they used were low frequency items. The same lexical measurement
done on the 400word published research report fragment resulted that the highly
proficient English user had scored 16 % in lexical sophistication. This is what
Read (2005) has asserted that the vocabulary use of higher proficiency students
was more sophisticated than that of those at the lower levels.
Scarcella and Zimmerman (1998) noted that there seemed to be an
agreement among scholars that the ability to access and use sophisticated lexical
register where setting demands is important in academic success. ESL students
who are unable to change lexical gears and shift from an informal, conversation
register to an analyticalacademic register encounter more and more difficulties as
the demands increase. This is given more emphasis especially in writing, because
